HIGH RECYCLED WASTE CONTENT ALKALI-ACTIVATED BINDER AS A FULL
REPLACEMENT FOR ORDINARY PORTLAND CEMENT
By
Brian Durkin

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Civil Engineering—Master of Science
2017

ABSTRACT
HIGH RECYCLED WASTE CONTENT ALKALI-ACTIVATED BINDER AS A FULL
REPLACEMENT FOR ORDINARY PORTLAND CEMENT
By
Brian Durkin
Aluminosilicate chemistry details the process known as geopolymerization, where alumina and
silica particles are released from materials in the presence of alkali metals, reorganized and
hardened into a new, amorphous material. Alkali-activated binders, following this chemistry
concept, can be produced in the form of a hydraulic cement, undergoing the hardening phase of
geopolymerization during hydration.

Alkali-activated binders have demonstrated improved

durability and strength properties compared to ordinary Portland cement (OPC). The replacement
of OPC with high recycled waste content alkali-activated binders would reduce the amount of
potentially harmful, stockpiled wastes globally as well as reduce the carbon footprint associated
with the OPC manufacturing process. This work details the process of developing a high recycled
waste content alkali-activated binder with the correct balance of strength, setting time and fresh
mix workability. The results indicated that brick waste and copper tailing can both be effective
aluminosilicate precursors in an alkali-activated binder at up to 20% content by total weight. An
alkali-activated binder containing brick waste was produced on a pilot scale and successfully used
in a concrete pavement field demonstration. The concrete used in the field, containing the alkaliactivated binder, demonstrated a favorable setting time and compressive strength compared to the
requirements for a Portland cement concrete pavement.

ACKNOWLEDGEMENTS

I am thankful for my advisor, Dr. Parviz Soroushian, for his consistent support and encouragement.
I would like to express my gratitude towards to Dr. Anagi Balachandra of Metna Co. for her
endless guidance and assistance throughout my research process. I appreciate the support of the
members of my M.S. guidance committee, Dr. Roozbeh Dargazany and Dr. Weiyi Lu. Throughout
this research, I benefited from discussions with other PhD and MS students working with Dr.
Soroushian including Faris Matalkah, Salina Aizad and Monisha Reddy.

iii

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................................ vii
LIST OF FIGURES ....................................................................................................................... ix
1.

Introduction ............................................................................................................................. 1
1.1. Summary of research ........................................................................................................ 1
1.1.1. Development of an Alkali-Activated Binder Incorporating Construction and
Demolition Brick Waste as an Aluminosilicate Precursor ...................................................... 1
1.1.2. Effects on the Quality of an Alkali-Activated Binder Containing Brick Waste Due
to the Scale Up Process When Manufactured on a Pilot Scale ............................................... 2
1.1.3. Alkali-Activated Binder Used as a Full Replacement for Portland Cement in a
Concrete Pavement: A Field Demonstration ........................................................................... 3
1.1.4. Development of an Alkali-Activated Binder Incorporating Copper Tailing Mine
Waste as an Aluminosilicate Precursor ................................................................................... 4

2. Development of an Alkali-Activated Binder Incorporating Construction and Demolition Brick
Waste as an Aluminosilicate Precursor........................................................................................... 5
2.1. Introduction ...................................................................................................................... 5
2.2. Materials ........................................................................................................................... 7
2.3. Formulation and Processing of Hydraulic Cement .......................................................... 8
2.4. Experimental Methods ................................................................................................... 10
2.4.1. X-Ray Fluorescence ................................................................................................ 10
2.4.2. Thermogravimetric Analysis .................................................................................. 11
2.4.3. Fourier Transform Infrared Spectroscopy .............................................................. 11
2.4.4. Alkalinity ................................................................................................................ 11
2.4.5. Set Time .................................................................................................................. 12
2.4.6. Heat of Hydration ................................................................................................... 13
2.4.7. Compressive Strength ............................................................................................. 13
2.5. Result and Discussion .................................................................................................... 14
2.5.1. Chemical Composition (XRF) ................................................................................ 14
2.5.2. Thermogravimetric Analysis .................................................................................. 14
2.5.3. Fourier Transform Infrared (FTIR) Spectroscopy .................................................. 16
2.5.4. Alkalinity, Total Dissolved Solids, and Conductivity ............................................ 17
2.5.5. Set Time .................................................................................................................. 18
2.5.6. Compressive Strength ............................................................................................. 20
2.5.7. Heat of Hydration ................................................................................................... 24
2.6. Conclusions .................................................................................................................... 25
3. Effects on the Quality of an Alkali-Activated Binder Containing Brick Waste Due to the Scale
Up Process When Manufactured on a Pilot Scale ........................................................................ 27
3.1. Introduction .................................................................................................................... 27
3.2. Materials ......................................................................................................................... 28
3.3. Alkali-Activated Cement Formulation and Processing Methods ................................... 29
iv

3.4. Experimental Methods ................................................................................................... 31
3.4.1. X-Ray Powder Diffraction (XRD) Analysis ........................................................... 31
3.4.2. Particle Size Analysis ............................................................................................. 32
3.4.3. Fourier Transform Infrared Spectroscopy .............................................................. 32
3.4.4. Thermogravimetric Analysis .................................................................................. 32
3.4.5. Heat of Hydration ................................................................................................... 32
3.4.6. Scanning Electron Microscopy (SEM) Analysis .................................................... 33
3.4.7. Compressive Strength ............................................................................................. 33
3.4.8. Setting Time ............................................................................................................ 34
3.5. Results and Discussion ................................................................................................... 34
3.5.1. Alkalinity ................................................................................................................ 34
3.5.2. X-Ray Powder Diffraction (XRD) Analysis ........................................................... 35
3.5.3. Particle Size Analysis ............................................................................................. 36
3.5.4. Fourier Transform Infrared Spectroscopy (FTIR) .................................................. 37
3.5.5. Thermogravimetric Analysis (TGA)....................................................................... 38
3.5.6. Heat of Hydration ................................................................................................... 41
3.5.7. Scanning Electron Microscopy (SEM) Analysis .................................................... 42
3.5.8. Compressive Strength ............................................................................................. 43
3.5.9. Setting Time ............................................................................................................ 45
3.6. Conclusion...................................................................................................................... 45
4. Alkali-Activated Binder Used as a Full Replacement for Portland Cement in a Concrete
Pavement: A Field Demonstration ................................................................................................ 48
4.1. Introduction .................................................................................................................... 48
4.2. Materials ......................................................................................................................... 49
4.3. Alkali Activated Binder Formulation and Processing Methods..................................... 50
4.3.1. Formulation and Processing.................................................................................... 50
4.3.2. Storage of Hydraulic Cement.................................................................................. 51
4.3.3. Quality Control ....................................................................................................... 52
4.4. Field Demonstration ....................................................................................................... 53
4.4.1. Production of Concrete in Ready Mix Plant ........................................................... 53
4.4.2. Field Construction and Testing ............................................................................... 54
4.5. Laboratory Results ......................................................................................................... 57
4.6. Field Performance .......................................................................................................... 57
4.7. Conclusion...................................................................................................................... 58
5. Development of an Alkali-Activated Binder Incorporating Copper Tailing Mine Waste as an
Aluminosilicate Precursor............................................................................................................. 60
5.1. Introduction .................................................................................................................... 60
5.2. Materials ......................................................................................................................... 62
5.3. Hydraulic Cement Formulations and Processing Method.............................................. 64
5.4. Experimental Evaluation of Hydraulic Cements ............................................................ 66
5.4.1. Alkalinity in Solution.............................................................................................. 66
5.4.2. Thermogravimetric Analysis .................................................................................. 67
5.4.3. Fourier Transform Infrared (FTIR) Spectroscopy .................................................. 67
5.4.4. Compressive Strength ............................................................................................. 68
v

5.4.5. Measurements of Heavy Metals Concentrations in Leachated Using Inductively
Coupled Plasma (ICP) Spectroscopy..................................................................................... 69
5.5. Results and Discussion ................................................................................................... 70
5.5.1. Alkalinity in Solution.............................................................................................. 70
5.5.2. Thermogravimetric Analysis (TGA)....................................................................... 70
5.5.3. Fourier Transform Infrared (FTIR) Spectroscopy .................................................. 72
5.5.4. Compressive Strength ............................................................................................. 73
5.5.5. Toxicity Characteristics per LEAF Procedure (ICP) .............................................. 75
5.6. Conclusion...................................................................................................................... 77
BIBLIOGRAPHY ......................................................................................................................... 79

vi

LIST OF TABLES

Table 1. Hydraulic cement formulations ........................................................................................ 9
Table 2. Sizes of the steel balls used in the milling process ......................................................... 10
Table 3. Chemical compositions (wt.%) of aluminosilicate precursors obtained via X-Ray
fluorescence spectroscopy ............................................................................................................ 14
Table 4. pH, total dissolved solids (TDS) and electrical conductivity test results for as-received
and activated waste bricks ............................................................................................................ 18
Table 5. pH, total dissolved solids (TDS) and electrical conductivity test results for alkali
activated cement formulations measured after different time periods .......................................... 18
Table 6. Set time test results for different cements measured at 0.3 water/cement ratio.............. 19
Table 7. Compressive Strength test results ................................................................................... 21
Table 8. Chemical compositions of brick waste, coal fly ash and ground granulated blast furnace
(GGBF) slag .................................................................................................................................. 28
Table 9. Alkali activated cement formulation............................................................................... 29
Table 10. Small and large milling jar dimensions ........................................................................ 30
Table 11. Steel ball milling processing methods .......................................................................... 31
Table 12. The pH, total dissolved solids (TDS), and conductivity test results at 30 minutes and
48 hours ......................................................................................................................................... 35
Table 13. Mean and median particle size values for hydraulic cements processed at laboratory
and pilot scales .............................................................................................................................. 37
Table 14. Compressive strength and fresh mix flow test results generated for standard mortars 45
Table 15. Compressive strength test results for concrete materials .............................................. 45
Table 16. Initial set times of cements processed at laboratory and pilot scales............................ 45
Table 17. The formulation of raw materials (proportions by weight) .......................................... 50
Table 18. Weights of the steel balls of different size .................................................................... 51
Table 19. 7 day compressive strength quality control testing....................................................... 53
Table 20. Concrete mix design ..................................................................................................... 54
vii

Table 21. Compressive strength test results at different ages for concrete specimens prepared in
field and curing in different conditions ......................................................................................... 57
Table 22. Chemical composition of aluminosilicate sources (%)................................................. 63
Table 23. Alternative formulations (wt.%) of raw materials used for mechanochemical
processing of hydraulic cements, and the additives used during mixing...................................... 65
Table 24. pH, total dissolved solids (TDS) and conductivity test results at 30 minutes and 24
hours .............................................................................................................................................. 70

viii

LIST OF FIGURES

Figure 1. Brick waste (a) crushed (b) after activation .................................................................... 7
Figure 2. Laboratory-scale ball mill................................................................................................ 9
Figure 3. Steel balls used in milling process................................................................................. 10
Figure 4. Shaking (left), and measurement of pH, TDS and conductivity (right) ........................ 12
Figure 5. Vicat apparatus .............................................................................................................. 12
Figure 6. TGA data for brick, sodium carbonate, and activated brick.......................................... 15
Figure 7. DTA data for brick, sodium carbonate, and activated brick.......................................... 16
Figure 8. FTIR for as-received brick, sodium carbonate, and brick activated mechanochemically
in the presence of sodium carbonate over different time periods ................................................. 17
Figure 9. Effects of lowering sodium silicate content and adding sodium sulfate on the compressive
strength produced by the hydraulic cement .................................................................................. 20
Figure 10. Mortar cube specimen (a) before compressive strength test (b) after test ................... 21
Figure 11. Compressive strength test results for the hydraulic cement Formulation 10-AB-LBWL
with different additives ................................................................................................................. 23
Figure 12. Water/cement ratios for achieving constant fresh mix workability with the hydraulic
cement Formulation 10-AB-LBWL incorporating different additives ......................................... 23
Figure 13. Rate of heat release during hydration of different hydraulic cements......................... 24
Figure 14. The cumulative heat released during hydration of different hydraulic cements ......... 25
Figure 15. Small (a) and large (b) milling jars.............................................................................. 30
Figure 16. X-Ray powder diffraction (XRD) analysis of the laboratory scale milled cement ..... 35
Figure 17. X-Ray powder diffraction (XRD) analysis of the pilot scale milled cement .............. 36
Figure 18. Cumulative size distributions of hydraulic cements produced at laboratory and pilot
scales ............................................................................................................................................. 37
Figure 19. Fourier transform infrared (FTIR) spectra for the blend of raw materials prior to milling,
and for hydraulic cements processed via laboratory- and pilot-scale milling .............................. 38

ix

Figure 20. Thermogravimetric analysis (TGA) curves for the blend of raw materials prior to
milling, and for the hydraulic cements processed at laboratory and pilot scales .......................... 40
Figure 21. Differential thermal analysis (DTA) curves for the blend of raw materials prior to
milling, and for the hydraulic cements processed at laboratory and pilot scales .......................... 40
Figure 22. The rate of heat released for hydraulic cements processed at laboratory and pilot scales
....................................................................................................................................................... 42
Figure 23. The cumulative heat released for hydraulic cements processed at laboratory and pilot
scales ............................................................................................................................................. 42
Figure 24. Scanning electron microscope (SEM) images at different magnifications of the
hydration products of the cement processed at laboratory scale................................................... 43
Figure 25. Scanning electron microscope images at different magnifications of the hydration
products of the cement processed at pilot scale ............................................................................ 43
Figure 26. The pilot-scale ball mill used for processing of the hydraulic cement ........................ 51
Figure 27. Steel balls of different size used in the milling process .............................................. 51
Figure 28. The type of plastic container used for storage of cement ............................................ 52
Figure 29. Addition of the hydraulic cement to the ready-mix concrete truck ............................. 54
Figure 30. Performing of slump test on fresh concrete ................................................................. 55
Figure 31. Different stages of field construction .......................................................................... 56
Figure 32. Field preparation of concrete specimens for laboratory testing .................................. 56
Figure 33. Surface appearance of the concrete walkway after different periods of exposure to
weathering effects ......................................................................................................................... 58
Figure 34. Copper tailing in as-received condition ....................................................................... 63
Figure 35. X-ray powder diffraction analysis of the copper tailing (Q= Quartz (SiO2), C= Calcite
(CaCO3) ........................................................................................................................................ 63
Figure 36. Steel ball-milling jar and apparatus ............................................................................. 65
Figure 37. Shaking apparatus ........................................................................................................ 66
Figure 38. pH, TDS and Conductivity apparatus .......................................................................... 67
Figure 39. Carver pellet press ....................................................................................................... 68
Figure 40. JASCO FTIR Spectrophotometer ................................................................................ 68
x

Figure 41. Flow table .................................................................................................................... 69
Figure 42. TGA data for hydraulic cement formulations and the copper mine tailing ................. 71
Figure 43. DTA data for hydraulic cement formulations and the copper mine tailing ................. 72
Figure 44. Fourier transform infrared (FTIR) spectroscopy data ................................................. 73
Figure 45. 7 day compressive strength measurements of formulations with additives ................ 74
Figure 46. Flow diameter measurements of formulations with additives..................................... 75
Figure 47. Heavy metal concentrations in the leachate of the hydrated paste B-12-A formulation
at 7 days of age versus the initial concentrations in copper mine tailing ..................................... 76
Figure 48. Percent immobilizations achieved with B-12-A formulation ...................................... 77

xi

1. Introduction

1.1. Summary of research

The motivation for this research was to produce a more sustainable alternative to Portland cement
for concrete production. The approach is based on an alternative cement chemistry which enables
large-volume use of waste raw materials, and adoption of energy-efficient processing techniques.
This dissertation is divided into four chapters; each chapter has been prepared as a stand-alone
paper written for submission to a peer-reviewed scientific journal. For convenience, the references
from each chapter have been consolidated and appear at the end of this thesis. The four main
chapters of the thesis are summarized in the following.

