Utilization of Waste Marble Powder in Cement Industry

By Ahmed N. Bdour¹ * and Mohammad S. Al-Juhani²
December 2011

  1. Associate Professor, Civil Engineering Department, College of Engineering, University of Tabuk, Saudi Arabia   * Corresponding Author
  2. Dean, College of Engineering, University of Tabuk, Saudi Arabia
Abstract
Presently large amounts of slurry are generated in marble cutting plants with a serious consequences on the environment and humans. This paper presents test results showing the feasibility of using Waste Marble Powder (WMP) in cement industry as a substitute limestone. Also, it describes the formulation of new lime-based (CR II) cementitious materials derived from thus industrial wastes. Powder mixtures were prepared and fired at different temperatures. For comparison, similar formulations were prepared with pre-treated and commercially available natural raw materials and processed in similar conditions.
The characterization included chemical composition, determined by X-ray fluorescence (XRF), thermal behaviour (DTA and TGA) and particle size distribution. Also, physical parameters of milled powders such as the specific surface area and the percent of weight retained in a fixed sieve (75 μm) were introduced.
The CR II clinker was found to contain common cementitious phases, such as C3A and C3S, but free lime and calcium aluminum oxide sulphate were also identified by high temperature XRD and NMR. The lime-based cement obtained from wastes shows a stronger hardening character than the standard material, which tends to show dusting phenomena due to the presence of a reasonable amount of free lime. However, some impurities present in the waste materials progress the overall reactivity of the mixture.
CR II binder is parallel to other conventional binder in the field of building construction like lime, cement and other admixture but is different from them as it utilizes WMP as a major component with other ingredients. Test results show that this WMP based cement is capable of improving hardened concrete performance up to 16%, enhancing fresh concrete behavior and can be used in architectural concrete mixtures containing white cement. CR II is an environment friendly product, cheaper and it involves small equipments with no convoluted technologies and it requires very low energy for manufacturing.

Keywords: Waste Marble Powder, Cement, Concrete, Limestone

Introduction

Environmental constraints, increasing industrial activity and rising costs of natural mineral resources, have forced the transforming industry to review the logistics of raw materials supply (Saak et al.,1999). Worldwide, there has been an increased interest in resource recovery, recycling and resource conservation as key parameters for any sustainable development plans (Miletic et al., 2003).

Waste marble powder is generated as a by-product during cutting of marble. The waste is approximately in the range of 20% of the total marble handled. The amount of waste marble powder generated at the study site every year is very substantial being in the range of 250-400 tones. The marble cutting plants are dumping the powder in any nearby pit or vacant spaces, near their unit although notified areas have been marked for dumping. This leads to serious environmental and dust pollution and occupation of vast area of land especially after the powder dries up (as shown in Figure 1). This also may leads to contamination of the underground water reserves (Branco et al., 2004; Vijayalakshmi et al., 2001).

Traditionally, WMP products are disposed of as soil conditioners or land fill. However, there might be reusing or recycling alternatives that should be investigated and eventually implemented. Thanks to Civil Engineering research, numerous uses of waste marble powder have been introduced, including use in tiles manufacturing, concrete mixes, subgrade fill, and modified binder (Huseyin et al., 2010; Nunes et al., 2009; Akbulut et al., 2007; and Bache H., 1981). Of our particular interest is the use of WMP in cement industry as a substitute of limestone for the production of clinker.

While marble blocks are cut by gang saws, water is used as a coolant. The blade thickness of the saws is about 5 mm and normally the blocks are cut in 20mm thick sheets. Therefore, out of every 25mm thickness of marble block, 5mm are converted into powder while cutting. This powder flows along with the water as marble powder. Thus, nearly 20 % of the total weight of the marble processed results into WMP. The produced WMP has nearly 35%-45% water content. The total waste generation from mining to finished product is about 50 % of mineral mined.

