- Email: [email protected]
Improvement on reactivity of cementitious waste materials by mechanochemical activation
Materials Letters 58 (2004) 903 – 906 www.elsevier.com/locate/matlet
Improvement on reactivity of cementitious waste materials by mechanochemical activation Jaesuk Ryou Department of Civil Engineering, Center for Advanced-Cement Based Materials, Northwestern University, 2145 Sheridan Road, Evanston, IL 60208, USA Received 5 June 2003; accepted 27 July 2003
Abstract Combinations of two waste materials were used to develop cementitious material through mechanochemical activation. Activation is accomplished through the use of attrition mill grinding. Energy absorbed by the material on impact may result in a modification of the particle surfaces by introducing dislocations, point defects, and other structural defects. This leads to materials with increased surface free energy, which makes them more reactive. Properties, including X-ray diffraction, initial set of time and particle size distribution, are determined. The results indicate that separate grinding of the two materials is most effective at activating the materials and provide the best properties of the paste. D 2003 Elsevier B.V. All rights reserved. Keywords: Waste materials; Attrition mill; X-ray diffraction; Initial set; Separate grinding
Throughout the history of Portland cement manufacturing, the process has generated cement kiln dust (CKD). CKD is a particulate matter entrained in the process air stream, resulting from the tremendous quantities of pulverized rock in the kiln system. Air pollution control equipment prevents it from entering the atmosphere. The United States’ cement industry generates about 5 million tons of cement kiln dust every year that is not recycled, and only 20% of which enters commerce through utilization and the remainder is sent to landfills [1,2]. CKD has been labeled by the Environmental Protection Agency as a high volume, toxic waste stream, and it has been the focus of studies over the last decade in order to determine the appropriate method of regulation in the future. Disposal of waste CKD is not only associated with the problem of land use but also with contamination of ground water from leaching of chemicals from the material. Similarly, another industrial by-product, fly ash, collected from the stack gases of power plants burning pulverized coal, is also commonly considered an extensive waste material. Annually, about 59 million tons of fly ash are generated in the United States [3]. The cement and concrete industry utilize approximately 20% of the total US output of fly ash, while the remainder is placed in landfills. Many fly E-mail address: [email protected] (J. Ryou). 0167-577X/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2003.07.047
ashes cannot be used in concrete because they are of poor and variable quality. Principal contaminants of fly ash include unburned carbon, iron oxides, and, in some cases, calcium sulfates and aluminates. The use of fly ash in cement and concrete may further be affected due to the recent EPA regulations, which results in fly ash produced with higher carbon content. CKD and fly ash have been used as a supplementary cementitious material in concrete in small amounts. Usage is limited due to deficiencies of the two materials. For example, CKD contains an excessive amount of fine particles and a high alkali content, which often impart adverse effects on concrete workability and durability. Most fly ash hydrates slowly and some contain a high unburned carbon content. These inadequacies are unacceptable for strength development and volumetric stability of cement and concrete materials. However, if the two materials are blended together, the alkalis from CKD may activate hydration of fly ash. The research involves properly blending CKD with fly ash to create a cementitious material in which the material deficiencies will be converted into benefits. The purpose of this research is to explore an effective way to substantially utilize CKD and fly ash by developing an environmentally friendly, sufficiently performing, and cost-effective cementitious product for future concrete materials.
904
J. Ryou / Materials Letters 58 (2004) 903–906
Table 1 Chemical composition of the materials Compound
Fly ash
Cement kiln dust
SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K2 O
51.65 23.52 9.39 4.73 1.07 1.42 1.73 2.32
17.67 5.06 2.75 56.99 0.91 6.55 0.30 3.43
One way to activate hydraulic reactivity of cementitious materials is the mechanical process called mechanochemical grinding. The most widely used and best known methods in mechanochemistry include treatment of different types of mill and other similar apparatus [4,5]. These lead to materials with increased surface free energy, which makes them more reactive. Along with grinding in conventional mills, this investigation is proposed to investigate attrition mills. Attrition mills are characterized by higher energy intensity than other types of machinery. A central shaft with arms continually stirs the particles and spherical media and provides a means to vary the grinding energy. In this investigation, the total media charge to powder weight ratio was five to one. The cement kiln dust was supplied by a cement company from a wet kiln process. The chemical composition is shown in Table 1. Cement kiln dust is a fine offwhite or light brown powder. Particles are often random in shape and size, as can be seen in Fig. 1. The kiln dust used is very coarse with a mean particle size of approximately 63 Am. The Class F fly ash was provided by a power plant located in Illinois, as shown in Table 1. Also, fly ash is generally composed of particles called cenospheres, which are round and either hollow or solid particles. Fig. 2 shows fly ash particles under a scanning electron microscope and the cenospheres clearly dominate
Fig. 2. SEM picture of fly ash at 600 magnification.
