You are on page 1of 126

Experimental investigation on self-healing

performance of cementitious composite


incorporating fly ash and ground granulated
blast furnace slag

その他(別言語等) フライアッシュおよび高炉スラグ微粉末を混入した
のタイトル セメント系材料の自己修復性能に関する研究
著者 NA Seung-Hyun
学位名 博士(工学)
学位の種別 課程博士
報告番号 甲第345号
学位授与年月日 2013-09-26
URL http://hdl.handle.net/10258/2662
EXPERIMENTAL INVESTIGATION ON SELF-HEALING

PERFORMANCE OF CEMENTITIOUS COMPOSITE

INCORPORATING FLY ASH AND GROUND

GRANULATED BLAST FURNACE SLAG

フライアッシュおよび高炉スラグ微粉末を混入したセメント系材料

の自己修復性能に関する研究

Seung-Hyun Na

Doctor of Philosophy

Division of Architecture, Civil and Environmental Engineering


MURORAN INSTITUTE OF TECHNOLOGY
ABSTRACT

Due to the limitation of the protection and conservation of the environment and natural

resources and high finance, long-term sustainable use of concrete is of importance issue in the

concrete industry all over the world. However, the deterioration of concrete structure is inevitable

when the material is exposed to extreme weather condition. Freeze-thaw durability is of great

important under cold climates. When exposed to the freezing and thawing cyclic conditions, the

lower frost resistance of concrete structure could be occurred to the ice expansion, accompanying

damaged surface and micro-cracks, and connect to the deterioration of the concrete structure.

Therefore, to solve this problem, it is necessary the technique to fill the micro cracks to extend the

service life of the concrete structures.

The main objective of current research is to investigate the effect of self-healing on damaged

concrete incorporating fly ash and ground granulated blast furnace slag.

In chapter 2, the reaction rate and self-healing ability in fly ash blended cement mixtures that deal

with reaction of fly ash paste, optimization replacement ratio of fly ash, variation of pore structure

and deterioration degree in mixtures were investigated. The experimental results revealed that

incorporating fly ash in cement paste would affect the hydration rate of fly ash and consumption of

calcium hydroxide. Velocity of reaction rate and the optimization of fly ash replacement ratio to fill

micro cracks in the concrete were suggested.

In chapter 3, the self-healing ability of fly ash blended concrete with or without air entraing agent

after deterioration that caused by frost and thaw cycling test was examined. Moreover, new damage

technique to make micro crack within air entrained concrete sample was suggested by means of

compression test. It was found that although fly ash had self-healing potential, air entrained agent

addition in the fly ash concrete sample is advisable because of the poor frost resistance due to the

less air content in fly ash blended concrete sample. In addition, it was also confirmed that the

i
repeated cyclic loading technique is useful and efficient way in order to make the crack in the

concrete in comparison to rapid freeze and thaw test, and this method will be able to apply

AE-concrete which has a high frost resistance.

In chapter 4, the self-healing ability of ground granulated blast furnace slag blended mortar with

different replacement ratios and finenesses was explored. It was confirmed that low fineness, high

replacement ratio of ground granulated blast furnace slag for sand exhibited higher self healing

ability with respect to the compressive, bending strength and accelerated carbonation test.

In chapter 5, comparison of self-healing ability of concrete mixture incorporating fly ash and

ground granulated blast furnace slag blended was performed in order to design more durable

self-healing concrete. It was found that self-healing efficiency can be decreased by increasing the

curing age and fly ash blended concrete sample has a higher self-healing ability in comparison to

ground granulated blast furnace slag blended concrete. In addition, new durability factor by taking

into the consideration of the self-healing effect equation was suggested.

Keywords: Fly ash, Ground granulated blast furnace slag, hydration velocity, self-healing ability,

freezing and thawing durability, micro cracks, repeated cyclic loading technique, optimization

replacement ratio

ii
TABLE OF CONTENTS

ABSTRACT ................................................................................................................... ii
TABLE OF CONTENTS.............................................................................................iii
CHAPTER 1 GENERAL INTRODUCTION ............................................................... 1
1.1 Introduction........................................................................................................ 2
1.2 World trend of self-healing ability ..................................................................... 4
1.3 Self-healing ability of concrete structures ......................................................... 6
1.4 Fly ash and slag hydration reaction ................................................................ 16
1.5 Use of Fly ash and slag as self-healing agent ................................................. 19
1.6 Problem definition............................................................................................ 24
1.7 Research aims and thesis organization ........................................................... 24
Reference ................................................................................................................ 27
CHAPTER 2 EXPERIMENTAL INVESTIGATION ON REACTION RATE AND
SELF-HEALING ABILITY IN FLY ASH BLENDED CEMENT MIXTURES ........ 32
2.1 Overview........................................................................................................... 33
2.2 Experimental program ..................................................................................... 35
2.2.1 Reaction of fly ash and calcium hydroxide reagents ................................ 35
2.2.2 Acceleration test for fly ash cement paste ................................................ 36
2.2.3 Measurement of velocity of fly ash reaction rate ...................................... 37
2.2.4 Self-healing of fly ash ................................................................................ 38
2.3 Results and discussion ..................................................................................... 40
2.3.1 Reaction of fly ash and calcium hydroxide reagent .................................. 40
2.3.2 Fly ash cement paste acceleration reaction test at 80oC .......................... 42
2.3.3 The reaction rate of fly ash cement paste ................................................. 43
2.3.4 Optimization of fly ash replacement ratio ................................................ 47
2.4 Conclusions ...................................................................................................... 55
References .............................................................................................................. 58
CHAPTER 3 SELF-HEALING EFFECT AND FROST RESISTANCE OF FLY ASH
BLENDED CONCRETE ........................................................................................... 60
3.1 Overview........................................................................................................... 61
3.2 Experimental program ..................................................................................... 62
3.2.1 Experimental materials ............................................................................ 62
3.2.2 Experimental methods .............................................................................. 64
3.3 Result and discussion....................................................................................... 67

iii
3.3.1 Self-healing ability for Non-AE concrete .................................................. 67
3.3.2 Self-healing ability of fly ash blended air entrained concrete .................. 70
3.3.3 New damage technique and visualization of micro crack ........................ 75
3.4 Conclusions ...................................................................................................... 77
References .............................................................................................................. 77
CHAPTER 4 SELF-HEALING ABILITY OF GROUND GRANULATED BLAST
BURNACE SLAG BLDED SYSTEM AFTER FROST DAMAGE ............................ 79
4.1 Overview........................................................................................................... 80
4.2 Experimental program ..................................................................................... 81
4.2.1 Experimental materials ............................................................................ 81
4.2.2 Experimental methods .............................................................................. 82
4.3 Results and discussions ................................................................................... 85
4.3.1 Frost resistance ......................................................................................... 85
4.3.2 Compressive and bending strength .......................................................... 86
4.3.3 Self healing ability .................................................................................... 87
4.4 Conclusions ...................................................................................................... 92
References. ............................................................................................................. 93
CHAPTER 5 COMPARISON OF SELF-HEALING ABILITY OF CONCRETE
INCORPORATING FLY ASH AND GROUND GRALUATED BLAST FURNANCE
SLAG ......................................................................................................................... 94
5.1 Overview........................................................................................................... 95
5.1.1 Self-healing efficiency ................................................................................... 96
5.2 Experimental program ..................................................................................... 98
5.2.1 Experimental materials ............................................................................ 98
5.2.2 Experimental methods .............................................................................. 99
5.3 Results and discussion ................................................................................... 102
5.3.1 Frost resistance ....................................................................................... 102
5.3.2 Self-healing effect on the change of relative dynamic modulus of elasticity
.......................................................................................................................... 102
5.3.3 New durability factor considering self-healing effect ............................. 105
5.3.4 Self-healing ability comparison............................................................... 107
5.3.5 Self-healing effect of fly ash and slag blended composites ..................... 107
5.4 Conclusions .................................................................................................... 109
Reference .............................................................................................................. 110
CHAPTER 6 CONCLUSIONS AND FUTURE WORK .......................................... 111
6.1 Overview......................................................................................................... 112

iii
6.2 Conclusions .................................................................................................... 112
6.2.1 Experimental investigation of reaction rate and self-healing ability in fly
ash blended cement mixtures .......................................................................... 112
6.2.2 Self-healing effect and frost resistance of fly ash blended concrete ....... 114
6.2.3 Self-healing ability of ground granulated blast furnace slag (BFS) blended
system after frost damage ................................................................................ 114
6.2.4 Comparison of self-healing ability of concrete mixture incorporating fly
ash and ground graduated blast furnace slag ................................................. 115
6.3 Future work ................................................................................................... 116
References ............................................................................................................ 116

iii
CHAPTER 1 GENERAL INTRODUCTION
1.1 Introduction
Concrete, which consists of water, cement, fine and coarse aggregates, has been widely used for a

long time in the civil engineering and construction field all over the world, as many benefits such as

low cost, extremely strong compression and so on. However, although there are many advantages,

deterioration of concrete structure is inevitable and it is especially dependent on various

environmental factors, especially for temperature and humidity. Due to the deterioration, cracks

occur into the concrete, leading to the reduction of service life time of concrete structure. Further,

when cracks introduces, the cost and amount of labor required for diagnosis and repair work should

be needed in order to restore the original performance.