Figure 1 · WMP pits at a local company site
Figure 1

In general, the clinker is the main component of cement, and is obtained by firing the appropriate mixture of raw materials at about 1500 °C. Common phases in Portland cement clinkers are: alite (3CaOSiO2, C3S), belite (2CaO·SiO2, C2S), tricalcium aluminate (3CaO·Al2O3, C3A), and tetracalcium aluminate ferrite (4CaO·Al2O3.Fe2O3, C4AF) (Giacomo et al., 2010).. Belite or sulphobelite-based cements contain the phases belite (C2S) and tetracalcium trialuminate sulphate (C4A3S¯) as their main constituents (Bonavetti et al., 2003). They do not contain alite or tricalciumaluminate, but may contain variable amounts of calcium aluminate ferrite. They also contain calcium sulphate (CS¯) in amounts higher than normal in Portland cement (Ho et al., 2001; Noguchi et al., 1999). One typical composition consists of: C2S 40%; C4A3S¯ 32%; C4AF 20%; CS¯ 8%. Sulphobelite cements may perform either as non-expansive high early strength cements, or as expansive cements, depending on the proportion of the individual phases in the clinker and the amount of interground calcium sulphate (Bonavetti et al., 2003). Two main advantages are generally attributed to these types of cements, namely: (i) energy savings upon firing (maximum clinkering temperature is 1350 °C; (ii) lower volume of CO2 emissions.

Free calcium oxide in small amounts (usually below 1 % of weight) is a regular constituent of Portland clinker, but larger amounts may be present if the maximum temperature in the production of the clinker is too low, the burning time is too short, or the CaO content in the raw material exceeds the acceptable range (lime saturation factor >100). Large amounts may cause expansion, strength loss and cracking of the hardened paste, due to a delayed hydration of free calcium oxide to calcium hydroxide, which takes place topochemically and is associated with an increase in volume (Giacomo et al., 2010). Thus, excessive amounts of free calcium oxide in clinker must be avoided. However, old cements have lime and/or pozzolans as main hardening phases and their durability is easily proved in several historical monuments and common buildings. It is then obvious than the porous structure of those mechanically weak structures accommodates larger dimensional changes without serious damage (Giacomo et al., 2010).

Environmental Problems Attributed to Waste Marble Powder

The WMP imposes serious threats to ecosystem, physical, chemical and biological components of environment. Problems encountered are:

It is therefore a social and legal responsibility of government and industry to solve the problem of WMP pollution (Corinaldesi et al., 1998). As such, development of country is only possible by sustainable balanced industrialization. Recently, the rapid social, economical, and environmental changes in Jordan stressed human society with unprecedented challenges. Thus, new approaches that consider industrial wastes as alternative raw materials becomes interesting, both technically and economically, for a wide range of applications. Of our particular interest is the use of WMP in cement industry as a substitute of limestone for the production of clinker.

Objectives

Because of the environmental threats associated with the WMP, their proper disposal has attracted a lot of attention of the environmentalists in the last years. In order to properly dispose of these hundreds to thousands of tonnes of powder, the use of innovative techniques to recycle them is important. Without the proper disposal of this powder material, the resulting stockpiles would cause major health risks for the public and the environment (Acchar et al., 2006).

This paper proposes a gainful utilization of waste marble powder as a part substitute of limestone in a cement plant. Bearing in mind the recent rapid economical growth and developments in the construction sector in Jordan. In particular, cement fabrication involves a huge consumption of natural raw materials (e.g. limestone and clay). Of our particular interest is the use of waste marble powder in cement industry as a substitute of limestone for the production of clinker.

This research describes attempts to define the compositions of waste-based mixtures and the corresponding processing conditions suitable to the production powder based cements. Also, this study assesses the properties of the final product after incorporating waste marble powder. Ultimately, as an outcome of this project, the incorporation of waste marble powder in cement industry could lead to a viable, environmentally friendly material with attractive properties.

Approach (Methodology)

For the purpose of this research, several WMP specimens were prepared. The raw material was provided by a local company, and then these materials were milled and sieved through 75 μm sieve size. After that, rigorous analyses were carried out at the Civil Engineering Laboratories and the Centre for Environmental Studies at the Hashemite University.

The characterization included chemical composition, determined by X-ray fluorescence (XRF, Philips X'UNIQUE II), thermal behaviour (DTA and TGA, Setaram) and particle size distribution (Coulter) in alcohol. Also, physical parameters of milled powders (about 2 h in a rings mill+porcelain jar) such as the specific surface area (SSA, by BET) and the percent of weight retained in a fixed sieve (75 μm) were determined (Ahn 2004; Singh et al., 2001).

For comparison, and to clarify the effect of minor components, present in the WMP materials, on the final product properties, similar compositions were also prepared with commercial high purity grade raw materials, M1 calcite are used as standard samples.

In order to prepare the corresponding cement (CM) carbonate (11%) and calcium sulphate (6% hemi-hydrate) is added to the clinker before milling. Since the available amount of material is small, setting characteristics of the optimized formulations (water/cement ratio=0.45) is inferred from temperature measurements of the fresh pastes. Samples have been placed in a special thermal insulating box in order to assure almost adiabatic conditions. Temperatures are continuously measured through the insertion of a thermocouple in the paste (Giacomo et al., 2010).