the composition. The mean particle size of the material is approximately 7.2 Am. The proportioning of cement kiln dust and fly ash was designated for 65% cement kiln dust and 35% fly ash by weight. Table 2 presents the mix combinations and grinding regimes that were tested during the investigation. The following notations are used to differentiate the mixes: G for grinding; F for fly ash; SG for separate grinding; IG for inter-grinding. For example, mix IG-8 in Table 2 was an inter-grinding of cement kiln dust and fly ash for 8 h. After the appropriate grinding regime, mixing was done with a water-to-binder ratio of 0.5. All mixing occurred in a standard paddle mixer. Three tests were used to indicate if the material was a potentially suitable cement replacement: X-ray diffraction, particle size distribution and initial set of time. X-ray diffraction analysis identifies the crystalline and amorphous phases of the material. Particle size distribution discerns the effects of grinding on the particle size. This was performed using a Sympa Tec laser analyzer. Initial set of pastes will be determined using a Vicat apparatus, commonly used in industry. Time of set was measured according to ASTM C 191 ‘‘Standard Test Method for Time of Setting of Hydraulic Cement by Vicat Needle’’.
Table 2 Mix combinations and grinding regimes
Fig. 1. SEM picture of cement kiln dust at 200 magnification.
Mix combination
CKD grinding time (h)
FA grinding time (h)
CKD + FA grinding time (h)
Total grinding time (h)
G-0 FG-4 FG-8 SG-8 SG-12 IG-4 IG-8
0 0 0 4 4 0 0
0 4 8 4 8 0 0
0 0 0 0 0 4 8
0 4 8 8 12 4 8
J. Ryou / Materials Letters 58 (2004) 903–906
905
Fig. 3. X-ray diffraction pattern for cement kiln dust and fly ash.
The X-ray diffraction pattern for cement kiln dust and fly ash was given in Fig. 3. The crystalline material of the cement kiln dust is of much greater intensity than that of the fly ash, while the fly ash has a greater amorphous phase. Also, the diffraction patterns were compared to analyze the extent of crystalline and amorphous material after grinding. It was expected that mechanochemical activation occurred if the bottom of the curve was increased by grinding, indicating a larger amorphous phase which is more reactive, and if there was a reduction in the crystalline peaks of the material. The X-ray diffraction pattern for mix combinations G-0 and SG12 was shown in Fig. 4. While having a small amorphous phase, mix G-0, with no grinding, also has a large quartz peak indicating non-amorphous material. As seen in Fig. 4, 12-h separate grinding of cement kiln dust and fly ash, SG12, was compared to the no ground mix and the result was
also lower crystalline peaks. The crystalline material of mix G-0 is much greater intensity than that mix SG-12, while mix SG-12 has a greater amorphous phase. Grinding changed the particle size distribution. Particles combined and were broken from the impact of the grinding media. Table 3 shows the mean particle size and percent finer than 0.45 Am for all grinding regime. The mean particle size is the size at which 50% of the particles are larger and 50% are smaller. It can be seen that the mean decreased for all of the mixes as the grinding time increased. Furthermore, there are significant decreases in mean particle size when separate grinding is incorporated into the grinding regime, such as SG-8 and SG-12. The corresponding trend is an increase in the percentage of particles finer than 0.45 Am, the lower limit of the laser measurement, as grinding time increases.
Fig. 4. X-ray diffraction pattern for mix combinations G-0 and SG-12.