1.1.1. Development of an Alkali-Activated Binder Incorporating Construction and Demolition
Brick Waste as an Aluminosilicate Precursor

Hydraulic cements were formulated with high contents of waste brick (activated by input of
mechanical energy in the presence of sodium carbonate), coal fly ash and sodium silicate. These
cements were processed using energy-efficient mechanochemical means without resorting to
elevated temperatures. They were based on alkali aluminosilicate chemistry, and used sodium
silicate and sodium carbonate as the primary sources of alkali metal (sodium). The hydraulic
cement formulations were refined, primarily by tailoring the raw materials formulations, to achieve
a desired balance of fresh mix workability, set time and hardened material strength. The
experimental results indicated that the combination of mechanochemical processing and alkali
aluminosilicate chemistry enable production of high-recycled-content hydraulic cements with low
1

energy content and carbon footprint for use in concrete construction. The experimental results
generated in this project also provided insight into the structural and chemical changes brought
about by input of mechanical energy onto properly selected blends of raw materials towards
production of hydraulic cements.

1.1.2. Effects on the Quality of an Alkali-Activated Binder Containing Brick Waste Due to the
Scale Up Process When Manufactured on a Pilot Scale

Pilot-scale mechanochemical processing conditions were devised based on theoretical
considerations to input an amount of energy to raw materials that is equivalent to that input in a
laboratory-scale implementation of the mechanochemical process for production of hydraulic
cements. The laboratory-scale process yielded a hydraulic cement based predominantly on the
alkali aluminosilicate chemistry that met standard requirements for General Use hydraulic cement.
The primary raw materials used in mechanochemical processing of this cement included waste
brick, coal fly ash and granulated blast furnace slag; minor quantities of dry alkaline materials
were also used in the process. Deviations from the theoretically determined pilot-scale processing
conditions were imposed by the limited availability of steel balls. Fundamental physico-chemical
and mineralogical characterization of the hydrated cements processed at laboratory and pilot scales
as well as their hydration products indicated that the scaled-up mechanochemical processing of the
hydraulic cement produced similar trends towards transforming the raw materials into a hydraulic
cement. The extent of this transformation in pilot-scale mechanochemical processing, however,
was not as thorough as that achieved in laboratory-scale mechanochemical processing. The
strength development characteristics of the cement processed at pilot scale were thus inferior to
that processed at laboratory scale, considering that the cement processed at pilot scale exhibited
2

an improved rate of strength gain at later ages. The experience gained in this research indicates
that further improvement of the pilot-scale processing conditions would be needed before the
qualities of the cement processed at laboratory scale can be achieved. These improvements would
involve reducing the fraction of the ball-mill jar volume filled with the raw materials and steel
balls, raising the ratio of steel balls to raw materials, and/or increasing the duration of milling.

1.1.3. Alkali-Activated Binder Used as a Full Replacement for Portland Cement in a Concrete
Pavement: A Field Demonstration

An alkali aluminosilicate cement formulated with high contents of waste materials was processed
at pilot scale, and used for industrial-scale production of concrete and implementation of a field
construction project. The concrete mix was designed with the sustainable hydraulic cement used
as full replacement for Portland cement. Field construction was implemented in late October in
mid-Michigan (USA). The hydraulic cement formulation was tailored towards concrete
construction at colder temperatures. The sustainable hydraulic cement was found to be compatible
with industrial-scale production of concrete and conventional field construction practices. Field
performance of the concrete pavement was monitored over one month during which temperatures
dropped below freezing, and substantial rainfall and some light snow were experienced. Concrete
performed quite satisfactorily during this period, exhibiting no signs of cracking or surface
damage. Compressive strength tests were performed on concrete specimens prepared in field and
cured either in field or in controlled laboratory environment. The strengths achieved in both curing
conditions met the 78- and 28-day compressive strength requirements for concrete materials used
in pavement construction. The experience gained in this project provided support for the scalability

3

of the hydraulic cement developed based on alkali aluminosilicate cement chemistry, and its
potential for full replacement of Portland cement in concrete construction.

1.1.4. Development of an Alkali-Activated Binder Incorporating Copper Tailing Mine Waste as
an Aluminosilicate Precursor

Hydraulic cements were formulated with high contents of copper tailing, coal fly ash and sodium
silicate. These cements were processed using energy-efficient mechanochemical means without
resorting to elevated temperatures. They were based on alkali aluminosilicate chemistry, and used
sodium silicate and sodium carbonate as the primary sources of alkali metal (sodium). The
hydraulic cement formulations were refined, primarily by tailoring the raw materials formulations,
to achieve the immobilization of heavy metals and a desired balance of fresh mix workability and
hardened material strength. The experimental results indicated that the combination of
mechanochemical processing and alkali aluminosilicate chemistry enable production of highrecycled-content hydraulic cements with low energy content, low carbon footprint and the ability
to immobilize heavy metals for use in concrete construction. The experimental results generated
in this project also provided insight into the structural and chemical changes brought about by
input of mechanical energy onto properly selected blends of raw materials towards production of
hydraulic cements.

4

2. Development of an Alkali-Activated Binder Incorporating Construction and Demolition
Brick Waste as an Aluminosilicate Precursor

2.1. Introduction

The manufacturing of Portland cement is an energy-intensive process which also involves
chemical decomposition of limestone, resulting in significant emissions of carbon dioxide [1, 2].
The technical work in progress focuses on development of more sustainable hydraulic cements
that suit concrete production. The work reported herein focused on formulation of high-recycledcontent hydraulic cements using waste brick.
The use of waste brick as supplementary cementitious material to partially replace Portland
cement (PC) clinker has been explored [3] in order to gain sustainability benefits [4, 5]. Similar
efforts with other waste and natural materials have not qualitatively reduced the high carbon
footprint and energy content of Portland cement. A departure from the conventional Portland
cement chemistry, based primarily on calcium silicates, could enable high-impact reduction of the
carbon footprint and energy content of the hydraulic cements used in concrete production [6, 7].
In recent years, the activation of fly ash, blast furnace slag, glass, and brick waste has been
investigated [8-11]. Replacement of 10-35% of Portland cement with these wastes has led to
improvements in the sustainability and some performance attributes of Portland cement concrete
[12]. A departure from the calcium silicate-based chemistry of Portland cement towards a
primarily alkali aluminosilicate chemistry promises to significantly raise the use of
byproducts/wastes as raw materials for production of hydraulic cements of significantly improved
sustainability that meet standard requirements for use in concrete production [13].

5

The focus of this work was on formulation of the raw materials for production of an alkali
aluminosilicate cement using waste brick as an aluminosilicate precursor. Waste brick constitutes
a major constituent of the construction and demolition (C&D) waste worldwide, comprising more
than half of the C&D waste generated in some regions [13]. Waste brick is also produced as
defective units in brick production plants. The dust generated during cutting of brick is another
source of waste brick. The large landfilled quantities of waste brick rationalize efforts to develop
markets for this major constituent of the C&D waste stream. The use of waste brick in cement
industry would yield important cost and sustainability benefits [14].
Clay, a major source of aluminosilicates, is the primary raw material used for production of brick.
During high-temperature firing of brick, a liquid phase develops, which subsequently forms a
glassy phase upon cooling. It is this glassy phase that binds the crystalline and any other phases
that make up brick [15]. Ground waste brick has been evaluated as a pozzolan in production of
Portland cement concrete [16]. The use of ground calcined brick clay as a pozzolanic material has
also been explored [17]. These effort promises to bring about improvements in the performance
and sustainability of Portland cement concrete [18]. The large landfilled quantities of waste brick
[19] rationalize efforts to develop markets for this constituent of the C&D waste stream.
Allahverdi and Kani [14] investigated alkali activation of construction waste brick using a mixture
of sodium hydroxide and sodium silicate (with a silica modulus of 0.6). A formulation with 8%
Na2O content with respect to the precursor weight produced a compressive strength of 40 MPa
after 28 days of curing. The work reported herein focused on the use of waste brick and other
abundant wastes of supplementary chemistry and mineralogy as raw materials (aluminosilicate
precursors) in sustainable production of a hydraulic cement based on the alkali aluminosilicate
chemistry.

6

2.2. Materials

Waste brick was collected from a waste management facility in mid-Michigan. Coal fly ash was
obtained from the Erickson power plant managed by the Lansing Board of Water & Light (LBWL)
in Lansing, Michigan. Ground granulated blast furnace (GGBF) slag was obtained from
LafargeHolcim (Ohio, USA). Clay was acquired from three different sources: (i) Greenstripe clay
from IONEMaterials; (ii) Kaolinite clay from Sigma Aldrich; and (iii) Red clay from Red Soil
Dirt Clay in East Tennessee (USA). These clays were calcined at different temperatures. Sodium
silicate, sodium carbonate, calcium oxide, and borax were purchased from Sigma-Aldrich in
powder form. Graded silica sand (ASTM C778) was used as fine aggregate in production of
standard mortar mixtures per ASTM C1157.

Figure 1. Brick waste (a) crushed (b) after activation

7

2.3. Formulation and Processing of Hydraulic Cement

Viable raw materials formulations were selected via trial-and-adjustment. Alkali aluminosilicate
hydrates can assume different stoichiometric attributes, and may also be supplemented with other
(e.g., calcium silicate) hydrates as inorganic binders. Reliance strictly on the chemical composition
of the raw materials for their formulation (proportioning) may be misleading because some
minerals within raw materials may not be made available by simple processing techniques which
avoid energy-intensive and polluting processing at elevated temperatures. Table 1 presents
alternative formulations of raw materials, devised initially based on chemical considerations and
modified via preliminary trial-and-adjustment efforts. Screening of these raw materials
formulations is a primary consideration in the work reported herein.
Processing of the blend of raw materials into hydraulic cement was accomplished in this work via
mechanochemical processing without resorting to elevated temperatures. Mechanochemistry
refers to chemical and physico-chemical changes of substances induced by input of mechanical
energy (e.g., milling) [20]. Initially, the waste brick was milled with sodium carbonate at a 4:1
ratio and with 1% moisture added. This combination and the remaining byproduct/waste
aluminosilicate precursors were milled together with sodium silicate and quick lime as alkali
activators. Mechanochemical processing of raw materials was accomplished using a conventional
ball mill
Figure 2). A steel ball: raw materials weight ratio of 8: 1 was used, with a milling duration of 2
hours. Minor quantities of some additives were blended with the hydraulic cement after milling.

8

No
1

Cement
Notation
1-ABLBWL

Table 1. Hydraulic cement formulations
Raw Material
Ash Brick Slag Na2SiO3 Na2CO3 CaO
45

20

20

7

5

Other

3

2

2-ABLBWL

30

20

24

7

6

3

10
Greenstripe
Clay

3

3-ABLBWL

40

20

20

7

5

3

5
Greenstripe
Clay

4

4-ABLBWL

40

20

20

7

5

3

5 Red Clay

45

20

20

7

5

3

0.5 Borax

45

20

20

7

5

3

1 Borax

45

20

20

7

5

3

45

20

20

7

5

3

48

20

20

4

5

3

48

20

20

4

5

3

5
6
7
8
9
10

5-ABLBWL
6-ABLBWL
7-ABLBWL
8-ABLBWL
9-ABLBWL
10-ABLBWL

Figure 2. Laboratory-scale ball mill

9

Notes

1 Borax
1 Na2SO4
1 Borax
2 Na2SO4

1 Borax

Clay
calcined
(12hr,
750°C)
Clay
calcined
(2 hr,
750°C)
Clay
calcined
(2 hr,
750°C)

Figure 3. Steel balls used in milling process

Table 2. Sizes of the steel balls used in the milling process
Ball Size Diameter (cm) Mass (g)
Small
2.50
65
Medium
3.50
200
Large
5.00
500

2.4. Experimental Methods

The test methods employed for characterizing the raw materials and the resulting hydraulic
cements are described in the following.

2.4.1. X-Ray Fluorescence

The chemical compositions of raw materials were assessed via x-ray fluorescence (XRF)
spectroscopy using a Bruker S2 Ranger XRF spectrometer. Loss on ignition (LOI) was measured
following a single-step procedure per ASTM D7348.

10

2.4.2. Thermogravimetric Analysis

Thermogravimetric analysis (TGA) is a method of determining the sample composition, including
volatiles and inert fillers. In this test, the sample mass loss with increasing temperature is
monitored, and compared with the thermal data for pure phases that are probably present in the
sample. The mass loss curve obtained is transformed into the derivative form where weight losses
are more visible. TGA was carried out using Perkin Elmer TGA 4000 at a heating rate of 15˚C/min
from 50 to 950oC under nitrogen atmosphere. For this purpose, about 100 mg of sample was heated
at a rate of 10˚C/min in nitrogen environment.