X-ray Photoelectron Spectroscopy (XPS) was used to characterize the outermost atomic layers on the surface of cement formulations aged for 1 and 2 months. Each sample was ground in an agate pestle and mortar to expose fresh surfaces for analysis. The powder obtained was compressed into holes 10 mm in diameter and 0.5 mm deep machined into a copper sample holder.

The prepared samples were immediately transferred into the XPS sample introduction chamber to minimize the effects of surface contamination and reaction. An area of approximately 4 mm × 4 mm on the compressed powder surface was analyzed using a Thermo VG Scientific Escascope spectrometer with an AlKα (1486 eV) X-ray source, operated at 280W(14 kV, 20mA). Wide scan survey spectra were obtained between 0 and 900 eV binding energy (BE) with a step size of 1.0 eV.

The evolution of compressive strength with curing time is also monitored on small test bars (~22×14×10 mm) cast from the cement pastes into iron moulds. Phases formed upon curing are determined by XRD (Rigaku Geigerflex D/max - Series, CuKα at 40 kV and 50 mA) and by DTA. Finally, characterization techniques were used to assess micro-structural and compositional challenges of the aged cements (Nuno et al, 2007). These included Raman spectroscopy, X-ray Photoelectron Spectroscopy (XPS), and Focused Ion Beam imaging (FIB).

Figure 2 · Location in the composition diagram of the
SiO2 – Al2O3 – CaO ternary system, of the waste marble
powder and the compositions investigated (CR II/CR II-P)
Figure 2

Results and Conclusions

Characterization of Clinkers

Figure 2 shows the location of the formulations in the ternary diagram S – A – C. The resulting loose powders were calcined at selected temperatures (10 °C/min heating rate, 1 h soaking) and named as CR II. The standard clinkers (hereby distinguished by a CR II-P) were prepared in a similar manner, but using the Calcitec M1 calcite commercial reagents.

Chemical and physical parameters of milled WMP and particle size distribution are shown Table 1, Table 2, and Table 3, respectively.

Table 1 · Chemical Analysis of Waste Marble Powder
Test Carried Out Test Value %
1. Loss on ignition 39.66
2. Silica 1.58
3. Alumina 0.99
4. Iron Oxide 0.22
5. Lime 54.17
6. Magnesia 3.87
7. Soda < .015
8. Potash < .015
Table 2 · Physical Properties of Waste Marble Powder
Property Result
Bulk Density (gm/cc) 1.25-1.65
Specific Gravity 2.75-2.98
Particle size < 343.2 µ
Table 3 · Particle Size Distribution of Waste Marble Powder
Particle Size (mm) % Finer by Volume
345.1 100
195.0 – 205.8 90
133-140 80
82.0 – 96.7 70
55.5 – 62.51 60
38.5 – 45.1 50
23.9 – 28.1 40
14.24 – 17.2 30
4.9 – 7.15 20
1.33 – 1.89 10
0.421 0.00
Figure 3 · XRD of CR II and CR II-P clinkers: W(γC2S in
CR II and βC2S in CR II); W° (W+G); F (calcium sulpho-
aluminate); F° (W+F), M (mayenite, C12A7); U (alite;
C3S: M1 – JCPDS 13-0272 e M3 – JCPDS 42-0551),
U* (U+W), T (C3A: JCPDS 38-1429); T* (T+U)
Figure 3

Figure 3 compares the very complex powder XRD patterns of CR II and CR II-P clinkers. Terms such as W° (W+G) mean that peak is common to both phases. CR II-P contains the expected hydraulic phases (C2S, C3S and C3A), while CR II contains, in addition to C2S and Gehlenite, C12A7 and calcium aluminum oxide sulphate.

This difference is certainly due to a sluggishtendency to equilibrium in the solid state and the presence of other minor constituents in the powder materials. This lime-based CR II clinker predictably contains a higher amount of sulphur but less than 0.2% weight of Na2O+K2O. The relatively low amount of alkalis might contribute to the low stability of the high temperature C2S polymorphs (alpha), while the high amount of sulphur (N2 wt.% as SO3) led to the formation of calcium aluminate oxide sulphate. Those phases are well known by their expansive characteristics and might be responsible for the observed dusting tendency of CR II samples. NMR technique has also been used to clarify the structure of cementitious phases, mostly after hydration. Common phases in Portland-type cement clinkers (C2S, C3S, C3A, and C4AF) were prepared with pure chemical reagents and characterized, to be used as references, aiming at an easier characterization of the actual complex clinkers.