906
J. Ryou / Materials Letters 58 (2004) 903–906
Table 3 Results of particle size test and initial time of set Mix combination
Mean particle size (Am)
Percent finer than 0.45 Am (%)
Initial time of set (h)
G-0 FG-4 FG-8 SG-8 SG-12 IG-4 IG-8
11.8 7.48 6.71 5.31 5.26 8.47 7.66
2.36 4.13 4.86 5.63 5.96 2.72 3.25
38.5 24.5 22.0 12.0 10.5 16.5 18.0
of the mixes with separate grinding showed much improved results over inter-grinding. However, initial set time does not appear to be dependent on the length of grinding. The longer ground mixes in the inter-grinding did not necessarily give the best results. Since separate grinding provided mixes with the shortest initial set times at all length of grinding, it is clear that separate grinding has some type of activation effect on the materials.
References Initial set time was chosen for testing as it indicates the amount of time available between mixing and setting of the paste; for industrial applications, that is the amount of time available for transportation and placement. Thus, the initial set time of the paste would provide important information for a new material. The results for all grinding regime are shown in Table 3. The no grinding mix, G-0, took over 38 h to reach initial set. Initial set decreased with the use of attrition grinding. Mixes SG-8 and SG-12 have lower set times than the other grinding regimes. All
[1] J.E. Halliday, R.K. Dhir, T.D. Dyer, An investigation of the hydration chemistry, of ternary blends containing cement kiln dust, Journal of Materials Science 34 (20) (1999 October) 4975 – 4983. [2] J.I. Bhatty, Alternative uses of cement kiln dust, Research and Development Information, Portland Cement Association, 1995. [3] J.I. American Concrete Institute (ACI) Committee 232, Use of Fly Ash in Concrete, American Concrete Institute, Detroit, Michigan, 1996. [4] V. Ronin, H. Hedlund, Advanced modification technologies of the portland cement based binders for different high performance applications, Proceedings of the 10th International Congress on the Chemistry of Cement, Gothenburg, Sweden, 1997. [5] J.S. Reed, Principles of Ceramics Processing, 2nd ed., Wiley, New York, 1995.
Improvement on reactivity of cementitious waste materials by mechanochemical activation Jaesuk Ryou Department of Civil Engineering, Center for Advanced-Cement Based Materials, Northwestern University, 2145 Sheridan Road, Evanston, IL 60208, USA Received 5 June 2003; accepted 27 July 2003
Abstract Combinations of two waste materials were used to develop cementitious material through mechanochemical activation. Activation is accomplished through the use of attrition mill grinding. Energy absorbed by the material on impact may result in a modification of the particle surfaces by introducing dislocations, point defects, and other structural defects. This leads to materials with increased surface free energy, which makes them more reactive. Properties, including X-ray diffraction, initial set of time and particle size distribution, are determined. The results indicate that separate grinding of the two materials is most effective at activating the materials and provide the best properties of the paste. D 2003 Elsevier B.V. All rights reserved. Keywords: Waste materials; Attrition mill; X-ray diffraction; Initial set; Separate grinding
Throughout the history of Portland cement manufacturing, the process has generated cement kiln dust (CKD). CKD is a particulate matter entrained in the process air stream, resulting from the tremendous quantities of pulverized rock in the kiln system. Air pollution control equipment prevents it from entering the atmosphere. The United States’ cement industry generates about 5 million tons of cement kiln dust every year that is not recycled, and only 20% of which enters commerce through utilization and the remainder is sent to landfills [1,2]. CKD has been labeled by the Environmental Protection Agency as a high volume, toxic waste stream, and it has been the focus of studies over the last decade in order to determine the appropriate method of regulation in the future. Disposal of waste CKD is not only associated with the problem of land use but also with contamination of ground water from leaching of chemicals from the material. Similarly, another industrial by-product, fly ash, collected from the stack gases of power plants burning pulverized coal, is also commonly considered an extensive waste material. Annually, about 59 million tons of fly ash are generated in the United States [3]. The cement and concrete industry utilize approximately 20% of the total US output of fly ash, while the remainder is placed in landfills. Many fly E-mail address: [email protected] (J. Ryou). 0167-577X/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2003.07.047
ashes cannot be used in concrete because they are of poor and variable quality. Principal contaminants of fly ash include unburned carbon, iron oxides, and, in some cases, calcium sulfates and aluminates. The use of fly ash in cement and concrete may further be affected due to the recent EPA regulations, which results in fly ash produced with higher carbon content. CKD and fly ash have been used as a supplementary cementitious material in concrete in small amounts. Usage is limited due to deficiencies of the two materials. For example, CKD contains an excessive amount of fine particles and a high alkali content, which often impart adverse effects on concrete workability and durability. Most fly ash hydrates slowly and some contain a high unburned carbon content. These inadequacies are unacceptable for strength development and volumetric stability of cement and concrete materials. However, if the two materials are blended together, the alkalis from CKD may activate hydration of fly ash. The research involves properly blending CKD with fly ash to create a cementitious material in which the material deficiencies will be converted into benefits. The purpose of this research is to explore an effective way to substantially utilize CKD and fly ash by developing an environmentally friendly, sufficiently performing, and cost-effective cementitious product for future concrete materials.