One of the most important issues of concrete is the durability that could be attributed with

extension of service life time. The durability could be affected by various properties such as

alkali-aggregate expansion reaction, freeze-thaw expansion, salt scaling by deicing salts, autogenous

and drying shrinkage, surface attack on exposure to ground waters containing surface ions, sea water

attack, and corrosion that caused by salts. Mineral admixtures as by-products, which are fly ash,

ground granulate blast furnace slag and silica hume, etc, have been widely used to improve the

mechanical properties of concrete.

Due to increasing the uses of admixtures into the concrete industry, the structure may depend on

environmental conditions that relate to the temperature and humidity. Particularly, among the

durability characteristic, freeze-thaw durability is of great important under cold climates. The

resistance involves spalling, scaling and cracking. When exposed to the freezing and thawing cyclic

conditions, the lower frost resistance of concrete structure could be occurred to the ice expansion

(spalling), accompanying damaged surface (scalling) and micro-cracks (cracking), and connect to

2
the deterioration of the concrete structure. In addition, due to the reduction of the frost durability as

freezing and thawing cycles, micro cracks within concrete structure occur and therefore carbonation

and chloride ingress is very rapid. Hence, for the extension of concrete structure, the technique of

cracks caused by frost damage should be needed and addressed.

Recently, researches on self-healing ability that is promising technique as intelligent materials has

been widely studied on the damaged concrete in laboratory and in real practical site, incorporating

micro capsule, fiber, expansive agent, fly ash and ground granulated blast furnace slag and so on.

Among them, fly ash and ground granulated blast furnace slag has been widely used in concrete

construction field to make C-S-H hydration products and improve the denser pore structure, leading

to long-term strength, low permeability and so on. However, the hydration of these materials takes

longer, this problem should be addressed. It is, however, assumed that the use of slow hydration in

fly ash and ground granulated blast furnace slag can heal and seal the micro cracks that caused by

low frost damage concrete structure.

3
1.2 World trend of self-healing ability
In 2005, RILEM Technical Committee 221 (Chairman Prof. E Schlangen) was established

regarding self-healing phenomena in cement-based materials. The purpose of the committee was to

develop specific way to make self-healing concrete incorporating various cementitious materials.

First international conference on self-healing materials held in the Netherlands in 2007.

To date, research on self-healing ability has been widely performed all over the world. The ability

is unique and promising solution to recover the damaged concrete that caused by various

deteriorations such as carbonation, autonomous and drying shrinkage and frost damage and so on.

Meanwhile, in 2007, JCI Technical Committee (JCI-TC075B 2007) was established and this

committee was to investigate the self-healing ability incorporating cementitious materials. The

committee was defined technical terms such as natural healing, autogenic healing and activated

repairing, as shown in Fig. 1.1.

Engineered healing
Autogenous
healing

Natural Autonomic Activated


healing healing repairing

Fig. 1.1 Definition of self-healing (JCI 2009)

4
For general mechanism of self-healing, the phenomenon has been observed in cementitious

materials for many years. One such example is on an 18th century bridge in Amsterdam, where

microcracks were self-healed by the re-crystallization of calcite (Nijland TG et al 2007). Afterwards,

Schlangen E. (2010) has been suggested that several possible causes can be responsible for the

self-healing phenomena, as shown in Fig. 1.2.

a. Formation of calcium carbonate or calcium hydroxide.

b. Blocking cracks by impurities in the water and loose concrete particles resulting from crack

spalling.

c. Further hydration of the unreacted cement or cementitious materials.

d. Expansion of the hydrated cementitious matrix in the crack flanks (swelling of C–S–H).

Fig. 1.2 Possible mechanisms for self-healing in cementitious materials (Schlangen E. 2010)

5
1.3 Self-healing ability of concrete structures
Extensive researches on the self-healing effect have been widely reported for different cement

based concrete structures and members (dam, bridge, offshore platform, asphalt etc.). In this section,

some examples of self-healing are presented according to the literature.

Glanville (1931) has reported for cracks of self-healing ability in bridges. Besides, for

underground civil infrastructure as tunnels, since underground structures as tunnels are generally

exposed to external water, high water-tightness reliability is needed to extend the service life of the

concrete structures. However, deterioration is inevitable so that the appearance of cracks in

underground structure would reduce the durability, which is reduced by water-tightness performance.

To solve this problem, self-healing of leaking cracks were studied extensively by Ahn and Kishi

(2012), who suggested a method for repairing cracks and have experimentally reported the crack

self-healing performance of cementitious composites using various mineral admixtures such as

expansion agent (CSA) and geo-materials for underground structures as tunnels, as shown in Fig. 13.

Incorporating the mineral admixtures, they were found that self-healing phenomenon is mainly

hydration behavior such as swelling effect, expansion effect and re-crystallization. Furthermore, they

were also investigated that cracks concrete could be completely healed after water immersion during

33 days, as given in Fig. 1.4

6
Fig. 1.3 Application concept of self-healing concrete for the water leakage of underground structure

as tunnels reported by Ahn and Kishi (2012)

7
Fig.1.4 Process of self-healing on self-healing concrete at water to binder ratio of 0.47 by Ahn and

Kishi (2012)

8
For self-healing of porous asphalt pavement, revelling, which is defined the loss of stones from

the road surface, is of importance to improve the lifetime of porous asphalt. In Netherlands, Erik

Schlangen and his co-researchers have reported self-healing ability of porous asphalt concrete

reinforced steel wool by measurements of four point bending fatigue resistance, as shown Fig. 1.5.

They were found that fatigue life extension ratio depends mainly on not only stain and but also

temperature. Further, it can be observed that, in the cases of different resting temperature and heating

temperature until 100ºC, temperature at 85 ºC is the optimal heating temperature to obtain best

healing effect, as given in Fig. 1.6.

Fig. 1.5 The schematic representation of induction healing in porous asphalt concrete reported by Q.

Liu et al (2012)

Fig. 1.6 Fatigue life extension ratio (at 300 microstrain) of the samples versus heating and resting

temperature reported by Q. Liu et al (2012)

9
Akira H. et al (2012) investigated the self-healing of longitudinal cracks in utility concrete pole

exposed to freeze and thaw environment for 34 years (Kano, Aisho Town, Echi County, Shiga

Prefecture in Japan) and reported the self-healing mechanism for evaluating the different analysis

methods such as powder X-ray diffraction, Scanning electron microscope and EPMA. Fig. 1.7

presents the utility concrete pole shape and section in which the pole diameter was 190 mm at the

top and 357 mm at the ground surface. From the practical observation, they were found that the

calcite remains to be precipitated at the surface of the pole (see Fig. 1.8) because Ca ion was

dissolved into the water, and then the water reacted with CO2 to form calcite, which also agrees with

results of power X-ray diffraction at the surface (Fig. 1.9).

Further, through observation of calcite precipitation by RGB method (see Fig. 1.10), they

observed the calcite in cracked part. It is interesting to note that no carbonation occurred near the

surface (healthy crack-free portion) and the crack widths about 200 um could not completely healed,

but 50 to 100 um cracks widths could be filled by precipitation of calcite,

Fig. 1.7 Shape and section of the utility concrete pole (section at the ground) reported by Akira H.

et al (2012)

10
Fig. 1.8 Observation of calcite precipitation precipitation in cracks inside the pole reported by

Akira H. et al (2012)

Fig. 1.9 Result of powder X-ray diffraction of calcite precipitation at the surface reported by Akira

H. et al (2012)

11
Fig. 1.10 Calcite precipitation observation by RGB method reported by Akira H. et al (2012)

Above observations, the self-healing performance has been investigated by previous researchers

by applying the civil infra structure. However, self-healing ability is limited under freeze and thaw

behavior and less information is available on rapid freezing and thawing test. Besides, with

decreasing the frost resistance, it is necessary to repair appropriately the cracks in order to extend the

service life of concrete structure. Hence, to restore the original performance of damage concrete,

self-healing ability is quite needed since it is a relatively cost-effective solution compared to

conventional repair methods. Next section, general mechanism of frost resistance is presented.