2.4.3. Fourier Transform Infrared Spectroscopy

FTIR spectroscopy was carried out using a JASCO FTIR Spectrophotometer. The wavenumber
range covered 400 cm-1 to 4000 cm-1. The spectra were recorded after 50 scans with a spectral
resolution of 4 cm-1.

2.4.4. Alkalinity

The alkalinity of hydraulic cements was measured by performing pH tests per ASTM C25. In this
test, 5g of cement was mixed with 45mL of deionized (DI) water, and shaken for 30 minutes. After
shaking, the solution was allowed to stand still for 30 minutes before taking measurements. The
solution was then shaken for 24 hours, and the measurement was repeated. The pH of solution was
measured together with the total dissolved solids (TDS) and electrical conductivity. Figure 4 shows
the setups used for shaking of the solution, and measurement of its pH, TDS and conductivity.

11

Figure 4. Shaking (left), and measurement of pH, TDS and conductivity (right)

2.4.5. Set Time

The set time of hydraulic cements was determined per ASTM C191. This test was performed using
a Vicat apparatus (Figure 5). The sample was prepared with a water/cement ratio of 0.30. The
hydraulic cement and any set retarder considered were mixed dry for 1 minute prior to mixing with
water for two minutes. Measurements were taken every 5 minutes after the start of hydration. The
final set time was recorded when the needle could not penetrate the sample beyond 25mm.

Figure 5. Vicat apparatus

12

2.4.6. Heat of Hydration

Heat of hydration tests were performed as per ASTM C186. Paste samples were prepared with a
water/cement ratio of 0.5, using 10g of hydraulic cement and 5g of deionized water. Immediately
after mixing, the paste samples were placed in the apparatus to begin measurements. A Calmetrix
I-Cal 2000 HPC was used to record measurements for approximately 72 hours.

2.4.7. Compressive Strength

Hydraulic cements were characterized following the ASTM C 1157 requirements. The ASTM
C109 procedures were followed to prepare mortar specimens with each hydraulic cement for
performance of compressive strength tests. The standard mortar mixtures used in these tests
comprised hydraulic cement: standard silica sand at 1: 2.75 weight ratio. The water/cement ratios
of these mixtures were adjusted to achieve a standard flow of at least 19mm. To improve the flow,
additives such as borax were mixed into the hydraulic cement. Water/cement ratios ranged from
0.49 to 0.54 for achieving the standard fresh mix workability. The mixing procedure involved
2 minutes of mixing the dry ingredients (cement, sand and additives) in a mixer at slow speed,
followed by the addition of water and 3 more minutes of mixing at medium speed. The fresh mix
was cast into molds and consolidated via external vibration. The specimens were moist-cured
inside molds at room temperature for 24 hours. They were then demolded, and cured at room
temperature in sealed condition (>95% relative humidity) until the test ages of 3 and 7 days.

13

2.5. Result and Discussion

2.5.1. Chemical Composition (XRF)

Table 3 presents the chemical compositions of the waste raw materials used in this investigation.
The waste brick is observed to comprise largely of silica (56.90%), alumina (17.01%), iron oxide
(13.02%), and potassium oxide (5.88%). All these ingredients, if made available, can contribute to
the active chemistry of hydraulic cements. The coal fly ash considered here has a calcium oxide
content of 22.2%, which qualifies it as an ASMT C618 Class C fly ash. The silica-to-alumina ratio
in the composition of his fly ash is close to 2. In GGBF slag comprised large of calcium oxide and
silica, which accounted for 78% of its weight. The alumina and magnesium oxide contents of this
slag were 9.54% and 10.85%, respectively.

Table 3. Chemical compositions (wt.%) of aluminosilicate precursors obtained via X-Ray
fluorescence spectroscopy
Waste Raw
SiO2 Al2O3 CaO SO3 Fe2O3 MgO Na2O K2O TiO2 P2O5
Material
C&D Brick
56.80 17.01 1.52 0.84 13.02
5.88 2.60 0.92
Waste
Coal fly ash
30.6 15.1 22.2
16.8 2.30
2.60
GGBF slag
37.5 9.54 40.8
0.43 10.85
0.45
-

2.5.2. Thermogravimetric Analysis

The TGA and DTA data for waste brick, sodium carbonate, and brick activated
mechanochemically with sodium carbonate over different time periods are presented in Figure 6
and Figure 7, respectively. The as-received waste brick exhibits a minor weight loss with a DTE
peak around 730oC, which could be a result of the presence of minor carbonates in brick. Sodium
14

carbonate exhibits relatively sharp peaks at about 100oC and 880oC associated with loss of
adsorbed water and decomposition of carbonate, respectively. Mechanochemical activation of
waste brick with sodium carbonate over 1 and 2 hours alters the TGA/DTA patterns of brick, and
increases its weight loss. Considering that mechanochemical activation was performed with water
added to solids a 1 wt.%, a DTA peaks appear below 100oC, with a higher temperature associated
with the longer mechanochemical activation. Broad peaks also appear at about 800 and 780oC for
waste bricks activated with sodium carbonate via 1 and 2 hours of milling. Activated bricks also
exhibited more than 10% weight loss, compared to about 1% weight loss of the as-received waste
brick. This rise in weight loss as well as the appearance of new peaks which shift with the duration
of milling point at changes that occur in the primarily aluminosilicate structure of brick as a result
of mechanochemical activation in the presence of sodium carbonate.

100
99
98
97

Weight (%)

96
95
94
93
92
91
90

brick as is
Na2CO3
brick-Na2CO3-1%water-2H milled
brick-Na2CO3-1%water-1H milled

89
88
50 100

200

300

500
600
400
50
Temperature (°C)

700

800

900

1000

Figure 6. TGA data for brick, sodium carbonate, and activated brick

15

100
0.0
-0.2

Derivative weight (% min)

-0.4
-0.6
-0.8
-1.0
-1.2
-1.4
-1.6
-1.8
50 100

brick as is
Na2CO3
brick-Na2CO3-1%water-2H milled
brick-Na2CO3-1%water-1H milled

200

300

400
500
50
Temperature (°C)

600

700

800

900

Figure 7. DTA data for brick, sodium carbonate, and activated brick

2.5.3. Fourier Transform Infrared (FTIR) Spectroscopy

Figure 8 shows the FTIR spectra of as-received brick, sodium carbonate, and brick activated
mecahnochemically in the presence of sodium carbonate and 1% water over different time periods.
The as-received brick shows the typical 1043 cm−1 band corresponding to asymmetric Si-O-Si
stretching vibration [21]. Sodium carbonate strong absorption bands at 1420 cm-1 and 881 cm-1,
and weak bands at 2970 cm-1, 1795 cm-1, 701 cm-1 and 694 cm-1 caused by CO32- [22]. The
activated brick powder contains a small amount of moisture (1%), producing a peak at 2970 cm−1
which corresponds to stretching of hydroxyl groups. The FTIR spectrum of the activated brick
exhibited a shift in the carbonate peak to 1445 cm-1. Similarly, the band corresponding to Si-O
stretching was shifted slightly to 1070 cm−1, further indicating a change in the chemical bond
environment of Si-O in brick due to milling with sodium carbonate. These shifts corroborate the
TGA trends, indicating that mechanical activation brings about fundamental structural and

16

chemical changes in the blend of raw materials considered here. Other sodium carbonate peaks
could be detected in the FTIR spectra of activated bricks. In addition, the peak around 3600 cm-1
to OH groups of physiosorbed water [23].

Brick

%T

Na2CO3

1043cm-1

Brick+Na2CO3-1%
water (2H)

1420cm-1

2970cm-1

Brick+Na2CO3-1%
2964cm-1
water (1H)

870cm-1

1070cm-1
1445cm-1
1081cm-1
1452cm-1

3600

3100

2600

2100

1600

1100

600

Figure 8. FTIR for as-received brick, sodium carbonate, and brick activated mechanochemically
in the presence of sodium carbonate over different time periods

2.5.4. Alkalinity, Total Dissolved Solids, and Conductivity

The pH, total dissolve solids and electrical conductivity test results for as-received and activated
bricks are summarized in Table 4. The values of pH, TDS and conductivity increase upon
activation of brick; this trend, however, can be attributed partly to the presence of sodium
carbonate in activated brick. Activation of brick could also contribute to this trend by making the
brick constituents more soluble. The results obtained for the hydraulic cements developed in this
investigation are presented in Table 5, with measurements taken after 24 and 48 hours. These
hydraulic cements produce high pH values which is reflective of their raw materials formulations

17

and mechanochemical processing condition. Their total dissolved solids and conductivity in
solution are also relatively high. The trends in evolution of these values with time are not consistent
between different types of hydraulic cement developed here. This reflects the opposing trends of
dissolution of cement constituents, and their subsequent hydration reactions and precipitation out
of solution.

Table 4. pH, total dissolved solids (TDS) and electrical conductivity test results for as-received
and activated waste bricks
Material
pH TDS, ppm Conductivity, ms/cm
As-received brick
10.49
131.5
0.260
Activated brick-Na2CO3-1%water-1H 11.48
1992
4.068
Brick-Na2CO3-1%water-2H
11.48
1776
3.622

Table 5. pH, total dissolved solids (TDS) and electrical conductivity test results for alkali
activated cement formulations measured after different time periods
Material
30min
48 hrs
pH
TDS, Conductivity,
pH
TDS , Conductivity,
ppm
mS/cm
ppm
mS/cm
2-AB-LBWL
13.19
388
0.792
13.33
3472
7.085
4-AB-LBWL
13.17
7248
14.79
13.28
6305
12.87
6-AB-LBWL (40min 13.12
1448
2.955
13.34
7276
14.85
Pilot-Scale Mill)
7-AB-LBWL
13.12
4458
9.097
13.26
2251
4.593
8-AB-LBWL
13.17
5352
10.92
13.21
7076
14.44
10-AB-LBWL
13.15
5027
10.26
13.23
4576
9.338

2.5.5. Set Time

Hydraulic cements based on the alkali aluminosilicate chemistry generally exhibit high initial rate
of reactions and thus relatively low set time. Sodium sulphate and borax were investigated to
increase the set time of hydraulic cements. The set time test results are summarized in Table 5.
Formulation 1-AB-LBWL, which did not incorporate borax, produced a set time of 10 minutes
18

when blended with 1 wt.% of sodium sulfate, and 16 minutes when blended with 1 wt.% sodium
sulfate and 1 wt.% borax. The retarding effect of borax can be attributed to the formation of a
dense calcium borate layer wrapping cement particles at early stages of hydration, making them
less reactive temporarily. The presence of borax has some negative effect on compressive strength;
the minor drop at 1 wt.% borax, however, is justified by the significant rise in set time. When
sodium sulfate was milled together with other raw materials (at 1 wt.% in 7-AB-LBWL, and 2
wt.% in 8-AB-LBWL), the resulting set times were 22 and 43 minutes, respectively. These
significant gains in set time, however, were accompanied with significant drops in compressive
strength. Another attempt made to increase set time involve reducing the sodium silicate content
of the hydraulic cement from 7% to 4%, and increasing the fly ash content instead (Formulation
10-AB-LBWL). This adjustment raised the set time significantly, and also produced high levels of
compressive strength (Figure 9). The addition of 1 wt.% sodium sulfate to this formulation
significantly raised its set time and also benefited its compressive strength; this adjustement,
however, was not required because Formulation 10-AB-LBWL produce an desired set time of 50
minutes without the addition of sodium sulfate.

Table 6. Set time test results for different cements measured at 0.3 water/cement ratio
Set
Added During
Notation
Formulation
Time,
Mixing
min
1-AB45% Ash, 25% Activated brick, 20% Slag, 7%
1% sodium
10
LBWL
Sodium silicate, 3% CaO
sulfate
2-AB30% Ash, 26% Activated brick, 24% Slag, 10%
1% borax
25
LBWL
Greenstripe clay, 7% Sodium silicate, 3% CaO
4-AB40% Ash, 25% Activated brick, 20% Slag, 7%
27
LBWL
Sodium silicate, 5% Red clay, 3% CaO
1% sodium
16
sulphate
6-AB45% Ash, 25% Activated brick, 20% Slag 7%
LBWL
Sodium silicate, 3% CaO, 1% Borax
1% borax, 1%
30
sodium sulphate
19

Table 6. (cont’d)

7-ABLBWL
8-ABLBWL

10-ABLBWL

45% Ash, 25% Activated brick, 20% Slag, 7%
Sodium silicate, 3% CaO, 1% Borax, 1% Sodium
sulphate
45% Ash, 25% Activated brick,20% Slag, 7%
Sodium silicate, 3% CaO, 1% Borax, 2% Sodium
sulphate

48% Ash, 25% Activated brick, 20% Slag, 4%
Sodium silicate, 3% CaO, 1% Borax

1% sucrose

17

-

22

-

43

1% sodium
sulphate
1% CaO
1% sodium
carbonate
5% slag

50
85
35
40
40

90

Setting Time (minutes)

80
70
60
50
40
30
20
10
0

7% sodium silicate &
1%sodium sulphate

4% sodium silciate & 1%
soidum sulphate

4% sodium silciate

Formulation

Figure 9. Effects of lowering sodium silicate content and adding sodium sulfate on the
compressive strength produced by the hydraulic cement

2.5.6. Compressive Strength

The compressive strength test results are presented in Table 7 together with the fresh mix flow test
data for different hydraulic cement formulations. A desired balance of fresh mix workability and
20

hardened material strength is provided by the hydraulic cements based on the 9-AB-LBWL and
10-AB-LBWL formulations. These two formulations are very similar, except that 10-AB-LBWL
incorporates 1 wt.% borax for raising the set time. Introduction of borax is observed to have minor
effects on compressive strength and fresh mix flow. The distinct feature of these two formulations
is their reduced sodium silicate content (4 wt.%) when compared with that in other formulations
(7 wt.%). Figure 11 shows the effectiveness of reducing the sodium silicate content of the
formulation, versus other measures, in raising the compressive strength obtained with the hydraulic
cement. Other benefits of the reduced sodium silicate include increased set time and fresh mix
flow, and favorable economics. Figure 12 shows the effectiveness of reducing the sodium silicate
content, versus other measures, in lowering the water/cement ratio required for producing a
constant fresh mix workability. Increasing the dosage of sodium sulfate, which was used as a set
retarder, produced lower compressive strengths. The clays considered in this investigation in asreceived or calcined condition did not yield notable benefits.