904
J. Ryou / Materials Letters 58 (2004) 903–906
Table 1 Chemical composition of the materials Compound
Fly ash
Cement kiln dust
SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K2 O
51.65 23.52 9.39 4.73 1.07 1.42 1.73 2.32
17.67 5.06 2.75 56.99 0.91 6.55 0.30 3.43
One way to activate hydraulic reactivity of cementitious materials is the mechanical process called mechanochemical grinding. The most widely used and best known methods in mechanochemistry include treatment of different types of mill and other similar apparatus [4,5]. These lead to materials with increased surface free energy, which makes them more reactive. Along with grinding in conventional mills, this investigation is proposed to investigate attrition mills. Attrition mills are characterized by higher energy intensity than other types of machinery. A central shaft with arms continually stirs the particles and spherical media and provides a means to vary the grinding energy. In this investigation, the total media charge to powder weight ratio was five to one. The cement kiln dust was supplied by a cement company from a wet kiln process. The chemical composition is shown in Table 1. Cement kiln dust is a fine offwhite or light brown powder. Particles are often random in shape and size, as can be seen in Fig. 1. The kiln dust used is very coarse with a mean particle size of approximately 63 Am. The Class F fly ash was provided by a power plant located in Illinois, as shown in Table 1. Also, fly ash is generally composed of particles called cenospheres, which are round and either hollow or solid particles. Fig. 2 shows fly ash particles under a scanning electron microscope and the cenospheres clearly dominate
Fig. 2. SEM picture of fly ash at 600 magnification.
the composition. The mean particle size of the material is approximately 7.2 Am. The proportioning of cement kiln dust and fly ash was designated for 65% cement kiln dust and 35% fly ash by weight. Table 2 presents the mix combinations and grinding regimes that were tested during the investigation. The following notations are used to differentiate the mixes: G for grinding; F for fly ash; SG for separate grinding; IG for inter-grinding. For example, mix IG-8 in Table 2 was an inter-grinding of cement kiln dust and fly ash for 8 h. After the appropriate grinding regime, mixing was done with a water-to-binder ratio of 0.5. All mixing occurred in a standard paddle mixer. Three tests were used to indicate if the material was a potentially suitable cement replacement: X-ray diffraction, particle size distribution and initial set of time. X-ray diffraction analysis identifies the crystalline and amorphous phases of the material. Particle size distribution discerns the effects of grinding on the particle size. This was performed using a Sympa Tec laser analyzer. Initial set of pastes will be determined using a Vicat apparatus, commonly used in industry. Time of set was measured according to ASTM C 191 ‘‘Standard Test Method for Time of Setting of Hydraulic Cement by Vicat Needle’’.
Table 2 Mix combinations and grinding regimes
Fig. 1. SEM picture of cement kiln dust at 200 magnification.