To date, frost resistance mechanism have been reported and studied by many researchers with

Power (1945), Kamada (1981), Fagerlund (1978), Katsura (1999) and Setzer (2004). Powers

suggested the hydraulic pressure theory. According to the theory, addition of air in concrete is

related to the protection of frost damage and this phenomenon was associated with the 9% expansion

of water upon freezing (Power 1945). Detail description, firstly, the external concrete is sealed due

to freezable water because the surface of concrete is freeze then internal concrete, after the freezable

water is accompanied to 9% expansion. Due to the expansion of freezable water, unfrozen water

penetrates pore in hardening paste. The moisture movement within the material pore structure may

be attributed to the expulsion of unfrozen water from the freezing sites and this mechanism has been

widely used in concrete science.

12
However, Kamada (1981) was pointed that this phenomenon did not consider the melting-point

depression in hardening cement composites having different pore structure (pore size). Based on the

hydraulic pressure theory by Powers, he suggested the melting-point depression theory and also

hypothesized that pore water volume is related to the inherent pore size. Kamada et al. (1996) have

investigated, by statistic approach and measurements with pore structure (mercury intrusion

porosimetery) and freeze and thaw cycling test (ASTM C 666 A), that pore volume in hardened

cement paste increased in the range of 17. 8 to 1000 nm (radii) until -18ºC freezes, implying that the

frost resistance may depend on the appearance of pores having high pore volume under 1000 nm

(radii). Since the higher pore volumes increase, the freeze and thaw durability can be decreased,

leading to the lower frost resistance.

Fagerlund (1977) proposed critical degree of saturation for assessing the freeze/thaw resistance of

concrete and emphasized that frost resistance of concrete with different saturation degree had

different tendency. For test method, this critical degree of saturation, SCR, is determined by a test in

which sealed specimens containing different amounts of water are subjected to a few

freeze/thaw-cycles. Other specimens are subjected to a test in which their ability to absorb water is

measured. This test yields a sort of potential degree of saturation which can be reached during very

moist conditions, which is called the capillary degree of saturation. The potential freeze/thaw

resistance, F, can be calculated as follow;

F = SCR-SCAP∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ (1.1)

where, F is the potential freeze/thaw resistance, SCR is the critical degree of saturation and SCAP is the

capillary degree of saturation.

13
Fig. 1.11 shows the result of frost durability parameter with normal concrete sample and

air-entrained concrete samples (Betong-handboken 1980). From this figure, it could be observed that

normal concrete with entrained air could be damaged by frost and thaw cycling, air-entrained

concrete could not be damaged even after longtime exposure to water absorption.

Fig. 1.11 Method of predicting the frost resistance of concrete

According to frost action mechanism by Powers and Kamada, Katsura et al. (1999) have been

suggested frost action mechanism considering super cooled water in concrete subjected to frost

environmental condition and reported that, when super cooled water freezes, growth rates of ice

crystals is significantly fast, the unfrozen water is rapidly moved due to volume expansion 9%. Thus,

pressure as movement is higher than tensile strength of concrete, leading to damage of pore structure.

However, they did not consider the movement of moisture and freeze and thaw cyclic behavior.

Max J. Setzer (2001) suggested micro–ice–lens model through thermodynamic approach based on

triple-point temperature (simultaneous and thermodynamically stable coexistence of vapor, liquid,

and solid water). The point temperature can shift in porous media. Fig. 1.12 shows the model and

14
detailed given that Part I shows the cooling and part II the heating phase of a freeze–thaw cycle. On

the left side (A) the macroscopic nonstationary system is given. Both temperature and pressure

profiles in the matrix with unfrozen water are plotted schematically. The moving boundary at the

macroscopic bulk-freezing level is seen. Due to the heat consumption there the temperature has a

nick point. The pressure in the unfrozen water starts below this temperature. On the right side (B) a

magnification by appr. 1 million is seen showing the triple-phase condition during cooling (I) and

heating (II). Here, fast local thermodynamic equilibrium is to be expected.

Fig. 1.12 Micro-ice-lens model (Max J. Setzer 2001)

15
1.4 Fly ash and slag hydration reaction
Various admixtures, which include fly ash, ground granulated blast furnace slag, silica fume and

so on, have been widely used in order to reduce CO2 emissions and to promote the mechanical

properties for a long time in concrete industry, especially dams, roads, nuclear power stations,

bridges and tunnels.

Among them, fly ash and ground granulate blast furnace slag are a promising solution, due to the

fact that the materials have good pozzolanic and latent hydraulic activities in comparison with

normal cement, thus, improved workability, long-term strength, reduced alkali silica reactivity,

lower porosity.

These two reactions are insignificantly different. The fly ash is reacted with calcium hydroxide

and water then produced C-S-H gel, which called pozzolanic reaction. The amount of cement

reduction will vary depending on the reactivity of the pozzolan.

In this section, therefore, fly ash and slag hydration reaction is presented, and the effect of cement

blended system on the hydration properties is reviewed from the literature. It is well known that

hydration rate of fly ash is lower than that of BFS, and fly ash addition may be attributed to lower

the compression strength at early ages. Kwangryul Hwang H. et al (2004) explored the prediction

model of compressive strength development of fly-ash concrete by taking into consideration of fly

ash replacement ratio, method and Blaine specific surface area. They were found that compressive

strength could be increased by the fly ash addition, which was replaced by fin aggregate (Fig. 1.13).

L. Lam et al (2000) examined the degree of hydration of high-volume fly ash/cement (FC)

systems by determining the non-evaporable water content using a selective dissolution method,

thermal gravimetry analysis and loss on ignition method. The results of hydration analysis indicated

16
that the decrease in Ca(OH)2 with curing ages indicates the progress of pozzolanic reaction that

consumed Ca(OH)2. In addition to the observation, preparing high-volume fly ash concrete at lower

w/b ratios can result in low compressive strength.

Fig. 1.13 Compressive strength development over time with age (“IN” means fly ash replacement for

cement, “EX” means for fine aggregate)

Kobayakawa M. (2001) examined the change of pore structure for fly ash blended paste by

evaluating the mercury intrusion porosimetery, ignition loss, BET (specific surface area) and EPMA

method. They reported that with increasing the curing ages (7, 28, 91, 182, 365 and 1095 days) and

with different fly ash replacement ratios (0.1 – 0.6 by mass of cement) would decrease the pore

volume in the range of 10 – 50 nm, and then increase the pore volume in the range of 3 to 10 nm in

the fly ash blended samples. They also found that pore volume of fly ash blended paste could be

changed by fly ash hydration rate (see Fig. 1.14)

E. Sakai et al (2005) investigated the influence of glass content and the basicity of the glass phase

of fly ash on the hydration of fly ash cement, as given in Fig. 1.15. They reported that glass content

affects the reaction of the fly ash during a period from 28 to 270 days and also found that reaction

17
ratio of fly ash at 360 days was almost the same for each fly ash replacement ratio. It was confirmed

that the pozzolanic reaction can occur between curing age of 91 and 360 days, while consuming

Ca(OH)2, because of the amount of Ca(OH)2 consumed by pozzolanic reaction.

Pore diameter
Curing
temperature
(40℃)
Pore volume (ml/ml)

Curing age (days)

Replcement ratio

Fig. 1.14 change of pore volume for fly ash blended system with different fly ash replacement ratios

Fig.1.15 Reaction ratio of fly ash (glass contents of 38.2% (F) and 76.6% (F`)

18
Many researchers have studied the effect of BFS on the hydration process and indicated that the

hydration degree of BFS is lower than that of cement (K.L Lin et al 2004, Tetsuya et al 2011, Yao

Luan et al 2012). This is an important issue in the concrete industry. Further, the BFS hydration

process depends on the presence of Ca(OH)2 as activator. H.-J. Chen et al (2012) investigated the

effect of curing conditions on the strength, porosity, and chloride ingress characteristics of concretes

made with high slag blast furnace cement (HBFC). They reported that The HBFC, which contains 45%

of blast furnace, had lower total porosity than the normal concrete with the same design strength due

to the fact that incorporation of HBFC in the concrete mixtures resulted in the formation of denser

microstructure within the cement paste matrix.

1.5 Use of Fly ash and slag as self-healing agent


S. Qian et al (2009) reported the self-healing behavior of pre-cracked fiber reinforced strain

hardening cementtious composites with different local waste materials such as BFS and limestone

power. They also confirmed that low water/cementitious material ratio and high percentage of

cementitious material can promote self-healing behavior after cracking and it was found that water

curing can increase the self-healing process and therefore enhance the fiber bridging behavior after

pre-cracking while it is not the case for air cured pre-cracked samples. Fig. 1.16 shows the crack

healing after water curing using environmental scanning electron microscopy. While most of the

cracks (crack width ranges from 10 to 60 micrometers) fully healed, a few crack with relatively large

width (60 micrometers) did show partial healing, as evident in Fig. 1.16. Particularly, in case of BFS

concrete, low water/cementitious material ratio and high percentage of cementitious material appear

to promote self-healing behavior.