Figure 10. Mortar cube specimen (a) before compressive strength test (b) after test

Notation

Table 7. Compressive Strength test results
Compressive Strength,
Added During Mixing
w/c
MPa
21

Flow,
cm

Table 7. (cont’d)
3d

1-ABLBWL

2-ABLBWL

3-ABLBWL
4-ABLBWL
5-ABLBWL
6-ABLBWL
7-ABLBWL
8-ABLBWL
9-ABLBWL
10-ABLBWL

7d
15.79

28d

None
Added 15% fly ash, using 85% of original
formulation
Added 15% fly ash, using 85% of original
formulation and 0.5% Borax
Added 15% fly ash, using 85% of original
formulation and 1% Borax
None
10% Red clay
5% Red Clay
5% Red Clay and 1% Borax
1% Borax
1.5% Borax
5% Greenstripe clay as received
5% Greenstripe clay calcined 3 hours
0.2% Amine and 2% Sodium Benzonate
1% Borax, 0.2% Amine and 2% Sodium
Benzonate
None
1% Borax

0.50

1% Borax

0.53

None

0.49

11.49

None
5% Na2SO4

0.54
0.50

12.58

17.80
15.44

19
19

None

0.50

10.30

18.0

19

None

0.51

10.10

13.70

19

None

0.50

12.9

21.2

19

None

0.50

12.47

20.89

19

22

14

0.50

6.89

16

0.50

7.74

16

0.50

7.32

16

0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.53

14.48
8.98
12.37
12.36
16.0
11.60
12.61
10.19
10.70

0.50

8.79

19

0.50
0.50

14.38
11.74

14.5
16

19.67
11.19
16.80
16.83

14.62

14
17
17
17
18
17
16
18
19

20.5
21.5

Compressive strength (MPa)

25
20
15

3 Day

10

7 Day

5
0

Base

1% Sodium
Sulphate

2% Sodium
Sulphate

Reduced
Sodium Silicate

Adjustment

Figure 11. Compressive strength test results for the hydraulic cement Formulation 10-AB-LBWL
with different additives

Water cement ratio (%)

0.55

0.50

0.45

0.40

0.35

Base

1% Sodium
Sulphate

2% Sodium
Sulphate

Reduced Sodium
Silicate

Adjustment

Figure 12. Water/cement ratios for achieving constant fresh mix workability with the hydraulic
cement Formulation 10-AB-LBWL incorporating different additives

23

2.5.7. Heat of Hydration

Figure 13 and Figure 14 compare the rate of heat release and the cumulative heat released versus
time, respectively, for hydraulic cement Formulations 6-AB-LBWL (with 7% sodium silicate
content) and 10-AB-LBWL (with 4% sodium silicate content). Lowering the sodium silicate
content is observed to reduce the initial rate of heat release, but to elongate the period of active
exothermic hydration reaction. The cumulative heat released for the hydraulic cement with lower
sodium silicate content exceeds that of the cement formulated with higher sodium silicate content.
These trends exhibited by the cement with reduced sodium silicate content are generally preferred
because it provides working time prior to setting for timely implementation of the steps involved
in concrete construction.

0.08

Heat evolution rate (W/g)

0.07
0.06
0.05
0.04

6-AB-LBWL

0.03

10-AB-LBWL

0.02
0.01
0.00

0.00

1.00

2.00

3.00

4.00

5.00

Time (Hours)

Figure 13. Rate of heat release during hydration of different hydraulic cements

24

Cumulative heat of hydration (J/g)

180
160
140
120
100

6-AB-LBWL

80

10-AB-LBWL

60
40
20
0

0.00

20.00

40.00

60.00

80.00

Time (hours)

Figure 14. The cumulative heat released during hydration of different hydraulic cements

2.6. Conclusions

Hydraulic cements based on alkali aluminosilicate chemistry can be formulated with waste brick
(activated mechanochemically in the presence of sodium carbonate), coal fly ash and blast furnace
slag used as raw materials. These hydraulic cements used a combination of sodium silicate and
calcium oxide as the primary alkali activators (and supplementary sources of silicon and calcium),
and were processed using mechanochemical effects without resorting to elevated temperatures.
The original formulations incorporated relatively high sodium silicate content, and provided
limited set time which compromises their ability to produce concrete materials with adequate fresh
mix working time to enable timely implementation of construction steps. Some generic set
retarders, that are compatible with the alkali aluminosilicate cement chemistry, were found to be
effective in elongating the set times without compromising the strength development ability of the
hydraulic cements. Lowering the sodium silicate content, however, was found to produce the best

25

combination of set time and compressive strength among the measures considered in this
investigation. Reduced sodium silicate contents lowered the initial rate of hydration reactions, but
elongated the period of active reactions.
Structural and chemical changes brought about by input of mechanical action onto judiciously
selected blends of materials can produce hydraulic cements with desired balance of qualities for
use in production of concrete. There seems to be a synergy between the alkali aluminosilicate
cement chemistry and mechanochemical processing for production of a new class of highrecycled-content hydraulic cements with significantly reduced energy content and carbon
footprint.

26

3. Effects on the Quality of an Alkali-Activated Binder Containing Brick Waste Due to the
Scale Up Process When Manufactured on a Pilot Scale

3.1. Introduction

Mechanochemistry [20] offers a sustainable alternative to energy-intensive processing of inorganic
materials, including hydraulic cements, at elevated temperatures. Mechanochemical processing of
inorganic materials can be accomplished using mills (e.g., ball mills) that are used commonly in
industrial processes for size reduction [24]. Mechanochemistry refers to the chemical and
physiochemical changes of substances induced by input of mechanical energy (e.g., via milling)
[20]. Steel ball milling generates energy through repeated transfer of impact energy at contact areas
between steel balls and raw materials.
Laboratory-scale investigations have demonstrated that ball-milling of a blend of aluminosilicate
precursors and alkaline compounds can induce mechanochemical effects that transform the raw
materials into hydraulic cements based on alkali aluminosilicate chemistry [25]. The work reported
herein was conducted to verify the scalability of the mechanochemical approach to processing of
hydraulic cements. This work involved a scale-up of the mechanochemical process from laboratory
scale with about 1 kg/hr production capacity to pilot scale with about 100 kg/hr production
capacity. The operation condition of the scaled-up ball mill was selected based on theoretical
considerations to input similar mechanical energy to raw materials as the laboratory-scale mill.
Scalability of the process was assessed by comparing the physical and mineralogical
characteristics as well as the strength development attributes of the hydraulic cements produced at
laboratory and pilot scales using similar raw materials.

27

3.2. Materials

Coal fly ash was obtained from the Erickson power plant operated by the Lansing Board of Water
& Light (LBWL) in Lansing, Michigan (USA). This fly ash with the chemical composition
presented in Table 8 (obtained via x-ray fluorescence spectroscopy) qualifies as ASTM C618 Class
C. The waste brick used in this study was collected from a landfill in mid-Michigan (USA). The
as-received brick waste was crushed to give a granular material of about 75 mm maximum particle
size. Ground granulated blast furnace (GGBF) slag was obtained from LafargeHolcim in Ohio
(USA). The chemical composition of this slag, obtained via x-ray fluorescence spectrostopy, is
presented in Table 8. Sodium silicate was purchased from JJ Chemicals in Fort Benning, GA
(USA). Sodium carbonate was purchased from Duda Energy LLC in Decatur, Alabama (USA).
Calcium oxide was purchased from Pete Lien and Sons Inc in Rapid City, South Dakota (USA).
Borax was purchased from Sigma-Aldrich in powder form. Graded standard silica sand (ASTM
C778) was used as fine aggregate to prepare mortar mixtures. Natural sand and crushed limestone
of 19 mm maximum particle size, meeting the ASTM C33 requirements, were used as fine and
coarse aggregates for preparation of concrete mixtures.

Table 8. Chemical compositions of brick waste, coal fly ash and ground granulated blast furnace
(GGBF) slag
Waste Raw
SiO2 Al2O3 CaO SO3 Fe2O3 MgO Na2O K2O TiO2 P2O5
Material
C&D Brick
56.80 17.01 1.52 0.84 13.02
5.88 2.60 0.92
Waste
Coal fly ash
30.6 15.1 22.2
16.8 2.30
2.60
GGBF slag
37.5 9.54 40.8
0.43 10.85
0.45
-

28

3.3. Alkali-Activated Cement Formulation and Processing Methods

For both the large and small mills, the raw materials presented in Table 9 were processed
mechanochemically via ball-milling (using steel balls) to produce a hydraulic cement powder with
a predominantly alkali aluminosilicate chemistry. As a first step, the brick waste was activated via
milling with sodium carbonate at a waste brick: sodium carbonate 4:1 weight ratio, with 1 wt. %
moisture. The remaining raw materials were then added to the mill, and further milling was
performed to produce the hydraulic cement.

Coal Fly
Ash
48%

Table 9. Alkali activated cement formulation
Sod.
Sod.
Calc.
GGBF Brick Waste
Carbonate
Silicate
Oxide
20%

20%

5%

4%

3%

Borax
1% wt of
cement

The appropriate milling time for production of hydraulic cement was first found through trial-andadjustment studies at laboratory scale. The goal was to find the minimum milling durations that
produce hydraulic cements that meet the primary ASTM C1157 requirements for the ‘General
Use’ (GU) type of hydraulic cement. The scale-up efforts were guided by the criterion that inputs
of mechanical energy at laboratory and pilot scales should be similar. For this purpose, the milling
duration was expressed in terms of the milling jar rotational speed-to-critical speed ratio (N),
depth-to-diameter ratio (β), powder (raw materials) volume-to-jar volume ratio (S), and x is the
scale-up index defined as the mass ratio of raw materials (powder) in the pilot-scale mill to that in
the laboratory-scale mill [26]. The consideration that the laboratory- and pilot-scale milling
processes should input similar energy to raw materials in order to induce comparable
mechanochemical effects, the required milling time (t) for the pilot-scale mill could be calculated
29

in terms of the milling duration of the laboratory-scale mill using Equation 1 below (where the
subscript 0 refers to the laboratory-scale mill) [26].

t

t0

N

1.3

= � N0 �

∗ �x ∗

β0 −0.2

�
β

S

∗ � S0 �

−1.2

(1)

The laboratory and pilot scale mills used in this investigation are shown in Figure 15, with the
corresponding dimensions presented in Table 10.

Figure 15. Small (a) and large (b) milling jars

Table 10. Small and large milling jar dimensions
Mill Type
Depth (cm) Diameter (cm) Volume (cm^3)
Laboratory Scale
21
28
12,930
Pilot Scale
91
61
265,940

The laboratory-scale mill was operated using an 8:1 steel ball: powder weight ratio and 2 kg of
powder, and required a total milling duration of 120 minutes to transform the blend of raw
materials into a hydraulic cement. Using Equation 1, it was found that with an 8:1 steel ball:
powder ratio, approximately 20 minutes would be required for pilot-scale transformation of raw
materials into hydraulic cement. Due to limits on the availability of balls, only a 4:1 steel ball:

30

powder ratio could be used in pilot-scale milling. In order to accommodate this difference in ratios,
the milling period was extended. Activation of brick required 20 minutes because the low amount
of powder (brick alone) allowed for achieving a steel ball-to-power weight ratio of 8:1; after
addition of the remaining raw materials, with a ball-to-powder weight ratio of 4: 1, an additional
milling duration of 40 minutes was required (verified using a trial-and-adjustment approach which
focused on production of cements with adequate strength development characteristics). The
operation conditions of the laboratory- and pilot-scale mills are summarized in Table 11.

Mill Scale
Laboratory
Pilot

Capacity,
kg
2
40

Table 11. Steel ball milling processing methods
Brick Activation
Additional Milling Duration
Time, min
of All Raw Materials, min
60
60
20
40

Total Milling
Duration, min
120
60

3.4. Experimental Methods

3.4.1. X-Ray Powder Diffraction (XRD) Analysis

The hydraulic cements processed at laboratory and pilot scales were subjected to x-ray powder
diffraction analysis using a Rigaku Miniflex XRD instrument. The resulting XRD spectra were
analyzed in order to identify the existing crystalline phases, and to assess the degree of
crystallinity.

31

3.4.2. Particle Size Analysis

The particle size distributions of the hydraulic cements were assessed using a Horiba LA-920 laser
particle size analyzer. Three replicated tests were performed for each hydraulic cements, and the
mean particle size distributions were obtained.

3.4.3. Fourier Transform Infrared Spectroscopy

The samples were subjected to Fourier transform infrared (FTIR) spectroscopy using a JASCO
FTIR spectrophotometer. The wavenumbers considered covered the 400 cm-1 to 4000 cm-1 Range.
The spectra were recorded after 250 scans with a spectral resolution of 4 cm-1.

3.4.4. Thermogravimetric Analysis

Thermogravimetric analysis (TGA) data were produced for the hydraulic cements produced at
laboratory and pilot scaled using a Perkin Elmer TGA 4000 instrument operated at a heating rate
of 10˚C/min from 50 to 950oC under a nitrogen atmosphere.

3.4.5. Heat of Hydration

Heat of hydration tests were performed per ASTM C186. Paste samples were prepared with a
water/cement ratio of 0.50, and tested using a Calmetrix I-CAL 2000 HPC test instrument over 72
hours in order to record the heat of hydration.

32

3.4.6. Scanning Electron Microscopy (SEM) Analysis

Scanning electron microscopy (SEM) was performed using a Neoscope JCM-5000 instrument on
paste specimens with water/cement ratio of 0.33 after 7 days of curing at room temperature.