Mix combination
CKD grinding time (h)
FA grinding time (h)
CKD + FA grinding time (h)
Total grinding time (h)
G-0 FG-4 FG-8 SG-8 SG-12 IG-4 IG-8
0 0 0 4 4 0 0
0 4 8 4 8 0 0
0 0 0 0 0 4 8
0 4 8 8 12 4 8
J. Ryou / Materials Letters 58 (2004) 903–906
905
Fig. 3. X-ray diffraction pattern for cement kiln dust and fly ash.
The X-ray diffraction pattern for cement kiln dust and fly ash was given in Fig. 3. The crystalline material of the cement kiln dust is of much greater intensity than that of the fly ash, while the fly ash has a greater amorphous phase. Also, the diffraction patterns were compared to analyze the extent of crystalline and amorphous material after grinding. It was expected that mechanochemical activation occurred if the bottom of the curve was increased by grinding, indicating a larger amorphous phase which is more reactive, and if there was a reduction in the crystalline peaks of the material. The X-ray diffraction pattern for mix combinations G-0 and SG12 was shown in Fig. 4. While having a small amorphous phase, mix G-0, with no grinding, also has a large quartz peak indicating non-amorphous material. As seen in Fig. 4, 12-h separate grinding of cement kiln dust and fly ash, SG12, was compared to the no ground mix and the result was
also lower crystalline peaks. The crystalline material of mix G-0 is much greater intensity than that mix SG-12, while mix SG-12 has a greater amorphous phase. Grinding changed the particle size distribution. Particles combined and were broken from the impact of the grinding media. Table 3 shows the mean particle size and percent finer than 0.45 Am for all grinding regime. The mean particle size is the size at which 50% of the particles are larger and 50% are smaller. It can be seen that the mean decreased for all of the mixes as the grinding time increased. Furthermore, there are significant decreases in mean particle size when separate grinding is incorporated into the grinding regime, such as SG-8 and SG-12. The corresponding trend is an increase in the percentage of particles finer than 0.45 Am, the lower limit of the laser measurement, as grinding time increases.
Fig. 4. X-ray diffraction pattern for mix combinations G-0 and SG-12.
906
J. Ryou / Materials Letters 58 (2004) 903–906
Table 3 Results of particle size test and initial time of set Mix combination
Mean particle size (Am)
Percent finer than 0.45 Am (%)
Initial time of set (h)
G-0 FG-4 FG-8 SG-8 SG-12 IG-4 IG-8
11.8 7.48 6.71 5.31 5.26 8.47 7.66
2.36 4.13 4.86 5.63 5.96 2.72 3.25
38.5 24.5 22.0 12.0 10.5 16.5 18.0
of the mixes with separate grinding showed much improved results over inter-grinding. However, initial set time does not appear to be dependent on the length of grinding. The longer ground mixes in the inter-grinding did not necessarily give the best results. Since separate grinding provided mixes with the shortest initial set times at all length of grinding, it is clear that separate grinding has some type of activation effect on the materials.
References Initial set time was chosen for testing as it indicates the amount of time available between mixing and setting of the paste; for industrial applications, that is the amount of time available for transportation and placement. Thus, the initial set time of the paste would provide important information for a new material. The results for all grinding regime are shown in Table 3. The no grinding mix, G-0, took over 38 h to reach initial set. Initial set decreased with the use of attrition grinding. Mixes SG-8 and SG-12 have lower set times than the other grinding regimes. All
[1] J.E. Halliday, R.K. Dhir, T.D. Dyer, An investigation of the hydration chemistry, of ternary blends containing cement kiln dust, Journal of Materials Science 34 (20) (1999 October) 4975 – 4983. [2] J.I. Bhatty, Alternative uses of cement kiln dust, Research and Development Information, Portland Cement Association, 1995. [3] J.I. American Concrete Institute (ACI) Committee 232, Use of Fly Ash in Concrete, American Concrete Institute, Detroit, Michigan, 1996. [4] V. Ronin, H. Hedlund, Advanced modification technologies of the portland cement based binders for different high performance applications, Proceedings of the 10th International Congress on the Chemistry of Cement, Gothenburg, Sweden, 1997. [5] J.S. Reed, Principles of Ceramics Processing, 2nd ed., Wiley, New York, 1995.