19
Fig. 1.16 Partial crack healing after water curing (S. Qian et al. 2009)

For fly ash self-healing, Termkhajornkit P. et al. (2009) have investigated the self-healing

possibility of fly ash–cement systems. They were studied about self-healing ability for autogenous

and drying shrinkage after 28days, in terms of compressive strength, chloride diffusion, the cracks

and/or the pore structure, the hydration of cement and fly ash and hydrated products such as C–S–H

gel and Ca(OH)2. Experimental investigation revealed that the fly ash–cement system has the

self-healing ability for cracks due to shrinkage and the self-healing ability increased when the fly ash

replacement ratio increases. In addition, he pointed out that when cement was replaced by fly ash,

the compressive strength decreased. For the solution of decrease on compressive strength, in real

concrete, this problem may be overcome if fine aggregate instead of cement is replaced by fly ash.

Above observations, fly ash and BFS has a self-healing ability due to the pozzolanic and latent

hydraulic reaction and incorporation of two types of materials can fill artificial cracks which were

introduced by three and four point bending test. In addition, different types of self-healing agents

have been developed in the world. Therefore, other self-healing materials are reviewed from the

literature as follow;

20
Many researchers have been studied the application of microencapsulation approach incorporating

healing agent to self-healing of materials (Schleicher L et al 1956; White SR et al. 2001; Ghosh SK

2006). When the cracks occur, the embedded microcapsules rupture. Then the healing agent contacts

the embedded catalyst, triggering polymerization and ensuring the closure of the near-by cracks (M.

Wu et al 2012). Fig. 1.17 shows the microencapsulation self-healing concept (White SR et al. 2001).

This concept can be expressed as follow; i). Cracks form in the matrix; ii) the crack ruptures the

microcapsules, releasing the healing agent into the crack plane through capillary action; and iii) the

healing agent contacts the catalyst, triggering polymerization thus ensuring the closure of the near-by

cracks

Fig. 1.17 Basic method of the microcapsule approach (White SR et al. 2001)

In the case of bacteria used, Van Tittelboom and coworkers (2009) studied the self healing

potential of cement-based materials by using bacteria (Bacillus sphaericus) with respect to water

permeability tests, ultrasound transmission measurements and visual examination. This Bacillus

sphaericus is able to precipitate CaCO3 in their micro-environment by conversion of urea into

ammonium and carbonate. They reported that pure bacteria cultures were not able to bridge the

cracks.

21
However, when bacteria were protected in silica gel, cracks were filled completely. In addition,

where crack filling is provided by the gel matrix together with the precipitated CaCO3 crystals can

be seen as a durable repair technique.

Virginie W and Henk M. Jonkers (2011) studied the effect of novel bacteria-based concrete on

self healing used bacterial isolate obtained from alkaline lake soil and they also reported that while

in this study the enhanced self-healing capacity of bacteria-based concrete has been quantified,

several other characteristics such as long-term (years) durability and cost efficiency of this novel

type of concrete needs to be resolved before practical application can be considered.

Kishi T, Ahn TH and researchers (2007, 2008 and 2010) have studied the effect of expansive

agent on self-healing behavior for the water leakage of underground civil infrastructures as tunnels.

They also reported about self-healing agent including expansive agent, geo-polymer material and

chemical regents and re-hydration products between cracks were observed, showing that, for

concrete beams incorporating expansive agent, a crack with an initial width of 0.22 mm was almost

healed after 1 month compared to normal concrete beam without expansive agent.

More recently, K. Sisomphon et al. (2012) have reported that Self-healing of surface cracks in

mortars incorporating calcium sulfoaluminate based expansive additive expansive additive and

crystalline additive and revealed that the mixtures with expansive agent showed favorable surface

crack closing ability.

University of Michigan by civil engineering professor Victor Li, who has developed the

Engineered Cementitious Composite (ECC) that is a unique type of high performance fiber

22
reinforced cementitious composite and that has high tensile ductility (tensile strain capacity) with

moderate fiber content, typically 2% by volume (Y. Yang et al. 2009). This material has been widely

used to recover the original performance after deteioration.

For advantage and disadvantage of different strategies as various self-healing agents, M. Wu et al

(2012) reported an advantage and disadvantage in self-healing ability of cement based mixtures

incorporating various self-healing agents and it summarized in Table 1.1. It can be seen from the

table that encapsulation, internal encapsulation, microcapsule and shape memory materials have the

self-healing performance, and appear to be having the problems of casting and high cost (const

intensive). In addition, it is pointed out that before applying practical sites, the choice of the agent

for suitable structure is needed to consider.

Table 1.1 A summarized comparison between different strategies (M. Wu et al. 2012)

23
1.6 Problem definition
Based on the above literatures, it is clear that fly ash and BFS has a self-healing ability to fill and

seal the cracks that were introduced by three and four point bending test. It can be said that this

cracks are different micro cracks of frost damage and experimental investigation on the effect of

self-healing on damaged concrete incorporating fly ash and ground granulated blast furnace slag is

limited. In addition, from micro scale (hydration of each hydrated products) until large scale

(concrete structure), a systematic and extensive investigation of the effect of the mineral admixtures

on self-healing composites is quite required. Some of the main problems of investigation in the

literature may be described as follows;

 The effect of hydration characteristic including velocity of reaction, replacement ratio, method

with deterioration degree in fly ash blended mixtures has not been studied. In addition,

optimization of fly ash replacement ratio to fill micro cracks by taking into consideration of

hydration properties in the long term period is limited.

 The effect of blast furnace slag replacement ratio, method and finenesses on self-healing effect

performance of BFS blended mortar has not been fully understood yet.

 Since AE-concrete exhibits excellent frost resistance property, a new damage technique is

needed to easily and quickly apply the concrete sample.

1.7 Research aims and thesis organization


The objective of current research is to investigate the effect of self-healing on damaged concrete

incorporating fly ash and ground granulated blast furnace slag. Various influential factors (fly ash

and BFS replacement ratio, damage degree, a new damage technique, evaluation index of concrete,

monitoring the self-healing concrete and so on) is considered to design the more durable and higher

the self-healing concrete. The scope of the thesis is summarized as follow.

24
This thesis is organized into seven chapters as presented in Fig. 1.18. The organization of the thesis

and brief introduction of each chapter are drawn;

Chapter 1 gives general introduction providing the necessity of the current research and presents a

review of the literature on self-healing materials and strategies;

Chapter 2 investigates experimentally the reaction rate and self-healing ability in fly ash blended

cement mixtures that deal with reaction of fly ash paste, optimization replacement ratio of fly ash,

variation of pore structure and deterioration degree in mixtures;

Chapter 3 investigates the self-healing ability of fly ash blended concrete with and with air entrain

agent after deterioration that is caused by frost and thaw cycling test and develops a new damage

method in order to introduce artificial micro cracks into the air-entrained concrete

Chapter 4 provides the self-healing ability of ground granulated blast furnace slag blended mortar

with different replacement ratios and finenesses;

Chapter 5 investigates the self-healing ability of air-entrained concrete incorporating fly ash and

granulated blast furnace slag based on developed deterioration technique and proposes the durability

factor considering self-healing ability; and

Chapter 6 gives the general summary and conclusions drawn from the results of this research,

furthermore, gives recommendations to the future studies.

25
Chapter 1 General introduction
Literature review, Problem definition, Research aims and Thesis organization

Chapter 2 Experimental investigation on Chapter 4 Self-healing ability of ground


reaction rate and self-healing ability in fly granulated blast furnace slag blended
ash blended cement mixtures system after frost damage
‧ Reaction of fly ash and calcium hydroxide ‧The effect of fineness, replacement ratio and method

‧ Prediction equation of reaction rate on self-healing ability of ground granulated blast

‧ Reaction of fly ash furnace slag blended mortar

‧ Self-healing effect investigation

‧ Optimization of fly ash replacement ratio

Chapter 3 Self-healing effect and frost


resistance of fly ash blended concrete
‧Verification of self-healing ability for fly ash
blended concrete

‧Evauation of frost resistance and compressive


strength development

‧Self-healing ability under accelerated enviro


nmental conditions

‧Proposal of a new introduction method of artificial


micro cracks in terms of compression test

Chapter 5 Comparison of self-healing ability of concrete mixture incorporating fly ash and
ground granulated blast furnace slag
‧Frost resistance, self-healing effect investigation ‧Development fly ash and slag blended self-healing concrete

‧Proposal of new durability factor by taking into consideration of self-healing effect

Chapter 6 Conclusions
Overview, summary and future work

Fig. 1.18 Outline of the dissertation

26
Reference

A.E. Richardson, K.A. Coventry, S. Wilkinson, Freeze/thaw durability of concrete with synthetic
fibre additions, Cold Regions Science and Technology, pp. 83-84 (2012)