3.4.7. Compressive Strength

The strength development characteristics of the hydraulic cements processed mechanochemically
at laboratory and pilot scales were characterized per ASTM C 1157. The ASTM C109 procedures
were followed to prepare mortar specimens for performance of compressive strength tests. The
standard mortar mixtures used in these tests comprised hydraulic cement: standard silica sand at
1: 2.75 weight ratio. The water/cement ratios of these mixtures were adjusted in order to achieve
a standard flow of 110 ± 5% (ASTM C1437). To improve the flow, additives such as borax and
clays were introduced to the hydraulic cement during mixing. The required water/cement ratio
ranged from 0.46 to 0.50. The mixing procedure comprised 2 minutes of mixing the dry ingredients
(cement and sand) in a rotary mixer at slow speed, followed by the addition of water and 3 more
minutes of mixing at medium speed. The fresh mix was cast into molds and consolidated via
external vibration. The molded specimens were stored in sealed condition at room temperature for
24 hours. They were then demolded, and cured at room temperature in sealed condition (>95%
relative humidity) until the test ages of 3 and 7 days.

33

3.4.8. Setting Time

Set time was measured following the ASTM C191 procedures using a Vicat apparatus. The
mixture comprised 300g of hydraulic cement and 90g of water, which were mixed for
approximately 2 minutes. The fresh mix was placed in the Vicat apparatus mold; a measurement
was taken every 5 minutes until the initial set time was reached.

3.5. Results and Discussion

3.5.1. Alkalinity

The pH, total dissolved solids (TDS) and electrical conductivity test results obtained for solutions
incorporating hydraulic cements processed at laboratory and pilot scales are presented in Table 12.
Measurements were taken 30 minutes and 48 hours after addition of cement to water. The trends
in TDS and conductivity indicate that the hydraulic cement processed at pilot scale are lower after
30 minutes and higher after 48 hours when compared with the cement processed at laboratory
scale. The pH values, however, are higher for the cement processed at pilot scale for measurements
taken at both 30 minutes and 48 hours when compared with the cement processed at laboratory
scale. One purpose of mechanochemical processing is to transform alkalis and aluminosilicates
into alkali aluminosilicates that can undergo timely dissolution and hydration upon addition of
water. The initially slow dissolution of the cement processed at pilot scale is not necessarily a
drawback since it accelerates beyond 30 minutes. The higher pH value of the cement processed at
pilot scale, however, could point at the relatively low efficiency of transforming the raw materials
into a hydraulic cement.
34

Table 12. The pH, total dissolved solids (TDS), and conductivity test results at 30 minutes and
48 hours
30 min
48 Hour
Mill Scale
TDS,
Conductivity,
TDS,
Conductivity,
pH
pH
ppt
mS/cm
ppt
mS/cm
Laboratory 13.15
5.027
10.26
13.23
4.576
9.338
Pilot
13.26
3.925
8.01
13.41
7.35
15.01

3.5.2. X-Ray Powder Diffraction (XRD) Analysis

The XRD patterns presented in Figure 16 and Figure 17 for the hydraulic cements processed at
laboratory and pilot scales are similar, pointing at the presence of quartz as the only identifiable
crystalline species, with other compounds occurring in amorphous form.

Figure 16. X-Ray powder diffraction (XRD) analysis of the laboratory scale milled cement

35

Figure 17. X-Ray powder diffraction (XRD) analysis of the pilot scale milled cement

3.5.3. Particle Size Analysis

The cumulative particle size distributions of hydraulic cements processed at laboratory and pilot
scales are shown in Figure 18; the mean and median values of particle size for the two cements are
presented in Table 13. Pilot-scale processing produced hydraulic cements with coarser particle size
when compared with the laboratory-scale processing the median particle sizes of hydraulic
cements processed at laboratory and pilot scales were 6.63 and 9.44 µm, respectively. This
observation implies that pilot-scale milling was not as efficient as laboratory-scale milling in terms
of size reduction. The initially low dissolution rate of the hydraulic cement processed at pilot scale
can be explained using the coarser particle size distribution of this cement. It is worth mentioning
that the particle size distribution achieved in pilot-scale processing is comparable to that of
conventional Type I Portland cement. Finally, one should distinguish between the comminution
and mechanochemical effects of milling; there is not necessarily a direct proportionality between
these two milling effects [27].

36

Figure 18. Cumulative size distributions of hydraulic cements produced at laboratory and pilot
scales

Table 13. Mean and median particle size values for hydraulic cements processed at laboratory
and pilot scales
Particle Size Laboratory Scale Pilot Scale
Mean (μm)
7.62
10.59
Median (μm)
6.63
9.44

3.5.4. Fourier Transform Infrared Spectroscopy (FTIR)

The FTIR spectra of the blend of raw materials prior to milling, and for hydraulic cements
processed at laboratory and pilot scales are presented in Figure 19. Milling at both laboratory and
pilot scales brings about major changes in the FTIR spectra of the blend of raw materials. Milled
hydraulic cements exhibited peaks at the 1000 cm-1 - 1150 cm-1 region, which corresponds to
polymerized silica. The hydraulic cement processed at laboratory scale produced a more notable
peak in this region. Both cements exhibited peaks in the 1400 cm-1 - 1600 cm-1 region, which points
at the presence of carbonates. These carbonates could have formed via reactions involving the
37

carbon dioxide in air during mechanochemical processing. The peak in this region is more defined
for the cement processed at pilot scale. A peak around 3600 cm-1,, which is more prounounced for
the cement processed at laboratory scale, is indicative of the presence of hydrates (or water) which
could have formed as a result of the adsorption of moisture from air and probably the consequent
hydration reactions.

Figure 19. Fourier transform infrared (FTIR) spectra for the blend of raw materials prior to
milling, and for hydraulic cements processed via laboratory- and pilot-scale milling

3.5.5. Thermogravimetric Analysis (TGA)

The TGA test data and the resulting differential thermal analysis (DTA) curves are presented in
Figure 20 and Figure 21. Mechanochemical processing (milling) of the blend of raw materials is
observed to bring about notable changes in the TGA test results. The changes in TGA test results
are somewhat more pronounced after laboratory-scale milling when compared with pilot-scale
milling, which confirms the observations made earlier regarding the more advanced state of
mechanochemical transformation realized via laboratory-scale milling. The DTA peaks observed
38

at 100 - 150°C point at the presence of physically adsorbed water and some less stable hydrates.
This peak is less pronounced for the cement processed at laboratory scale. It should be noted that
mechanochemical activation of waste brick was performed in the presence of a minor quantity of
water. This could be the reason why the blend of raw materials (prior to milling), which
incorporated non-activated brick, has a smaller DTA peak in this region. One of the raw materials,
borax (sodium tetraborate decahydrate) incorporates chemically bound water, and could contribute
to the DTA peak at the 100 – 150oC region. A second set of DTA peaks appeared within the 400 500°C temperature range, which point at the presence of hydroxides. These hydroxides may have
formed during milling due to the use of minor amount of water for mechanochemical activation of
waste brick, or after milling due to hydration reactions of the resultant cement with the moisture
present in air. This peak is also more pronounced for the cement subjected to laboratory-scale
processing. The peaks observed at 700 - 900°C can be attributed to the presence of carbonates. At
approximately 700°C all materials exhibited peaks which are likely due to the presence of sodium
carbonate. This peak is more pronounced for the cement processed at laboratory scale.
Additionally, between 800°C and 900°C peaks were observed for the blend of raw materials (prior
to milling), and for the cement processed at pilot scale. The carbonates which produce these peaks
may have formed via reactions of the raw materials constituents with the carbon dioxide present
in air, which would be pronounced during milling and also upon formation of reactive compounds
after milling.

39

Figure 20. Thermogravimetric analysis (TGA) curves for the blend of raw materials prior to
milling, and for the hydraulic cements processed at laboratory and pilot scales

Figure 21. Differential thermal analysis (DTA) curves for the blend of raw materials prior to
milling, and for the hydraulic cements processed at laboratory and pilot scales

40

3.5.6. Heat of Hydration

The heat of hydration test results are presented in Figure 22 and Figure 23. The hydraulic cement
processed at pilot scale exhibited a relatively high initial rate of heat release, which could not be
retained over time. For both cements, a well-defined heat release peak occurred after only few
minutes; the cement processed at laboratory scale exhibited a less defined peak after about one
hour. The cumulative heat release for the cement processed at pilot scale was, after 3 days, about
70% less than that processed at laboratory scale. This finding is consistent with all previous results
which indicate that pilot-scale processing of the cement needs adjustment before it can produce a
cement quality that, similar to the cement processed at laboratory scale, meets the ASTM C1157
strength development requirements. These adjustments relate to the fraction of the jar volume
filled with raw materials and steel balls, the weight ratio of balls to raw materials, and the milling
duration. It should be noted that relatively high initial rate of heat release noted for the cement
processed at pilot scale, in spite of its relatively low overall reactivity, can be attributed to the
greater availability of the less compounded alkalis in this cement. There is a relatively high rate of
heat release associated with the rapid release of the alkali metal cations to the solution, and the
subsequent alkali attack on aluminosilicate precursors.

41

Figure 22. The rate of heat released for hydraulic cements processed at laboratory and pilot
scales

Figure 23. The cumulative heat released for hydraulic cements processed at laboratory and pilot
scales

3.5.7. Scanning Electron Microscopy (SEM) Analysis

The scanning electron microscope images captured at different magnifications for the hydration
products of cements processed at laboratory and pilot scales are shown in Figure 24 and Figure 25,
respectively. The hydration process of the cement processed at laboratory scale seems to be more
thorough, with seemingly amorphous products of hydration filling the voids and rendering binding
42

effects (Figure 24). These trends seem to be less advanced in the case of the hydraulic cement
processed at pilot scale (Figure 25).

(a) 1000x
(b) 1500x
(c) 2000x
Figure 24. Scanning electron microscope (SEM) images at different magnifications of the
hydration products of the cement processed at laboratory scale

(a) 1000x
(b) 1500x
(c) 2000x
Figure 25. Scanning electron microscope images at different magnifications of the hydration
products of the cement processed at pilot scale

3.5.8. Compressive Strength

ASTM C1157 requires standard mortar 3- and 7-day compressive strengths of 13 and 20 MPa,
respectively, for ‘General Use’ (GU) hydraulic cements. The compressive strength test results
presented in Table 14 indicated that the standard mortar made with the cement processed at
laboratory scale met these requirements (within ±1%). The cement processed at pilot scale,

43

however, failed to meet the ASTM C1157 requirements for General Use hydraulic cement. This
was particularly true at 3 days of age, but the rate of strength gain picked up between 3 and 7 days.
The flow of the fresh mortar mix prepared with the cement processed at pilot scale was somewhat
higher than that prepared with the cement processed at laboratory scale. The shortfalls of the
cement processed at pilot scale can be feasibly overcome by reducing the fraction of the jar filled
with the raw materials and steel balls, raising the steel ball-to-raw materials weight ratio, and/or
increasing the duration of pilot-scale milling. The duration of air storage of the cement processed
at pilot scale prior to testing was longer than that for the cement processed at laboratory scale.
Storage of cement in bagged/sealed condition is another measure that could benefit the strength
development qualities of the cement processed at pilot scale.
The compressive strength test results produced for concrete materials are summarized in Table 15.
Similar to mortar mixtures, the fresh mix workability of the concrete mixtures prepared with the
cement processed at pilot scale was better than that for the cement processed at laboratory scale.
This allowed for lowering the water/cement ratio of the concrete prepared with the cement
processed at pilot scale without compromising its fresh mix workability. In spite of this, the cement
produced at pilot-scale produced concrete with significantly reduced 3-day compressive strength
when compared with the cement processed at laboratory scale. The rate of concrete strength rise,
however, increased between 3 and 7 days. The concrete compressive strength test results follow
trends similar to those of standard mortar compressive strengths. The measures suggested for
improving the pilot-scale processing of cement would thus be the same as those devised based on
the standard mortar compressive strength test results.

44

Table 14. Compressive strength and fresh mix flow test results generated for standard mortars
Compressive Strength, MPa
Cement Processing
Flow,
W/C
Scale
cm
3-day
7-day
Laboratory-Scale 0.50
12.9
21.2
19
Pilot-Scale
0.50
6.00
14.3
20

Table 15. Compressive strength test results for concrete materials
Compressive Strength, MPa
Cement Processing
W/C
Scale
3-day
7-day
Laboratory-Scale 0.45
23.8
29.8
Pilot-Scale
0.35
7.4
21.9

3.5.9. Setting Time

ASTM C1157 requires initial set times occurring within the 45 to 420 minutes range for General
Use hydraulic cements. The initial set time test results presented in Table 16 indicate that cements
processed at both laboratory and pilot scales meet this requirement. The lower reactivity of the
cement processed at pilot scale actually benefits its ability to retain the fresh mix workability by
elongating its set time.

Table 16. Initial set times of cements processed at laboratory and pilot scales
Type
Set Time, min
Small Mill
50
Big Mill
60

3.6. Conclusion

A scaled-up milling process was developed to reproduce the energy input of laboratory-scale
milling that successfully produced a mechanochemically processed hydraulic cement based on the

45

alkali aluminosilciate chemistry that met the ASTM C1157 requirements. Some deviations from
the theoretical scaled-up conditions were imposed (particularly in terms of the weight ratio of steel
balls to raw materials) due to different constraints (e.g., availability of steel balls). The duration of
the pilot-scale milling was adjusted to account for the deviations from the theoretical scale-up
requirements. The specific hydraulic cement that was subject of pilot-scale processing comprised
predominantly of waste aluminosilicate precursors (waste brick, coal fly ash, and granulated blast
furnace slag). Relatively small quantities of dry alkaline materials were also used as raw materials.
Waste brick required mechanochemical activation in the presence of an alkaline material and
minor moisture prior to blending with other raw materials for mechanochemical production of
hydraulic cement.
The trends in evolution of the physico-chemical, mineralogical and hydraulic activity of the
cement produced at pilot scale were similar to those produced at laboratory scale. Laboratory-scale
mechanochemical processing of the hydraulic cement, however, produced higher levels of specific
surface area, solubility, reactivity, and strength development characteristics. Improvements that
would benefit the qualities of the hydraulic cement processed at pilot scale include lowering the
fraction of the jar volume filled with the raw materials and steel balls, raising the weight ratio of
steel balls to raw materials, and increasing the milling duration.
The hydraulic cement processed at pilot scale fell significant short of the cement processed at
laboratory scale in terms of 3-day strength development. At 7 days of age, however, the strength
obtained with the cement processed at pilot scale was closer to (through still smaller than) that
obtained with the cement at laboratory scale.