Figure 4 · 29Si MAS NMR spectra
of the CR II and CR II-P clinkers
Figure 4

Figure 4 compares the 29Si MAS NMR spectra of the CR II and CR II-P clinkers. Two major peaks between −65 and −75 ppm can be observed. Such chemical shifts correspond to Q0 or Q1 mono-silicates, totally depolymerised tetrahedral. The βC2S polymorph (peak at −71.3 ppm) and the C3S phase (peak at −73.5 ppm) are present in both CR II and CR II-P, but the later seems to be a better crystallized material (sharper resonances) and contain a higher amount of C2S (higher peak area). In the WMP based clinker, the peaks at −67.4 and −69.5 ppm are attributed to Q0 and Q1 sites in crystalline silicates, also presumed in the broad shoulder between −67 and −69 ppm in the C1 spectrum. In CR II, the clear presence of γC2S (peak at −73.7 ppm) may also contribute to dusting behavior of this powder when exposed to ambient conditions. The delay in the setting process displayed by the CR II clinker may also be explained by its lower C2S content.

Figure 5 · 27Al MAS NMR spectra
of the CR II and CR II-P clinkers
Figure 5

The 27Al MAS NMR spectrum of the CR II clinker (Figure 5) shows a peak at 67.7 ppm which, using the XRD information, can be attributed to C12A7. Besides C12A7, the spectrum of the CR II-P sample exhibits extra peaks, attributed to C3A (peak at 54.0 ppm) and probable hydrates (peak at ca. 17 ppm). Due to low iron oxide content in the raw materials, it is not surprising that the presence of C4AF was not detected in CR II (Akbulut and Gure, 2007).

To summarize, Table 4 gives main identified phases in the clinkers by the two techniques.

Table 4 · Summary of main detected phases on clinkers by using distinct techniques
Clinker XRD NMR
CR II C2S; C12A7; C2AS; calcium aluminum oxide sulphate β – C2S; γ-C2S; C3S; C12A7; C2AS
CR II-P C2S; C3S; C4A β – C2S; C3S; C3A; C – S – H; C – A – H

Behavior of Cements

Table 5 shows some characteristics of CM II and CM II-P cements. Figure 6 shows the evolution of their compressive strength with curing time. These values are much lower than those reported for Portland cement (about 10% after 7 days of curing), meaning that the hardening process is much slower. However, long-term resistance is more comparable. For example, samples cured for 3 months show compressive strength of about 32 MPa.

Table 5 · Initial setting time and mechanical strength development upon curing of cements
Property CM II CM II-P
Initial setting time (min) ≅628 ≅347
Compressive strength - 7 days (MPa) 3.58 5.69
Compressive strength - 28 days (MPa) 3.79 11.22
Figure 6 · Evolution of the compressive
strength of cements with curing time,
CM II and CMII-P formulations
Figure 6

Comprehensive characterization of the 28 day aged samples was conducted by different techniques. Using DTA, decomposition reactions (between 110 and 350 °C) of C – A – H and C – S – H hydrates seem to be stronger in CM II-P, revealing their presence in higher amounts. The absence or small expression of such kind of peaks in CM II cement is compatible with its weak hydraulic character.

Conclusions

In conclusion, the WMP based clinkers were found to contain the expected cementitious phases and a good agreement was obtained between the characterizations techniques used. In particular, the dusting phenomena observed in CR II clinkers was explained by the formation of calcium aluminum oxide sulphate and also by the presence of γC2S.

On its standard counter-sample (CR II-P) the βC2S is stabilized instead. Since the total amount of alkalis in CR II-P is even lower, the non occurrence of expansive (dusting) reactions in this material means that the lack of those mineralizing agents is not the controlling factor of the unwanted phenomenon. So that sulphur content should be the determining factor, inducing the formation of detected expansive sulphur-containing phases.

The WMP based cement obtained from wastes shows a stronger hardening character than the standard material, which tends to show dusting phenomena due to the presence of a reasonable amount of free lime (as the result of its expansive reaction with ambient moisture). Some fluxing impurities (e.g. alkalis) present in the waste materials improve the overall reactivity of the mixture.

This newly develop binder is parallel to other conventional binder in the field of building construction like Lime, cement and other admixture but is different from them as it utilizes WMP with other ingredients. Test results show that this WMP based cement is capable of improving hardened concrete performance up to 16%, enhancing fresh concrete behavior and can be used in architectural concrete mixtures containing white cement. The new binder is environment friendly, cheaper in cost from the other conventional material also it involves small equipment with no complicated technologies and requires very low energy for manufacturing.