Ahn TH and Kishi T, The effect of geomaterials on the autogenous healing behavior of cracked
concrete, In : Proceeding of 2nd ICCRRR, Cape Town, South Africa. pp. 235–40 (2008)

Ahn TH and Kishi T., Crack self-healing behavior of cementitious composites incorporating various
mineral admixtures, J Adv Concr Technol , Vol.8, No.2, pp.171–86 (2010)

Ahn TH, Development of self-healing concrete incorporating geo-materials: a study on its


mechanism and behavior in cracked concrete. PhD dissertation, Department of Civil Engineering,
The University of Tokyo, Japan (2008)

Betong-handboken, Svensk Byggjanst, Stockholm, pp.430-433 (1980)

Brown PW, Bothe JV., The stability of ettringite. Adv Cem Res; 5(18):47–63 (1993)

Collepardi M., Damage by delayed ettringite formation – a holistic approach and new hypothesis.
Concr Int; 21(1), pp. 69–74 (1999)

De Muynck W, De Belie N, Verstraete W., Microbial carbonate precipitation improves the durability
of cementitious materials: a review. Ecol Eng;36, pp. 118–36 (2010)

Edwin R. Dunstan Jr., P.E, 2011, How does pozzolanic reaction make concrete “green”?, 2011 World
of coal ash conference, 9-12 May, in Denver, CO. USA (2011)

Etsuo Sakaia, *, Shigeyoshi Miyaharab , Shigenari Ohsawa, Seung-Heun Lee, Masaki Daimona,
Hydration of fly ash cement, Cement and Concrete Research 35, pp. 1135 – 1140 (2005)

Fagerlund, G, The critical degree of saturation method of assessing the freeze-thaw resistance of
concrete. Materials and Structures; 10 58, pp. 217 – 229 (1977)

Ghosh SK, Functional coatings and microencapsulation: a general perspective. Functional Coatings
(2006)

Heinz D, Ludwig U., Mechanism of secondary ettringite formation in mortars and concretes
subjected to heat treatment. In: Seanlon JM, editor. Concr durability. Katharine and Bryant Mather
Int Conf, SP 100, vol. 2. Detroit: American Concrete Institute. pp. 2059–71 (1987)

27
How-Ji Chen, Shao-Siang Huang, Chao-Wei Tang, M.A. Malek, Lee-Woen Ean, Effect of curing
environments on strength, porosity and chloride ingress resistance of blast furnace slag cement
concretes: A construction site study, Construction and Building Materials 35, pp. 1063–1070 (2012)

Jacobsen S, Marchand J, Homain H., SEM observations of the microstructure of frost deteriorated
and self-healed concrete. J Cem Concr Res; 25:1781–90 (1995)

Japan Concrete Institute, Autogenous-healing in cementtious materials, Technical Committee Report


of JCI-TC075 B, 7, (2012) (in Japanese)

Jonkers HM, Schlangen E, A two component bacteria-based self-healing concrete. In: Alexander MG,
Beushausen HD, Dehn F, Moyo P, editors. Concrete repair, rehabilitation and retrofitting II. Boca
Raton: CRC Press, Taylor and Francis Group. pp. 119–20 (2009)

Jonkers HM, Thijssen A, Muyzer G, Copuroglu O, Schlangen E, Application of bacteria as


self-healing agent for the development of sustainable concrete. Ecol Eng; 36, pp. 230–5 (2010)

Jonkers HM., Self-healing concrete: a biological approach. In: van der Zwaag S, editor. Self-healing
materials: an alternative approach to 20 centuries of materials science. The Netherlands: Springer; pp.
195–204 (2007)

K.L Lin, K.S Wang, C.Y Lin, C.H Lin, The hydration properties of pastes containing municipal solid
waste incinerator fly ash slag, Journal of Hazardous Materials, Volume 109, Issues 1–3, pp. 173-181
(2004)

Kim Van Tittelboom, Nele De Belie,Willem De Muynck and Willy Verstraete, Use of bacteria to
repair cracks in concrete, Cement and Concrete Research 40 pp. 157–166 (2010)

Kishi T, Ahn TH, Hosoda A, Takaoka H., Self-healing behavior by cementitious recrystallization of
cracked concrete incorporating expansive agent. In: Proceedings of the first international conference
on self-healing materials, Noordwijk aan zee, the Netherlands; April 18–20 (2007)

Kwangryul H., Takafumi N. and Fuminiro T., Prediction model of compressive strength
development of fly-ash concrete, Cement and Concrete Research, Volume 34, Issue 12, pp.
2269-2276 (2004)

Li VC., On engineered cementitious composites (ECC) – a review of the material and its
applications. J. Adv. Concrete Technol.1 (3):215–30 (2003)

Max J. Setzer, Mechanical Stability Criterion, Triple-Phase Condition, and Pressure Differences of

28
Matter Condensed in a Porous Matrix, Journal of Colloid and Interface Science 235, pp. 170–182
(2001)

Max J. Setzer, Micro-Ice-Lens Formation in Porous Solid, Journal of Colloid and Interface Science
243, pp193–201 (2001)

Mehta PK., Sulfate attack on concrete – a critical review. materials science of concrete III. The
American Ceramic Society; pp. 105–30 (1993)

Min Wu, Björn Johannesson and Mette Geiker, A review: Self-healing in cementitious materials and
engineered cementitious composite as a self-healing material, construction and Building Materials
28 pp. 571–583 (2012)

Mustafa Sahmarana, Victor C. Li b, Durability of mechanically loaded engineered cementitious


composites under highly alkaline environments, Cement & Concrete Composites 30 pp. 72–81
(2008)

Nijland TG, Larbi JA, van Hees RPJ, Lubelli B, de Rooij MR, Self-healing phenomena in concretes
and masonry mortars: a microscopic study. In: van der Zwaag S, editor. Proceedings (CD) first
international conference on self-healing materials. Dordrecht: Springer (2007)

P. Kumar mehta and Pauld J. M. Monteiro, Concrete (Microstructure, Properties, and Materials,
Third edition), McGraw-Hill

Pimienta P and Chanvillard G, Retention of the mechanical performances of Ductal specimens kept
in various aggressive environments. In: Fib symposium (2004)

Pipat T., Toyoharu N., Yoichi Y. and Toshiki S., Self-healing ability of fly ash–cement systems,
Cement & Concrete Composites 31 pp. 195–203 (2009)

Powers, T. C., A Working Hypothesis for Further Studies of Frost Resistance of Concrete, Journal of
American Concrete Institute, Vol.16, No.4, pp.245-272 (1945)

Ramachandran SK, Ramakrishnan V, Bang SS., Remediation of concrete using micro-organisms.


ACI Mater J 98:3–9 (2001)

Richardson, A.E., Polypropylene fibres in concrete with regard to durability. Structural Survey, vol.
21, no. 2. MCB UP Ltd., UK, pp. 87–94 (2003)

Richardson, A.E., Compressive strength of concrete with polypropylene fibre additions. Structural

29
Survey, vol. 24, no. 2. MCB UP Ltd., UK, pp. 138–153 (2006)

Richardson, A.E., Coventry, K., Early life freeze/thaw durability of Type 1 polypropylene fibre and
ground granulated blast furnace slag concretes. Coventry University and UWM Center for
By-Products Utilization, Second International Conference on Sustainable Construction Materials and
Technologies, June 28 - June 30. Università Politecnica delle Marche, Ancona, Italy (2010)

S. Qian, J. Zhou, M.R. de Rooij, E. Schlangen, G. Ye and K. van Breugel, Self-healing behavior of
strain hardening cementitious composites incorporating local waste materials, Cement and Concrete
Composites, Volume 31, Issue 9, October, pp. 613-621 (2009)

Schlangen E., Fracture mechanics. CT5146 Lecture Notes. In: Hua X. Self-healing of Engineered
Cementitious Composites (ECC) in concrete repair system. Master thesis, Delft University of
Technology (2010)

Schleicher L, Green BK. Manifold record material. US patent 2730456 (1956)

Scrivener KL, Damidot D, Famy C., Possible mechanisms of expansion of concrete exposed to
elevated temperatures during curing (also known as DEF) and implications for avoidance of field
problems. Cem Concr Aggr 21, pp. 93–101 (1999)

Sisomphon K, Copuroglu O., Some characteristics of a self-healing mortar incorporating calcium


sulfo-aluminate based agent. In: Proceedings of the 2nd international conference on durability of
concrete structures, Sapporo, Japan; pp. 157–64 (2010)

Taylor HFW, Cement chemistry. second ed. London: Thomas Telford Publishing (1997)

Tetsuya ISHIDA, Yao LUAN, Takahiro SAGAWA and Toyoharu NAWA: Modeling of early age
behavior of blast furnace concrete based on micro-physical properties, Cement and Concrete
Research , 41, pp. 1357-1367 (2011)