46

The trends in concrete strength development characteristics with cements processed at laboratory
and pilot scales were similar to those achieved with the standard hydraulic cement mortar used in
evaluation of the strength development attributes of hydraulic cements per ASTM C1157.
The particle size distribution of the cement processed at pilot scale was coarser than that of the
cement processed at laboratory scale, and the cumulative (exothermic) release of heat of hydration
was considerably smaller for the cement processed at pilot scale. These findings further indicate
that the energy input of the pilot-scale milling process was not adequate to render the physicochemical effects that effectively transform the raw materials into a hydraulic cement.
The set times of the hydraulic cements processed at laboratory and pilot scales met the ASTM
C1157 requirement for ‘General Use’ hydraulic cements. The fresh ix workability obtained with
the cement processed at pilot scale was better than that processed at laboratory scale. The
water/cement ratio required for the cement processed at pilot scale is thus lower than that processed
at laboratory scale. This could well be due to the smaller specific surface area and lower reactivity
of the cement processed at pilot scale.

47

4. Alkali-Activated Binder Used as a Full Replacement for Portland Cement in a Concrete
Pavement: A Field Demonstration

4.1. Introduction

The manufacturing of Portland cement is an energy-intensive process that results in a significant
carbon footprint [1, 2]. Efforts have been made to replace Portland cement as the binder in concrete
with the objective of mitigating the environmental impacts associated with manufacturing of
Portland cement [3-5]. The alkali aluminosilicate chemistry has been an option considered for
development of a sustainable class of hydraulic cements [28-33]. This cement chemistry offers
several advantages over the conventional Portland cement chemistry, including enhanced
durability, a reduced carbon footprint and energy content [33-36]. Fly ash, granulated blast furnace
slag and metakaolin can be used as aluminosilicate precursors in the conventional approach to
production of alkali aluminosilicate binders via alkali activation. The robust chemistry of alkali
aluminosilicate cements allow for the use of broad categories of (waste or natural) inorganic
materials as aluminosilicate precursors. The large-volume use of wastes as raw materials in
production of alkali aluminosilicate cements should address the need to ensure consistent
satisfaction of the targeted performance requirements in spite of the variability of waste raw
materials [37].

Binders based on alkali aluminosilicate chemistry have been developed largely as aluminosilicate
precursors that are transformed into a solid binder by the addition of highly alkaline solutions. This
“two-part” approach to producing an inorganic binder deviates from the common practice of
producing a solid binder using Portland cement where water (in lieu of alkaline solutions) induces
48

hydration reactions. In spite of this, the important sustainability and performance advantages of
this class of binder has led to its growing applications in airfield [38] and highway [39, 40]
pavements, and infrastructure repair [35, 41-44].

The work reported herein is part of efforts to devoted to development of hydraulic cements based
on alkali aluminosilicate chemistry that can transform into a solid binder by simple addition of
water and curing at room temperature. This work involved production of an alkali aluminosilicatebased hydraulic cement at pilot scale, its use for production of concrete in a ready-mix truck, and
construction of a walkway using common field practices. The particular hydraulic cement
considered here was formulated with waste brick as one aluminosilicate precursor. The field
construction was implemented in mid-Michigan on October 18, 2017.

4.2. Materials

Waste brick, coal fly ash and ground granulated blast furnace slag were used as aluminosilicate
precursors. The coal fly ash used for this investigation was obtained from the Erickson Power Plant
operated by the Lansing Board of Water & Light (LBWL) in Lansing, Michigan (USA). The waste
brick used for this study was obtained from a landfill in mid-Michigan (USA). As-received waste
brick was crushed to produce particles of about 75 mm size. Ground ganulated blast furnace
(GGBF) slag was obtained from LafargeHolcim in Ohio (USA). The alkaline constituents of raw
materials included sodium carbonate, sodium silicate and calcium oxide. Sodium silicate was
purchased from JJ Chemicals in Fort Benning, Georgia (USA). Sodium carbonate was purchased
from Duda Energy LLC in Decatur, Alabama (USA). Calcium oxide was purchased from Pete
Lien and Sons Inc in Rapid City, South Dakota (USA). Borax was purchased from Sigma-Aldrich
49

in powder form. Graded standard silica sand (ASTM C778) was used as fine aggregate to prepare
standard mortar mixtures. Concrete mixtures used natural sand and coarse aggregate of 19 mm
maximum particle size, that were provided by High Grade Materials Co. of Lansing, Michigan
(USA).

4.3. Alkali Activated Binder Formulation and Processing Methods

4.3.1. Formulation and Processing

The raw materials formulation used for processing of hydraulic cement is introduced in Table 17.
These raw materials were mechanochemically processed via steel ball-milling using the pilot-scale
mill shown in Figure 26. This mill measures 91cm in depth and 61cm in diameter. The steel balls
(Figure 27) used in the mill were of three different sizes (Table 18). As a first step in
mechanochemical processing of raw materials, brick waste, sodium carbonate and 1% moisture
were milled for 20 minutes to activate the waste brick. Subsequently, the remaining raw materials
( of Lansing, Michigan (USA).) were added to the mill, and milling was continued for an additional
40 minutes. This milling operation yielded 40 kg of binder. The pilot-scale mill was operated
seventeen times to produce 680 kg of cement for use in the field construction project.

Coal Fly
Ash
48%

Table 17. The formulation of raw materials (proportions by weight)
Brick
Sodium
Sodium
Calcium
GGBF
Waste
Carbonate
Silicate
Oxide
20%

20%

5%

4%

50

3%

Borax
1% wt of
raw
materials

Figure 26. The pilot-scale ball mill used for processing of the hydraulic cement

Figure 27. Steel balls of different size used in the milling process

Table 18. Weights of the steel balls of different size
Ball type Mass, g Diameter, cm
Small
65
2.5
Medium
200
3.5
Large
500
5.0

4.3.2. Storage of Hydraulic Cement

The hydraulic cement was produced in the period from 9/18/2017 through 10/18/2017. In order to
reduce the chance of reactions between the moisture in air and he hydraulic cement during the
storage time of up to 1 month, the manufactured cement was stored in a plastic container (Figure
28).

51

Figure 28. The type of plastic container used for storage of cement

4.3.3. Quality Control

During the pilot-scale production of hydraulic cement, samples were taken every other day (during
weekdays) for the purpose of quality control through preparation of standard mortars and
performance of set tome and compressive strength tests per ASTM C403 and ASTM C109. The
7-day compressive strength test results summarized in Table 19 are the mean values of
compressive strengths measured for three standard cubic mortar specimens produced with the
cement samples obtained on a daily basis. The compressive strength test results exhibit random
variations, occurring mostly within the 11-14 MPa range at 7 days of age. These values are about
40% smaller than those obtained with similar cements processed at laboratory scale. The milling
conditions at laboratory scale were more favorable, particularly as far as the weight ratio of steel
balls to the raw materials is concerned. The transformation of raw materials into a hydraulic cement
was thus more advanced under the laboratory-scale milling conditions considered prior to this
scale-up effort. It was judged that the quality of the cement processed at pilot scale was acceptable
for construction of sidewalks in the field project that was subject of this investigation. Set times
were measured in outdoor conditions in preparation for the field construction work. The initial set
time was measured at 60 minutes for outdoor temperatures of about 18oC. and increased to 120
minutes as the outdoor temperature dropped to about 10oC. As an adjustment for construction at
52

lower temperatures, the sodium silicate content in the formulation of raw materials was increased
from 4% to 5% in order to accelerate the setting and the early-age strength gain processes. This
additional 1% sodium silicate content, however, could not be added during mechanochemical
processing of the hydraulic cement. It was simply blended with the original hydraulic cement
considering that the temperature during field construction was about 10oC.

Table 19. 7 day compressive strength quality control testing
7-Day
Cement
Compressive
Sample
Strength, MPa
1
11.00
3
14.16
5

12.30

7

12.60

9

14.35

11

10.06

13

10.90

4.4. Field Demonstration

4.4.1. Production of Concrete in Ready Mix Plant

The hydraulic cement was transported to a ready-mixed concrete plant operated in Lansing,
Michigan (USA) by High Grande Materials Co. The concrete mix design used for the field
construction work is introduced in Table 20. The ready-mix concrete truck was first loaded with
about half of the required quantities of natural sand and coarse aggregate. The hydraulic cement
was then added (Figure 29), followed by the remaining natural sand and coarse aggregate, and
finally water and a normal dosage of an air entraining agent. The total volume of concrete prepared
53

was close to 2 cubic meter. Mixing in the ready-mix truck followed the procedures commonly used
for normal Portland cement concrete.

Table 20. Concrete mix design
Material
Kg/m3
Alkali activated binder
445
Natural sand
1029
Limestone coarse aggregate 1305
Water (w/c = 0.35)
156

Figure 29. Addition of the hydraulic cement to the ready-mix concrete truck

4.4.2. Field Construction and Testing

The concrete prepared in the ready-mixed concrete plant was transported over about 30 kilometer
to an apartment complex in East Lansing, Michigan. Concrete was poured at about 4:45 pm at a
temperature of 10-15oC in sunny condition. The slump of concrete was measured at 89 mm (Figure
30), and its initial set time at 90 minutes. The fresh concrete was placed and finished conveniently
although its bulk rheology and surface characteristics could be characterized as more cohesive than
a normal Portland cement concrete. Figure 31 shows different stages of construction; compatibility
54

of the concrete produced with the new hydraulic cement with conventional field construction
practices was verified through successful construction of the sidewalk.
Concrete specimens were prepared (Figure 32) using samples of fresh concrete obtained from the
ready-mix truck for testing in laboratory.

Figure 30. Performing of slump test on fresh concrete

55

Figure 31. Different stages of field construction

Figure 32. Field preparation of concrete specimens for laboratory testing

56

4.5. Laboratory Results

Compressive strength tests were performed on the cylindrical samples taken from the field
concrete mix. Half of the samples were cured in outdoor condition, and the other half in a
controlled laboratory environment in sealed condition. Compressive strengths were measured at 2,
7 and 28 days of age. The Michigan Department of Transportation specifications for concrete
pavements require minimum compressive strengths of 17.9 MPa and 24.1 MPa after 7 and 28 days
of curing, respectively. The compressive strength results summarized in Table 21 indicate that
both the outdoor and controlled curing conditions produced compressive strengths that meet the
requirements for use in pavement construction.

Table 21. Compressive strength test results at different ages for concrete specimens prepared in
field and curing in different conditions
Age, days
Curing Condition
2
7
28
Outdoor

2.85 MPa 19.4 MPa 24.2 MPa

Controlled Indoor 7.31 MPa 21.0 MPa 28.9 MPa

4.6. Field Performance

The outdoor temperature during the first month after construction ranged from -6oC to 21oC; the
field work was also exposed to substantial rainfall and some light snow during this period. No
obvious cracking or surface deterioration was detected after one month. The surface appearances
of the sidewalk after different periods of exposure to weathering effects are shown in Figure 33.
The walkway surface appeared to change color over time, starting from a brown, and converting
to increasingly bright grey. It should be noted that the cylindrical specimens cured outdoors
57

exhibited this same color change while those cured in controlled laboratory condition retained their
original brown color.

Figure 33. Surface appearance of the concrete walkway after different periods of exposure to
weathering effects

4.7. Conclusion

A high-recycled wastes content alkali aluminosilicate cement was manufactured at pilot scale for
use in industrial-scale production of concrete and performance of a field demonstration project.
The cement was used as a full replacement for Portland cement in concrete.
The cement developed in the project exhibited longer set times and lower rates of strength
development at colder temperatures. The hydraulic cement formulation was tailored to suit
concrete construction at colder temperatures. The tailored formulation exhibited adequate retention
of fresh mix workability for industrial-scale production of concrete, its transportation to the job
site, and timely implementation of field construction steps.

58

The concrete walkway constructed with the sustainable cement was exposed to moderate to
freezing temperatures as well as substantial rainfall and some light snow during a one-month
period after construction. No cracking or damage can be observed on the concrete surface.
Concrete specimens were prepared in field, and cured either in field or in controlled laboratory
condition. While the colder field curing produced somewhat reduced compressive strengths, both
field and laboratory curing produced 7- and 28-day compressive strengths which meet the
requirements relevant to pavement applications of concrete materials.
The experience gained with pilot-scale production of cement, industrial-scale production of
concrete, and field construction of a walkway point at the scalability of the high-recycled-content
alkali aluminosilicate cement, and its compatibility with industrial-scale production of concrete
and conventional field construction practices. The cement developed in the project offers the
potential for full replacement of Portland cement in concrete construction.

59

5. Development of an Alkali-Activated Binder Incorporating Copper Tailing Mine Waste
as an Aluminosilicate Precursor

5.1. Introduction

Environmental concerns are major issues in the mining industry. Current mining practices generate
significant quantities of wastes (mine tailings), including the ore processing wastes. The toxic
constituents of these wastes and their current disposal methods in tailing dams and ponds pose
hazards to the environment. Contamination of the surrounding soils and groundwater supplies are
examples of such hazards. When stockpiled, mine tailings can alter pH levels of soil, turning them
acidic, which has strong negative impacts on the surrounding ecosystem [45]. Mine wastes have
accumulated in tremendous quantities across the world. Value-added and large volume
applications, which simultaneously immobilize their toxic constituents, would yield significant
sustainability benefits.
Copper is a metal with vast applications that is mined worldwide and especially in the United
States [46]. The United States Geological Survey (USGS) has indicated that there is a growing
demand for copper; the global consumption of copper increased by 39.5% between the years 2003
and 2016 [46]. The United States Geological Survey’s most recent publication indicates that 19.4
million tons of copper were produced globally, and that reserves totaled 720 million tons [46]. A
past study by Robert Gordon indicated that the quality of copper ores (the percentage of usable
copper) has continuously decreased over time; the ratio of waste materials per copper ore mined
is thus increasing with time [47]. Today, copper ore is estimated to be composed of just 0.5 - 2%
copper metal with the remains being tailings [45, 47, 48]. Based on these numbers, between 970
million and 3.88 billion tons of copper tailings were produced globally in 2016 alone. Given the
60

decreasing quality of copper ores, this supply of tailings will continue to rise at an increasing rate
if the mining of copper continues to increase as it has for the past several decades.
Prior studies have indicated that copper tailings, when used as a partial substitution for Portland
cement in concrete, provide improved strength, abrasion resistance and water impermeability [49,
50]. The hazardous heavy metals often found in copper tailings have also been found to stabilize
as a result of hydration of cement in concrete [49]. A past investigation that is more relevant to
this work has indicated that a geopolymer binder based on copper tailings and fly ash could achieve
compressive strengths greater than 21 MPa at 7 days of age. This study also noted that the copper
tailings were highly reactive and therefore made significant contributions to the strength
development of concrete during the first 7 days of curing [51].
Alkali-activated cements constitute a large family of inorganic binders which solidify upon
exposure of various aluminosilicate precursors to alkaline solutions. The aluminosilicate
precursors commonly used in alkali-activated binders include coal fly ash, slag and kaolinite clay;
mine tailings have also been used in this application [52-54]. The alkalis used for activation of
these aluminosilicate precursors include sodium hydroxide and sodium silicate [52, 53].
Depending on the specifics of their formulation and curing condition, alkali-activated binders can
provide competitive or superior strength, impermeability, chemical resistance, fire resistance and
sustainability features when compared with Portland cement concrete [55, 56]. Alkali-activated
binders have a more robust chemistry, and can be produced using broader selections of raw
materials. This is an important advantage that allows for potential use of some market-limited
wastes (generally as aluminosilicate precursors) in alkali-activated binders. At the same time, one
should be aware of the variations in material properties resulting from the use of alternative
aluminosilicate precursors in alkali-activated binders.