Acknowledgements

The authors would like to thank the anonymous referees whose comments have helped to improve the quality of this manuscript. The staff of the Civil Engineering Laboratories and the Centre for Environmental Studies at the Hashemite University and the University of Tabuk for their assistance and support during the course of this research. Also, the Higher Council of Science and Technology (HCST), for their generous financial support.

References

  1. Acchar W, Vieira FA, Hotza D. Effect of marble and granite sludge in clay materials. Material Sci Eng A (2006);419:306 – 9.
  2. Ahn S., D.W. Fowler, An Experimental Study on the Guidelines for Usi ngHigher Contents of Aggregate Microfines in Portland Cement Concrete, in: Research Report 102-1F, ICAR (Eds.), Texas, USA, 2001.
  3. Akbulut H, Gurer C. Use of aggregates produced from marble quarry waste in asphalt pavements. Build Environ 2007;42:1921 – 30.
  4. Bache HH. Identified cement ultra-fine particle-based materials. In: Malhotra VM, editor. Superplasticizers in concrete, SP-68. Farmington Hills, (MI): American Concrete Institute; 1981. p. 3 – 35.
  5. Bonavetti V., H. Donza, G. Menéndez, O. Cabrera, E.F. Irassar, Limestone filler cement in low w/c concrete: a rational use of energy, Cement and Concrete Research 33 (6) (2003) 865 – 871.
  6. Branco F., Brito J., and Santos J., High-performance concrete with recycled stone slurry Nuno Almeida, Cement and Concrete Research 37 (2007) 210 – 220
  7. Corinaldesi V, Moriconi G. The influence of mineral additions on the theology of self-compacting concrete. In: Malhotra VM, editor. Superplasticizers and other chemical admixtures in concrete, SP-217. Farmington Hills, (MI): American Concrete Institute; 2003. p. 227 – 40.
  8. Giacomo M., Tarun R. N. and Valeria C. Characterization of marble powder for its use in mortar and concrete Construction and Building Materials 24 (2010) 113 – 117
  9. Ho S., Sheinn M., Ng C., and Tam T., The use of quarry dust for SCC applications, Cement and Concrete Research 32 (4) (2002) 505 – 511.
  10. Huseyin Y. A. , Metin G., Mustafa D., and Ilker T. Utilization of waste marble dust as an additive in cement production. Materials and Design, 31 (2010) 4039 – 4042
  11. Hwang EH, Ko YS, Jeon JK. Effect of polymer cement modifiers on mechanical and physical properties of polymer-modified mortar using recycled artificial marble waste fine aggregate. J Indus Eng Chem (2008);14:265 – 71.
  12. Miletic S, Ilic M, Milosevic M. Mihajlov, A. Building materials based on waste stone sludge. San Sebastia´ n: WASCON; 2003.
  13. Munoz-Montano J. Project Reference BRST985531. Recycling of waste originated in the process of cutting natural stone. In: BRITE/ EURAM3; 2003.
  14. Noguchi T., S.G. Oh, F. Tomosawa, Rheological approach to passing ability between reinforcing bars of self-compacting concrete, Proceedings of First International RILEM Symposium on Self-Compacting Concrete, RILEM Publications, S.A.R.L., Stockholm, (1999), pp. 59 – 70.
  15. Nunes S, Milheiro Oliveira P, Sousa Coutinho J, Figueiras J. Interaction diagrams to assess SCC mortars for different cement types. Constr Build Mater (2009);23(3):1401 – 12.
  16. Nuno Almeida_, Fernando Branco, Jose´ Roberto Santos Recycling of stone slurry in industrial activities: Application to concrete mixtures. Building and Environment 42 (2007) 810 – 819
  17. Singh S, Vijayalakshmi V. Marble slurry – a new building material. In: TIFAC News and Views (http://www.tifac.org.in/news/marble.htm); 2004.
  18. Saak W., Jennings H., and Shaf S., Characterization of the rheological properties of cement pastes for use in self-compacting concrete, Proceedings of First International RILEM Symposium on Self-Compacting Concrete, RILEM Publications, S.A.R.L., Stockholm, (1999), pp. 83 – 94.
  19. Vijayalakshmi V, Singh S, Bhatnagar D. Marble slurry a new resource material for entrepreneurs. Jodhpur: National Seminar on Small Scale Mining; 2001.

***

Copyright © 2011, ECO Services International