Virginie Wiktor Henk M. Jonkers, Quantification of crack-healing in novel bacteria-based


self-healing concrete, Cement & Concrete Composites 33, pp. 763–770 (2011)

White SR, Sottos NR, Geubelle PH, Moore JS, Kessler MR, Sriram SR, Autonomic healing of
polymer composites. Nature; 409(6822):794 (2001)

Yao Luan, Tetsuya Ishida, Toyoharu Nawa and Takahiro Sagawa: Enhanced Model and Simulation
of Hydration Process of Blast Furnace Slag in Blended Cement, Journal of Advanced Concrete
Technology, 10(1) pp. 1-13 (2012)

30
Yoshimichi Aono, Fumiaki Matsushita, Sumio Shibata and Yukio Hama, Nano-structural Changes of
C-S-H in Hardened Cement Paste during Drying at 50°C. Journal of Advanced Concrete Technology.
Vol., 5, No. 3, pp. 313-323 (2010)

31
CHAPTER 2 EXPERIMENTAL INVESTIGATION ON REACTION

RATE AND SELF-HEALING ABILITY IN FLY ASH BLENDED

CEMENT MIXTURES

32
2.1 Overview
Durability of concrete would be adversely degraded by micro cracks due to freezing/thawing cycles and

drying/wetting cycles. Hence, it is necessary to repair appropriately the cracks in order to extend the service life of

concrete structures. On the other hand, it has been well known that concrete has originally self-healing function due

to rehydration of cementitious materials such as anhydrous cement in the microstructure. Self-healing of concrete is

a phenomenon that micro cracks in a concrete structure naturally close under an environment with water supply. To

restore the original performance of damaged concrete, self-healing ability is quite needed since it is a relatively

cost-effective solution compared to conventional repair methods. Fly ash, which has been increasingly used as a

concrete admixture in last decades, has such kind of appropriate self-healing ability in the presence of calcium

hydroxide. The filling of micro cracks, with hydrated products of fly ash and cement, can decrease the transfer of

substances into damaged concrete, and hence, longer life span of concrete structures can be achieved appropriately.

In 2009, the Japan Concrete Institute committee has reported the definition of self-healing such as autogenous

healing, natural healing, autonomic healing, engineered healing and activated repairing (JCI 2009). It was

concluded that in the case of low water to cement ratio and high fly ash replacement ratio, the additional hydration

would be caused easily by water supply around the micro cracks because of the pozzolanic reaction of fly ash. This

pozzolanic reaction produces C-S-H gel that may heal micro cracks and result in better performance of cement

mixtures. However, it is not sufficient to understand the self-healing mechanism in long-term period, to explain the

hydration process of fly ash blended cement mixtures.

There are many studies in the literatures that have dealt with hydration property of cement based on X-ray

diffraction (XRD)-Reitveld analysis. They showed that XRD-Reitveld analysis is able to investigate the hydration

characteristics of cement, considering the quantitative technique, curing condition and particle size distribution in

cement and amorphous admixtures (Yamaguchi et al. 1960; Asaga et al. 1992; Itoh et al. 2002; Whitfield and

Mitchell 2003; Sagawa et al. 2004; Scrivener et al. 2004; Hoshino et al. 2005). Additionally, the reaction model of

cement paste has been developed by Sagawa et al (2005); this model considers the thickness of hydrated products.

33
Sakei et al. (1997, 2004) performed a fundamental study on the reaction between fly ash and calcium hydroxide

to measure the calcium hydroxide and hydration water using thermogravimetry and differential thermal analysis

(TG-DTA). To calculate the reaction rate and unhydrated products of fly ash, the selective dissolution method has

been reported (Ohsawa et al. 1999; Matsui and Ikabata 1999; Yamamoto and Kanazu 2004; Termkhajornkit et al.

2005).

Based on the above literatures, there is lack of information about long-term behavior of hydration process,

hydrated products and self-healing ability in fly ash blended cement mixtures for different types of fly ash and

cement.

The objective of this study is to investigate the hydration process of fly ash and consumption of calcium

hydroxide and self-healing performance in fly ash cement mixtures for long-term examination period, taking into

consideration effect of cement types, fly ash types, fly ash replacement ratio and curing conditions. The observation

of hydration rate of fly ash and calcium hydroxide was conducted by the combination of selective dissolution and

XRD-Rietveld analysis. Thermogravimetry and differential thermal analysis (TG-DTA) was applied to quantify

amount of calcium hydroxide in hardened paste. In order to introduce the micro cracks in sample, the freeze/thaw

cycling test was applied to pose damage in mortar incorporating fly ash, and then, self-healing effect on accelerated

carbonation tests, compressive strength, bending strength, relative dynamic modulus (RDM) of elasticity, point

count test and modification of pore volume were evaluated. The porosity and pore structure modification were

measured by mercury intrusion porosimetry (MIP). Finally, the obtained results from this study are valuable

information for practical design codes in which the understanding of self-healing ability of fly ash is quite needed

to prolong the service life of concrete structures.

34
2.2 Experimental program
The experimental work in this study consists of three parts. The first part deals with reaction rate of fly ash and

consumption of calcium hydroxide. The second part concerns with reaction rate of fly ash cement paste based on

acceleration test at 80℃, and the velocity of fly ash reaction rate. While the last part of this study concerns with

self-healing effect in mortars incorporating fly ash, which exposed damage by freezing/thawing cycles, based on

mechanical tests such as compressive and bending strengths, relative dynamic modulus of elasticity, accelerated

carbonation test, point count test and porosity. Details of the experimental plan and investigated fly ash blended

cement samples are given below.

2.2.1 Reaction of fly ash and calcium hydroxide reagents


The physical properties and chemical composition of ordinary Portland cement (OPC) and fly ash are shown in

Tables 2.1 and 2.2; mineral composition and glass phase for all investigated types of fly ash are shown in Table 2.3.

The experimental plan for fly ash samples 4A and 2B is shown in Table 2.4. The materials were mixed 2 minutes

and then casted in cubic mould with dimension of 50×50×50 mm. After casting, 4A and 2B samples were cured in

water at 60℃ and 40℃, respectively, for 1 day. After 1 day, they were demoulded and then moved into water till

the planned ages at 60℃ and 40℃ for 4A and 2B samples, respectively. At the required ages, samples were soaked

in acetone to stop the hydration, dried at 105℃, and then ground by ball mill. The amount of unhydrated fly ash

was estimated based on the selective dissolution method (Ohsawa et al. 1999; Matsui and Ikabata 1999; Yamamoto

and Kanazu 2004; Termkhajornkit et al. 2005). The rate of hydration of cement minerals was calculated by the

results obtained by the XRD-Rietveld analysis. The XRD-Rietveld analysis was performed after adding 10% of

Al2O3, as internal reference, to the sample. The sample and internal reference were blended uniformly till becoming

a harmonized texture. Cu Kα X-ray diffraction system was used for conducting XRD-Rietveld analysis, with 45 kV

tube electric voltage, 40mA tube current, scanning rage of 2θ 20-70° and step scan width of 0.02°. The Rietveld

analysis program used in this study was Rietan-2000 software developed by Izumi and Ikeda (2000). Target

minerals for analysis were C3S, C2S, C3A, C4AF, Mullite, Quartz and Magnetite. In addition, the thermogravimetry

and differential thermal analysis (TG-DTA) was performed to obtain the amount of calcium hydroxide in hardened

paste. An ignition loss was measured according to JIS R5202.


35
Table 2.1 Characteristics of cement.

Types of Density Blain Chemical composition(%) Mineral composition(%)


cement (g/cm3) (cm2/g) SiO2 Al2O3 Fe 2O3 CaO MgO SO3 C3S C2S C3A C4AF
N 3.16 3250 21.5 5.4 2.9 64.3 1.9 1.8 52 23 10 9
H 3.13 4650 20.4 4.8 2.5 64.4 2.3 3.2 62 12 9 7
L 3.24 3330 25.9 3.2 3.7 63.1 0.1 2.4 27 54 2 11

Table 2.2 Properties of fly ash used for reaction rate tests.

Blain fineness Chemical composition(%) Density Insoluble


Types of Fly ash
(cm2/g) SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O SO3 TiO2 MnO (g/cm3) matter(%)
4A 3900 73.1 16.9 3.0 1.6 1.2 0.3 1.0 0.2 0.7 0.0 2.11 93.4
4B 4120 59.4 24.9 6.0 3.2 1.0 0.6 1.2 0.1 1.0 0.1 2.19 89.1
4C 4060 52.7 24.2 12.4 3.2 1.0 0.4 1.4 0.4 0.8 0.1 2.41 84.4
2A 3300 62.0 22.3 4.7 2.8 1.9 1.2 0.3 1.1 1.1 0.0 2.21 91.3
2B 4240 46.9 23.9 11.0 2.8 4.1 1.2 0.2 3.3 1.2 0.1 2.39 87.5

Table 2.3 Mineral composition and glass phase for fly ash determined by XRD-Rietveld analysis.