61

5.2. Materials

Coal fly ash was used as an aluminosilicate precursor in this investigation. The fly ash was
obtained from the Erickson power plant operated by the Lansing Board of Water and Light
(LBWL) in Lansing, Michigan. The chemical composition of this ash is presented in Table 22.
This fly ash qualifies as Class C fly ash per ASTM C618
Ground granulated blast furnace slag, obtained from LafargeHolcim, Ohio, was used as another
aluminosilicate precursor. Table 22 presents the chemical composition of this slag.
The copper tailing used as an aluminosilicate precursor was acquired from Montana Resources
Co. in Butte, Montana. Montana Resources mines copper ore and, through a smelting and floating
process, separates copper from the tailing. The copper tailing was shipped with a sandy
morphology as seen in Figure 34. The crystalline structure of the copper tailing was examined with
an x-ray powder diffraction (XRD) analysis and the results are presented in Figure 35. The XRD
analysis indicates that the copper tailing is a crystalline material with large amounts of quartz and
some calcite present. The chemical composition and the mineralogy of the copper tailing is
presented in Table 22. The copper tailing was composed highly of silica oxide (SiO2) and alumina
oxide (Al2O3) which suggests that it could be an excellent aluminosilicate precursor.

62

Figure 34. Copper tailing in as-received condition

9000

Q

8000
7000

Intensity

6000
5000
4000
3000
2000
1000

Q

Q Q
C

Q

Q

Q

Q

10
12
15
17
19
21
24
26
28
30
33
35
37
39
42
44
46
48
51
53
55
57
60
62
64
67
69
71
73
76
78

0
2θ (degree)

Figure 35. X-ray powder diffraction analysis of the copper tailing (Q= Quartz (SiO2), C= Calcite
(CaCO3)

Table 22. Chemical composition of aluminosilicate sources (%)
Waste Raw SiO2 Al2O3 CaO SO3 Fe2O3 MgO Na2O K2O TiO2 P2O5
Material
Copper
44.28 18.14 <2.72 0.72 3.78
8.67 0.81 1.73 0.87
Tailings
Coal fly
30.6
15.1
22.2
16.8
2.30
2.60
ash
Waste
61.63 1.79 20.71 0.78 4.55
1.2
5.9 1.02 0.18 0.77
Glass
63

Table 22. (cont’d)
GGBF slag

37.5

9.54

40.8

-

0.43

10.85

-

0.45

-

-

Waste glass was obtained from the Granger Recycling Center in Lansing, Michigan [10, 57-61].
Kaolinite [15-17], sodium silicate, sodium hydroxide, sodium carbonate, calcium oxide, borax and
citric acid were purchased from Sigma-Aldrich in powder form. A polycarboxylate-based
superplasticizer was purchased from Fritz-Pak Corporation (Dallas, Texas). A graded standard
silica sand (ASTM C778) was used as fine aggregate for mortar samples.

5.3. Hydraulic Cement Formulations and Processing Method

Multiple cement formulations were investigated in this investigation. Each formulation was
designed to maximize the contents of waste materials while producing a desired balance of set
time and strength per ASTM C1157 while providing adequate moisture stability.
Mechanochemical processing was used to convert the raw materials into a hydraulic cement based
predominantly on alkali aluminosilicate chemistry. Mechanochemical processing refers to the use
of mechanical energy to induce physio-chemical phenomena that activate and compound the raw
materials [62]. Mechanical energy was input to the raw materials via ball-milling Figure 36 their
blend over 2 hours with a steel ball-to-powder weight ratio of 8:1. The formulations of raw
materials that were investigated are introduced in Table 23.

64

Figure 36. Steel ball-milling jar and apparatus

Table 23. Alternative formulations (wt.%) of raw materials used for mechanochemical
processing of hydraulic cements, and the additives used during mixing
Raw Materials
Sample
Ash Slag Copper Tailing Na2SiO3 NaoH Na2CO3 CaO
B-11-A 45
20
20
7
5
3
B-12-A 45
20
20
7
5
3

Several different additives/supplementary materials were added in small dosages to various
hydraulic cements during preparation of mortar mixtures. Some of these materials were activated
prior to being added to the hydraulic cements. Kaolinite was calcined at 750°C for 12 hours. Waste
glass was crushed into fine particles and mechanochemically processed with sodium carbonate
with a glass: sodium carbonate weight ratio of 4:1 and with 2% moisture for 2 hours. It was
determined, through multiple trials, that the mixing of sand, powder and additives for 3 minutes in
a mortar mixer at medium speed was adequate for producing homogenous mixtures.

65

5.4. Experimental Evaluation of Hydraulic Cements

5.4.1. Alkalinity in Solution

The alkalinity levels of the hydraulic cements were measured by performing pH tests per ASTM
C25. In this test, 5g of cement was mixed with 45mL of deionized (DI) water, and shaken (Figure
37) for 30 minutes. After shaking, the solution was allowed to stand still for 30 minutes before
making measurements. The solution was then shaken for 24 hours, and measured again. The pH
level of the solution was measured together with the total dissolved solids (TDS) and electrical
conductivity. Figure 37 shows the apparatus by which these samples were stirred and Figure 38
shows the equipment utilized for the pH, TDS, and electrical conductivity measurements.

Figure 37. Shaking apparatus

66

Figure 38. pH, TDS and Conductivity apparatus

5.4.2. Thermogravimetric Analysis

Thermogravimetric analysis (TGA) records the mass loss of a sample with increasing temperature
at a constant rate. The results of a TGA test can help determine the physical attributes and chemical
composition of a material. The rates of mass loss at different temperatures can also be found as
derivatives of the mass loss test data versus temperature (DTA). TGA was executed using samples
of approximately 100mg with a Perkin Elmer TGA 4000 at a heating rate of 10˚C/min from 50 to
950oC under a nitrogen atmosphere.

5.4.3. Fourier Transform Infrared (FTIR) Spectroscopy

The Fourier transform infrared spectroscopy (FTIR) test was performed in order to detect the
formation of chemical bonds via mechanochemical processing of raw materials. To execute an
FTIR test, the cement powder was compressed into pellets (Figure 39). A pellet made of 0.125g
of potassium bromide was used to establish the baseline. A pellet composed of 0.125g of potassium
bromide and a small amount of the cement powder was then tested. The FTIR test was run by
utilizing the JASCO Spectophotometer (Figure 40) .The tests were performed using wavelengths
67

ranging from 400 cm-1 to 4000 cm-1, with the data recorded over 250 scans at a spectral resolution
of 4 cm-1.

Figure 39. Carver pellet press

Figure 40. JASCO FTIR Spectrophotometer

5.4.4. Compressive Strength

Mortar specimens were prepared per ASTM 1157C requirements. Immediately after mixing, the
fresh mix flowability was measured on a flow table (Figure 41) per ASTM C1437-07. The fresh
mix was then used to prepare 50mm cube specimens. The molded specimens were sealed and
stored for 24 hours at room temperature. After 24 hours, they were demolded, sealed and allowed
68

to cure at room temperature for 3, 7 and 28 days. The compressive strength values were measured
at these ages per ASTM 1157C.

Figure 41. Flow table

5.4.5. Measurements of Heavy Metals Concentrations in Leachated Using Inductively Coupled
Plasma (ICP) Spectroscopy

Sample powders were subjected to inductively coupled plasma (ICP) analysis in order to measure
the concentrations of different trace metals. This was necessary to determine whether the samples
had successfully immobilized the dangerous trace metals present in the raw materials prior to
milling. The samples were turned into solutions of approximately 1% - 2% HNO3 prior to the ICP
analysis. A blank sample was prepared by the same method used to prepare the unknown samples.
The blank sample was run to check the cleanliness of the sample preparation procedure. A series
of standard solutions with varying concentrations of multi-element standards were prepared in
order to determine the concentrations in unknown sample solutions.

69

5.5. Results and Discussion

5.5.1. Alkalinity in Solution

The results, presented in Table 24 indicate that, after shaking for 30 minutes, the milled cement
powder provided higher levels of pH, conductivity and total dissolved solids when compared with
a similar blend of raw materials that was not milled. This observation points at the rise in solubility
of raw materials upon milling. After 24 hours, the measurements revealed that both B-11-A
samples experienced significant reductions in total dissolved solids and electrical conductivity
while the B-12-A samples remained constant. The pH levels were shown to consistently exceed
13, which is ideal for alkali-activated cements.

Table 24. pH, total dissolved solids (TDS) and conductivity test results at 30 minutes and 24
hours
Material
30-minute results
24-hour results
pH
TDS
Conductivity pH
TDS
Conductivity
Copper Mine Tailing 9.06 58.12 ppm 118.3 μS/cm 9.08 57.63 ppm 117.8 μS/cm
B-11-A (blend)
13.48 24.04 ppt 49.04 mS/cm 13.54 14.33 ppt 29.23 mS/cm
B-11-A (milled)
13.60 24.52 ppt 50.04 mS/cm 13.60 21.01 ppt 42.88 mS/cm
B-12-A (blend)
13.33 14.01 ppt 28.59 mS/cm 13.46 13.88 ppt 28.21 mS/cm
B-12-A (milled)
13.37 14.49 ppt 29.57 mS/cm 13.43 14.12 ppt 28.85 mS/cm

5.5.2. Thermogravimetric Analysis (TGA)

The TGA test results are presented in Figure 42 . The DTA test data is presented in Figure 43.
Several peaks can be observed in Figure 43. The copper tailing exhibits peaks at approximately
550°C and 750°C, while B-11-A and B-12-A show peaks around 100°C, 400°C and 700°C. The
peaks are shifted between the copper tailing to the binder samples, indicating chemical changes
70

during the milling process. The peaks at 100°C indicate the presence of hydrates. The peaks at
400°C indicate chemically bound hydrates. The peaks observed at approximately 700°C indicate
the presence of carbonates. The formulation B-12-A shows substantially greater mass loss at this
temperature indicating that a larger amount of carbonates had formed during the milling process.

Figure 42. TGA data for hydraulic cement formulations and the copper mine tailing

71

Figure 43. DTA data for hydraulic cement formulations and the copper mine tailing

5.5.3. Fourier Transform Infrared (FTIR) Spectroscopy

The samples show below, in Figure 44, observe several different peaks. The samples B-11-A and
B-12-A observe larger peaks at approximately 3400 cm-1, 1400 cm-1 and 1050 cm-1. The peaks at
3400 cm-1 indicate the presence of water while the peaks at 1400 cm-1 are indicative of carbonates,
which both support the observed peaks on the DTA plot in Figure 43. The peaks at 1050 cm-1
indicate the presence of silica. The presence of silica is expected due to the large quantity of
aluminosilicate present in each formulation. The copper tailing sample observes this peak as well,
which is expected because it contains 44% silica oxide content as shown in Table 22.

72

Figure 44. Fourier transform infrared (FTIR) spectroscopy data

5.5.4. Compressive Strength

The initial effort relies upon superplasticizer to produce hydraulic cements that met requirements
relevant to fresh mix workability and set time. Superplasticizer was an effective additive in
Formulations B-11-A and B-12-A. The specifics of the composition of commercially available
superplasticizers, however, are protected as trade secrets; these specifics are also subject to change
over time. Therefore, one cannot rely upon them as generic raw materials for formulation of a new
class of hydraulic cement. The relatively high cost of commercially available superplasticizers is
another concern. Efforts were thus made to find replacements for superplasticizer in formulation
of hydraulic cements, considering that formulations used with superplasticizer offered 7-day
compressive strength of 21 to 23 MPa. The replacements considered here included borax, citric
acid, activated glass, and activated kaolinite. These replacements were considered due their
potential role in improving the fresh mix workability and set time. Activated glass and activated
kaolinite were particularly effective in matching these qualities offered by superplasticizer. Most
73

replacements were selected to lower the rate of reactions, enabling the fresh cement paste to retain
its workability over longer periods. The 7-day compressive strengths obtained with activated glass
and activated kaolinite were 21 MPa and 19 MPa, respectively (see Figure 45). The flow test data
presented in Figure 46 confirm that the replacements considered here for superplasticizer matched
its ability to improve fresh mix workability. Activated glass and activated kaolinite were used with
the hydraulic cement at additive: cement weight ratio of 10: 90. Their favorable economics makes
them less costly than superplasticizer, which was used at 0.5% by weight of cement. It should be
noted that borax and citric acid, which were also considered as replacement for superplasticizer,
were used at 1% by weight of cement.

Compressive Strength (Mpa)

25
20
15
B-11-A

10

B-12-A

5
0

S/p

Act. Glass

Act. Kaolinite

Borax

Citric Acid

Additive

Figure 45. 7 day compressive strength measurements of formulations with additives

The flow diameters shown in Figure 46 indicate the samples replacing the superplasticizer with
activated glass and activated kaolinite both produced a material with similar or superior flowability
compared to the superplasticizer sample. Borax and citric acid proved to be effective in increasing

74

the flowability of the samples (Figure 46), however they were found to cause significant reductions
in compressive strength (Figure 45).