Types of fly ash Mullite Quartz Magnetite Glass phase Ignition loss(%)
4A 4.6 7.8 0.5 85.1 2.0
4B 10.0 6.7 2.4 78.5 2.5
4C 10.5 8.8 4.4 73.0 3.4
2A 11.3 9.2 0.6 76.3 2.6
2B 8.9 2.8 2.2 80.7 5.4

Table 2.4 Experimental plan for hydration rate test

Types of W/B Curing Curig age


Ca(OH)2 : Fly ash
fly ash (%) Temperature(℃) (days)
4A 60 7,14,21,28 and 56
1:1 50
2B 40 1,4,5,7 and 28

Note: W/B = Water / (Fly ash + Calcium hydroxide reagent).

2.2.2 Acceleration test for fly ash cement paste


In order to investigate the insoluble residue contents (unhydrated amount fly ash) until the final reaction in fly

ash cement paste, the acceleration test at 80℃ of five fly ash blended cement samples, which are 4A, 4B, 4C, 2A

and 2B in addition to N sample as a reference mixture, was conducted. Each sample includes different type of fly

ash and with 10% by mass fly ash replacement ratio. The experimental plan is given in Table 2.5. All samples

mixed according to JIS R5201 before they were casted into 40×40×160 mm prisms, and then cured in water at 20℃

for 1day. After 1day, the samples were cut as 3 mm thick, and cured in water at 80℃. At the testing age, the

samples were soaked in acetone to stop the hydration reaction, dried at 105oC and further ground by ball mill until

particles became smaller than 75 μm.

36
Table 2.5 Experimental plan for acceleration test

Fly ash W/C W/B Curing Curig age


Designation
(mass %) (%) (%) Temperature(℃) (days)
4A
4B
5hr, 1, 2, 3, 4,
4C 10 45
50 80 in water 7, 14, 28, 56
2A
and 94
2B
N 0 -

Note: W = water, C = Cement, W/B = Water / (Fly ash + Cement).

2.2.3 Measurement of velocity of fly ash reaction rate


Four fly ash cement paste samples, which are 4A1, 4A3, 2B1 and 2B3 with two types of fly ash and with fly ash

replacement ratios by mass 10% and 30% in addition to N sample as a reference mixture, were tested to measure

the reaction rate in the mixtures. The experimental plan is shown in Table 2.6. Moreover, the effect of different

curing conditions was also considered. All samples mixed according to JIS R5201, and then cured in water at 20oC

for 7 days. After 7 days, the samples were cut as 3 mm thick, and cured in water at 5℃, 20℃ and 40℃. At the

testing age, the samples were soaked in acetone to stop the hydration reaction, dried at 105oC and further ground

by ball mill until 75 μm. The amount of unhydrated fly ash was estimated based on the selective dissolution method,

and the rate of hydration of cement minerals was calculated by the results obtained by the XRD-Rietveld analysis;

both methods apply the same procedures explained in section 2.1. TG-DTA was conducted to observe the amount

of the calcium hydroxide and calcium carbonate in hardened paste (Ohsawa et al. 1999). An ignition loss was tested

according to JIS R5202.

Table 2.6 Experimental plan for reaction rate test

Types of Fly ash W/C W/B Curing Curig age


Designation
fly ash (mass %) (%) (%) Temperature(℃) (days)
4A1 10 45.5
4A
4A3 30 38.5
5, 20 and 40 14, 21, 35, 63
2B1 10 50 45.5
2B in water and 198
2B3 30 38.5
N - 0 0

Note: W/B = Water / (Fly ash + Cement).

37
2.2.4 Self-healing of fly ash
Ten mixtures were tested to evaluate the self-healing ability of mortar incorporating fly ash, considering different

types of cement, fly ash, fly ash replacement ratio and curing conditions. The experiments used ordinary Portland

cement (N), high early strength Portland cement (H) and low heat Portland cement (L). The fly ash characteristic

properties, such as blain fineness, ignition loss and glass phase, were given in Table 2.7. The mortar mixtures were

prepared with the proportions summarized in Table 2.8. It is well known that the reaction rate of fly ash is late, in

fact that may cause a decrease of the compressive strength at early age. This drawback can be improved by using

high early strength Portland cement type, therefore, the high early strength Portland cement type (H) was chosen in

order to avoid the decrease of compressive strength at early ages. In addition, low heat Portland cement has a large

amount of C2S (belite), unlike ordinary Potland cement and that may improve the long-term reaction. The physical

properties and chemical composition of ordinary Portland cement (N), high early strength Portland cement (H) and

low heat Portland cement (L) are shown in Table 2.1.

The cement replacement ratio of these two types cement could be calculated from mineral composition. The ten

mortar mixtures consist of one OPC mortar sample (N) without fly ash replacement and eight fly ash cement mortar

samples with mixing ratio 7 to 3 for high early strength Portland cement to low heat Portland cement, respectively,

and with fly ash replacement ratio as 10%, 20% and 30% by volume. The water to cement ratio (W/C) in this

experiment varied as 45%, 55% and 65%. After casting, samples were cured under a condition of 20℃ for 4 weeks

in water. In order to introduce the micro cracks in the sample, the freeze/thaw test was performed until relative

dynamic modulus of elasticity (RDM) 60% and 80%, in accordance with JIS A 1127. The relative dynamic

modulus of elasticity was determined by measuring the resonant frequency and was calculated as the percentage

fraction of square transverse frequency after cycles of freeze/thaw test to square transverse frequency at 0 cycle

of freeze/thaw test. After deterioration of samples, different curing conditions were applied as 20℃ for 1 week and

40℃ for 4 weeks in water. Then, tests for physical properties, which are compressive test, bending test, accelerated

carbonation depth test and point count procedure, of samples at testing ages were performed.

38
Table 2.7 Properties of the fly ash used for self-healing ability tests.

Blain fineness Ignition Glass


Types of fly ash
(cm2/g) loss(%) Phase(%)
A 4090 1.5 88.8
B 4040 2.1 86.2
C 3160 1.9 88.4
D 2830 1.7 80.6

Table 2.8 Mortar mix proportion

Types Degree of Curing age and Unit weight (kg/m3)


W/C F/(S+F)
Designation of deterioration temperature after cement
(%) (vol.%) W F S
fly ash (%) deterioration(℃)
N H L
N - - 342 621 - - - 1241
HL - - 60 and 80 342 - 434 187 - 1241
FA20 55 20 342 - 434 187 97 1119
FA10 10 342 - 434 187 195 994
20 at 1 week
FA30 A 30 and 342 - 434 187 293 870
FAX 45 40 at 4 weeks 316 - 492 211 221 1125
in water
FAY 65 60% 380 - 409 175 183 937
FB B 20 342 - 434 187 195 994
FC 55 C 342 - 434 187 207 994
FD D 342 - 434 187 205 994

Note: W = water, N = Ordinary Portland cement, H = High early strength Portland cement, L = Low heat Portland

cement, F = Fly ash, S = Sand.

In the point count test, which was used to quantify the change of the number of cracks after deterioration, the

number of cracks in sample was analyzed using a modified point count procedure which was suggested by T.

Matsumura et al. (1997) who investigated experimentally the relationship between the number of cracks and degree

of frost damage by means of image analysing microscope (with 100×magnification). It was inferred that the

decrease of relative dynamic modulus can increase the number of cracks in concrete sample after the frost damage

has taken place. In the accelerated carbonation test, at condition of 20℃ and 60% RH, after 4 weeks the samples

were dried in advance, and then this test was carried out with 5% CO2 concentration at 1, 4 and 13 weeks. Finally,

the PoreMaster 33 mercury intrusion porosimetry (MIP) was used to measure the pore structure distribution and its

modification in the mortar samples, so that the influence of fly ash and the curing condition on the change of pore

structure volume in studied samples can be examined. In order to determine the hydration rate of fly ash for normal

slab concrete in Asahikawa, Japan, the natural temperature and humidity were investigated for 2 years. The water

to cement ratio (W/C) of normal concrete in this experiment was 50%.

39
2.3 Results and discussion
2.3.1 Reaction of fly ash and calcium hydroxide reagent
The methods based on selective dissolution (Ohsawa et al. 1999; Matsui et al. 1999) can be adopted as the

calculation method of reaction rate of unhydrated fly ash. The phase constitution of fly ash in both unreacted

process and reaction process is considered in Fig. 2.1. Glass contents in the unreacted process can be calculated by

deducting the amount of crystalline phase and ignition loss from unreacted fly ash amount. The glass contents can

be divided into two parts which are acid soluble amount (As) and acid insoluble matter (Au). As for the crystal

phase part, assuming that it does not dissolve in acid, acid insoluble matter can be obtained by excluding the

amount of crystal phase from the test results for the insoluble residue quantity. Thus, the calculation method of

reaction ratio by selective dissolution method can be explained as follows. When fly ash reacts, it becomes the

substance where portion of insoluble matter dissolves in acid. Here, selective dissolution test is carried out, and the

unreacted part of acid insoluble matter is measured as an insoluble residue. Then, the dissolved contents can be

estimated from reacted weights of paste. The following patterns are considered for the calculation method.