20

Flow (cm)

16
12
8

B-11-A
B-12-A

4
0

S/p

Act. Glass

Act.
Kaolinite

Borax

Citric Acid

Additive

Figure 46. Flow diameter measurements of formulations with additives

5.5.5. Toxicity Characteristics per LEAF Procedure (ICP)

The degree of immobilization of heavy metals was evaluated using the “synthetic precipitation
leaching procedure (SPLP)” suggested by USEPA (1996). The immobilization efficiency was
calculated using the following equation [63]:

(2)

where, η(t) % is the immobilization efficiency (%) of the hydraulic cement treated mechanically
for a time interval of t(s), C(t) (mg L-1) is the heavy metal concentration of leachate, Vleach is the

75

leachate volume (L), q24h (mg kg-1) is the initial heavy metal concentration in untreated soil
(Equation (2)), and Wsolid (kg) is the solid weight used in the test. The test results are presented in
Figure 47, and the values of percent immobilization values calculated using the above equation are
presented in Figure 48. Percent immobilizations achieved with B-12-A formulation Figure 48.
These results point at the high degree of heavy metals immobilization provided by the binder
chemistry based on alkali aluminosilicate hydrates. The Cu, Ba, Ti and Mn constituents of the
copper mine tailing were 100% immobilized; the degree of immobilization of the Sr constituent of
mine tailing was more than 95%. These results suggest that abundant wastes which could be
hazardous could offer the potential for value-added use as aluminosilicate precursors in the
approach to production of alkali aluminosilicate cements developed in this work, yielding safe
cement hydrates and concrete materials which effectively immobilize the heavy metals
constituents of the hazardous aluminosilicate precursor.

1,000.0000

Concentration (ppm)

100.0000
10.0000
1.0000

Copper Tailing

0.1000

Paste

0.0100
0.0010
0.0001

Cu

Ba

Ti

Sr

Mn

Figure 47. Heavy metal concentrations in the leachate of the hydrated paste B-12-A formulation
at 7 days of age versus the initial concentrations in copper mine tailing

76

100
99

% Immobilization

98
97
96
95
94
93
92

Cu

Ba

Ti

Sr

Mn

Figure 48. Percent immobilizations achieved with B-12-A formulation

5.6. Conclusion

Due to its high silica and alumina contents, copper mine tailing can be an effective aluminosilicate
precursor in alkali-activated binders. The results of this investigation indicate that copper tailing
can be an effective aluminosilicate precursor at quantities as high as 20% in formulation of highrecycled-content alkali aluminosilicate hydraulic cements subjected to energy-efficient
mechanochemical processing. Copper mine tailing incorporates heavy metals at concentrations
which mitigate its use in many applications. An experimental investigation was conducted in order
to assess the potential for immobilization of the hazardous constituents of copper mine tailing by
the cement hydration products. Results of the ICP spectroscopy tests performed for this purpose
indicated that the heavy metals present in the copper mine tailing were substantially immobilized
by the hydrated cement paste at an early age of 7 days, considering that the hydration
(geopolymerization) process is not complete at this age. This finding suggests that abundant wastes
which are rich in aluminosilicates could feasibly be used as aluminosilicate precursors in the
77

cement production process developed here in spite of their relatively high heavy metals
concentrations which prevent them from use in other applications.
The use of copper mine tailing as an aluminosilicate precursor in production of cement led to
increased demand on water content for achieving desired fresh mix workability. Viable solutions
involving the use of generic additives such as activated kaolinite or waste glass were developed to
effectively resolve this problem.

78

BIBLIOGRAPHY

79

BIBLIOGRAPHY

1.
Kong, D.L. and J.G. Sanjayan, Damage behavior of geopolymer composites exposed to
elevated temperatures. Cement and Concrete Composites, 2008. 30(10): p. 986-991.
2.
Damtoft, J., et al., Sustainable development and climate change initiatives. Cement and
concrete research, 2008. 38(2): p. 115-127.
3.
Reig, L., et al., Properties and microstructure of alkali-activated red clay brick waste.
Construction and Building Materials, 2013. 43: p. 98-106.
4.
VILAS SANKAR, K., S. LATHESWARY, and K. VALSARAJAN, Effect of Partial
Replacement of Cement with Rice Husk Ash. Journal of the Institution of Engineers. India. Civil
Engineering Division, 2010. 90(FEV): p. 25-30.
5.
Hossain, M., et al., Performance of Concrete in presence of Supplementary Cementitious
Material.
6.
Matalkah, F., et al., Use of non-wood biomass combustion ash in development of alkaliactivated concrete. Construction and Building Materials, 2016. 121: p. 491-500.
7.
Matalkah, F., et al., Characterization of Alkali-Activated Nonwood Biomass Ash–Based
Geopolymer Concrete. Journal of Materials in Civil Engineering, 2016: p. 04016270.
8.
Kumar, S. and R. Kumar, Mechanical activation of fly ash: Effect on reaction, structure
and properties of resulting geopolymer. Ceramics International, 2011. 37(2): p. 533-541.
9.
Pereira, A., et al., Mechanical and durability properties of alkali-activated mortar based on
sugarcane bagasse ash and blast furnace slag. Ceramics International, 2015. 41(10): p. 1301213024.
10.
Redden, R. and N. Neithalath, Microstructure, strength, and moisture stability of alkali
activated glass powder-based binders. Cement and Concrete Composites, 2014. 45: p. 46-56.
11.
Reig Cerdá, L., et al., Properties and microstructure of alkali-activated red clay brick waste.
2013.
12.
Baronio, G. and L. Binda, Study of the pozzolanicity of some bricks and clays.
Construction and Building Materials, 1997. 11(1): p. 41-46.
13.
Sassoni, E., et al., Valorization of brick waste by alkali-activation: A study on the possible
use for masonry repointing. Ceramics International, 2016. 42(13): p. 14685-14694.
14.
Naceri, A. and M.C. Hamina, Use of waste brick as a partial replacement of cement in
mortar. Waste management, 2009. 29(8): p. 2378-2384.

80

15.
Khatib, J. and S. Wild, Pore size distribution of metakaolin paste. Cement and Concrete
Research, 1996. 26(10): p. 1545-1553.
16.
Gonçalves, J., et al., Performance evaluation of cement mortars modified with metakaolin
or ground brick. Construction and Building Materials, 2009. 23(5): p. 1971-1979.
17.
Sabir, B., S. Wild, and J. Bai, Metakaolin and calcined clays as pozzolans for concrete: a
review. Cement and Concrete Composites, 2001. 23(6): p. 441-454.
18.
Lin, K.-L., et al., Waste brick’s potential for use as a pozzolan in blended Portland cement.
Waste Management & Research, 2010. 28(7): p. 647-652.
19.
Kinuthia, J. and R. Nidzam, Towards zero industrial waste: Utilisation of brick dust waste
in sustainable construction. Waste management, 2011. 31(8): p. 1867-1878.
20.
Balaz, P., Mechanochemistry in nanoscience and minerals engineering. 2008: Springer
Science & Business Media.
21.
Teng, L.-W., et al., A Study of Crystalline Mechanism of Penetration Sealer Materials.
Materials, 2014. 7(1): p. 399.
22.
Huang, C. and P.F. Kerr, Infrared study of the carbonate minerals. Am. Mineral, 1960.
45(3-4): p. 311-324.
23.
Raichlin, Y., A. Millo, and A. Katzir, Investigations of the structure of water using mid-IR
fiberoptic evanescent wave spectroscopy. Physical review letters, 2004. 93(18): p. 185703.
24.
Cho, K.A., I. Choi, and I.W. Kim, Ball milling to build the hybrid mesocrystals of
ibuprofen and aragonite. Journal of Nanomaterials, 2015. 2015: p. 1.
25.
Matalkah, F., et al., Mechanochemical synthesis of one-part alkali aluminosilicate
hydraulic cement. Materials and Structures, 2017. 50(1): p. 97.
26.
Iwasaki, T., et al., Scale-up methodology for tumbling ball mill based on impact energy of
grinding balls using discrete element analysis. Advanced Powder Technology, 2010. 21(6): p. 623629.
27.
Avvakumov, E.G.e., M. Senna, and N. Kosova, Soft mechanochemical synthesis: a basis
for new chemical technologies. 2001: Springer Science & Business Media.
28.

Davidovits, J., Geopolymer chemistry and applications. 2008: Geopolymer Institute.

29.
Duxson, P., et al., The role of inorganic polymer technology in the development of ‘green
concrete’. Cement and Concrete Research, 2007. 37(12): p. 1590-1597.
30.
Juenger, M.C.G., et al., Advances in alternative cementitious binders. Cement and
Concrete Research, 2011. 41(12): p. 1232-1243.
31.

Provis, J.L., Alkali-activated materials. Cement and Concrete Research, 2017.
81

32.
Provis, J.L. and S.A. Bernal, Geopolymers and Related Alkali-Activated Materials. Annual
Review of Materials Research, 2014. 44(1): p. 299-327.
33.
Provis, J.L., A. Palomo, and C. Shi, Advances in understanding alkali-activated materials.
Cement and Concrete Research, 2015. 78: p. 110-125.
34.
McLellan, B.C., et al., Costs and carbon emissions for geopolymer pastes in comparison to
ordinary portland cement. Journal of Cleaner Production, 2011. 19(9): p. 1080-1090.
35.
Muszynski, L. Corrosion protection of reinforcing steel using pyrament blended cement
concrete. in BLENDED CEMENTS IN CONSTRUCTION. PAPERS PRESENTED AT THE
INTERNATIONAL CONFERENCE, UNIVERSITY OF SHEFFIELD, UK, 9-12 SEPTEMBER
1991. 1991.
36.
van Deventer, J.S., et al., Microstructure and durability of alkali-activated materials as key
parameters for standardization. Journal of Sustainable Cement-Based Materials, 2015. 4(2): p.
116-128.
37.
Craig Heidrich, J.S., Marita L. Berndt, Stephen Foster, Kwesi Sagoe-Crentsil, Pathways
and barriers for acceptance and usage of geopolymer concrete in mainstream construction. 2015.
38.
Glasby, T., et al. EFC geopolymer concrete aircraft pavements at Brisbane West Wellcamp
Airport. in Concrete Institute of Australia Conference, 27th. 2015.
39.
Provis, J.L., et al., Demonstration projects and applications in building and civil
infrastructure, in Alkali Activated Materials. 2014, Springer. p. 309-338.
40.
Van Deventer, J.S., J.L. Provis, and P. Duxson, Technical and commercial progress in the
adoption of geopolymer cement. Minerals Engineering, 2012. 29: p. 89-104.
41.
Buchwald, A., ASCEM® cement–a contribution towards conserving primary resources
and reducing the output of CO 2. Cem. Int, 2012. 10(5): p. 86-97.
42.
Husbands, T.B., P.G. Malone, and L.D. Wakeley, Construction Productivity Advancement
Research (CPAR) Program. Performance of Concretes Proportioned with Pyrament Blended
Cement. 1994, DTIC Document.
43.
Mcintosh, A., et al., Selection and characterisation of geological materials for use as
geopolymer precursors. Advances in Applied Ceramics, 2015. 114(7): p. 378-385.
44.
Ozyildirim, C., A FIELD INVESTIGATION OF CONCRETE PATCHES CONTAINING
PYRAMENT BLENDED CEMENT. FINAL REPORT. 1994.
45.
Adriano, S.D.D.C., Environmental impacts of metal ore mining and processing: A review.
Journal of Environmental Quality, 1997.
46.

Brininstool, M., Copper, in USGS 2017 Minerals Yearbook. 2017. p. 54-55.

82

47.
Gordon, R.B., Production residues in copper technological cycles. Resources,
Conservation, and recycling 2002. 36: p. 87-106.
48.
Kossoff, D., et al., Mine tailings dams: characteristics, failure, environmental impacts, and
remediation. Applied Geochemistry, 2014. 51: p. 229-245.
49.
Onuaguluchi, O., Copper tailings as a potential additive in concrete: consistency, strength
and toxic metal immobilization properties. Indian journal of engineering and material sciences,
2012. 19(2): p. 79-86.
50.
Gupta, R.C., Utilization of copper tailing in developing sustainable and durable concrete.
Journal of Materials in Civil Engineering ASCE, 2017. 29(5): p. 1-7.
51.
Zhang, L., Synthesis and Characterization of Fly Ash Modified Mine Tailings-Based
Geopolymers. Construction and Building Materials, 2011. 25(9): p. 3773.
52.
Yodsudjai, W., Application of Fly Ash- Based Geopolymer for Structural Member and
Repair Materials. Advances in Science and Technology, 2014. 92: p. 74-83.
53.
Najimi, M., Alkali-Activated Natural Pozzolan/Slag Binder for Sustainable Concrete, in
Civil and Environmental Engineering and Construction. 2016, University of Nevada, Las Vegas:
Ann Arbor, MI. p. 431.
54.
Phair, J., Green chemistry for sustainable cement production and use. Green chemistry: an
international journal and green chemistry resource: GC, 2006. 8(9): p. 763.
55.
Duxson, P., The role of inorganic polymer technology in the development of "green
concrete". Cement and concrete research, 2007. 37(12): p. 1590-1597.
56.
Chen, L., Preparation and Properties of Alkali Activated Metakaolin-Based Geopolymer.
Materials, 2016. 9(9): p. 767.
57.
Shi, C. and K. Zheng, A review on the use of waste glasses in the production of cement
and concrete. Resources, Conservation and Recycling, 2007. 52(2): p. 234-247.
58.
Kou, S. and C. Poon, Properties of self-compacting concrete prepared with recycled glass
aggregate. Cement and Concrete Composites, 2009. 31(2): p. 107-113.
59.
Islam, G.M.S., M.H. Rahman, and N. Kazi, Waste glass powder as partial replacement of
cement for sustainable concrete practice. International Journal of Sustainable Built Environment,
2016.
60.

Figg, J., Reaction between cement and artificial glass in concrete. 1981.

61.

Baxter, S.Z., C. Meyer, and W. Jin, Use of waste glass in concrete. 1998, Google Patents.

62.
Balaz, P., Mechanochemistry in nanoscience and minerals engineering 2008, Springer
Science & Business Media.
83

63.

Montinaro, S., et al., Immobilization of heavy metals in contaminated soils.

84