Glass contents(A ) Crystalline Ignition


Unreacted
Non dissolved acid Dissolved acid phase loss
process
( Au ) (As ) ( Cr ) (%)

Glass contents(A )
Crystalline Ignition
Reaction Non dissolved Dissolved acid
phase loss
process acid Reaction Dissolved acid
( Cr ) (%)
( Au - Ar ) ( Ar ) (As )

Fig. 2.1 Schematic reaction models

Pattern 1: it is related to the melting ratio of glass phase and crystal phase (Cr). This pattern ( is calculated as

follows:

40
As  Ar
Re c  (2.1)
Au  As  Cr

Pattern 2: it is related to the melting ratio for glass phase. This pattern Rg is expressed as shown in Eq. (2.2):

As  Ar
Rg  (2.2)
Au  As

Pattern 3: it can be estimated from the melting ratio for acid insoluble matter. This pattern Rus is given as shown

in Eq. (2.3):

Ar
Rus  (2.3)
Au  Cr

Pattern 4: It can be obtained from the melting ratio for the acid insoluble matter in the glass phase. This pattern

Rau is calculated as in Eq. (2.4):

Ar
Rau  (2.4)
Au

Fig. 2 shows the comparison of the consumption of calcium hydroxide contents and reaction rate of fly ash for each

pattern. Relationship between the reaction rate and the calcium hydroxide consumption for fly ash cement paste can

be expressed as given in Eq. (2.5):

74.09
Wch  (Wcc  )
Ca(OH ) 2  W0  100 .09 (2.5)
(100 Rw)  0.01

Where, W0 is calcium hydroxide contents within 1g of un-reacted sample (g/g), Wch is the calcium hydroxide

weights which include in unreacted sample 1g (g/g), Wcc is the calcium carbonate weights in hydrated sample 1g

(g/g) and Rw is the amount of bound water in 1 g within hydration sample (%).

By observing Figs. 2.2 (a) and (b), which represent patterns 1 and 2, respectively, it can be inferred that, although

calcium hydroxide consumption is almost zero, the reaction rate is not identical to zero. This is due to the fact that

pozzolanic reaction exists in fly ash and calcium hydroxide.

However, it cannot be determined which part of fly ash that reacting with calcium hydroxide. The data plotted in

Figs. 2(c) and (d), which explains behavior of patterns 3 and 4, respectively, reveals that when the consumption of

calcium is almost zero, there is no contribution of calcium to reaction rate. It is obvious from the comparison given

in Fig. 2.2 that pattern 4 gives the best response for reaction rate of fly ash compared to other patterns 1, 2 and 3.

41
The reason for this response is due to the assumption of pattern 4, in which the reaction rate is calculated based on

the melting ratio for the acid insoluble matter in the glass phase, meaning that there is no contribution of crystalline

phase (Cr) to reaction rate. From the results above, the reaction of fly ash with calcium hydroxide can be

considered, when acid insoluble matter in the glass phase exists, as given by Eq. (4).
0.5 0.5
2B y = 1.8622x + 0.1164 2B
0.5 y = 1.5903x + 0.0994
0.4 0.5
0.4 R² = 0.977
R² = 0.977 2B y = 1.8622x + 0.1164 2B
y = 1.5903x + 0.0994 4A
0.4
0.3 4A 0.4 R² = 0.977
R² = 0.977 0.3
rate (g/g)

rate (g/g)
4A
0.3
0.2 4A 0.3
rate (g/g)

0.2

rate (g/g)
y = 1.1333x + 0.0634
Reaction

0.2

Reaction
0.1 0.2
0.1
R² = 0.8367
y = 1.1333x + 0.0634
Reaction

y = 0.9849x + 0.0551

Reaction
0.1
0 0.1
R² = 0.8367 0 R² = 0.8367
y = 0.9849x + 0.0551
0
-0.1 R² = 0.8367 -0.10
-0.1 0 0.1 0.2 0.3 -0.1 0 0.1 0.2 0.3
-0.1 -0.1
Consumption of0.1
CH contents(g/g) Consumption of CH contents(g/g)
-0.1 0 0.2 0.3 -0.1 0 0.1 0.2 0.3
Consumption of CH contents(g/g) Consumption of CH contents(g/g)

0.5 (a) Pattern 1 0.5 (b) Pattern 2


2B
0.5 0.5 y = 2.0412x + 0.032
0.4 y = 1.7182x + 0.0269 2B 0.4
2B
R² = 0.977 R² = 0.977
0.4 0.4 y = 2.0412x + 0.032 4A
rate (g/g)
rate (g/g)

0.3 y = 1.7182x + 0.0269 2B 0.3


4A
R² = 0.977 R² = 0.977
4A
rate (g/g)
rate (g/g)

0.3 0.3
0.2 4A 0.2
Reaction
Reaction

0.2 0.2
0.1 0.1 y = 1.199x + 0.0095
Reaction

y = 1.0337x + 0.0082
Reaction

0.1 R² = 0.8367 0.1 R² = 0.8367


0 y = 1.0337x + 0.0082 0 y = 1.199x + 0.0095

R² = 0.8367 R² = 0.8367
0 0
-0.1 -0.1
-0.1 0 0.1 0.2 0.3 -0.1 0 0.1 0.2 0.3
-0.1 -0.1
-0.1 Consumption
0 of0.1
CH contents(g/g)
0.2 0.3 -0.1 Consumption
0 of CH contents(g/g)
0.1 0.2 0.3
Consumption of CH contents(g/g) Consumption of CH contents(g/g)

(c) Pattern 3 (d) Pattern 4

Fig. 2.2 Comparison of the consumption of CH contents and reaction rate of fly ash

2.3.2 Fly ash cement paste acceleration reaction test at 80oC


Fig. 2.3 presents the comparison of the pozzolanic reaction rate and curing age (days) for 5 types of fly ash

cement paste. It can be inferred from the figure that the reaction rate for sample 4C, at early age, is higher than

other samples. At 94 days, the reaction rate for sample 2A is increased to 100 %, while samples 4C and 2B almost

have the same reaction rate. The results shown in Fig. 2.3 reveal that acid insoluble matter of the glass phase of fly

ash (FA) reacts almost 100%, for all types of fly ash in the study at 94 days. In Fig. 2.4, the relationship between

remaining contents of calcium hydroxide and curing ages (days) is described. The amount of calcium hydroxide in

42
the hardened paste can be calculated as shown in Eq. (2.5). In comparison with N sample, the remaining contents of

calcium hydroxide in fly ash cement samples after 2 days decrease because of its pozzolanic reaction. For the

remaining calcium hydroxide contents at each age, it is found that there is no significant difference among the types

of fly ash. Although the pozzolanic reaction rate is almost 100%, the calcium hydroxide in paste remains after 56

days. Therefore, in case of cement paste with 10% by mass fly ash replacement ratio, it can be considered that the

glass phase and insoluble residue contents in fly ash could react with the calcium hydroxide.

100 0.3

Remaining contents of Ca(OH)2(g/g)


80
Reaction rate(%)

60 0.2

4A
40
4B
20 4C 0.1
4A 4B
2A
4C 2A
0
2B 2B N
0
-20
1 10 100
1 10 100
Age(days) Age(days)

Fig. 2.3 Comparison of the pozzolanic reaction rate Fig. 2.4 Remaining contents of calcium hydroxide
of different fly ash types versus curing age (days). versus curing age (days).

2.3.3 The reaction rate of fly ash cement paste


2.3.3.1 Amount of calcium hydroxide
In Fig. 2.5, the amount of remaining calcium hydroxide in hardened fly ash paste based on the XRD-Rietveld

analysis is plotted versus testing age for different curing conditions. From the figure, it can be revealed that in the

case of cement paste incorporating fly ash (samples 4A and 2B), the amount of remaining calcium hydroxide is

lower than N sample which is the case of OPC paste without fly ash. Also, it is obvious that fly ash cement samples

with fly ash replacement ratio by mass 30%, and with high curing temperature in water (samples 4A3-40℃ and

2B3-40℃), yielded smaller remaining amount of calcium hydroxide compared to other samples with smaller fly

ash replacement ratio (samples 4A1-40℃ and 2B1-40℃). This observation confirms that large amount of fly ash as

30% replacement ratio and high curing temperature in water can increase the consumption amount of calcium

hydroxide. The reason for this behavior is due to the pozzolanic reaction of fly ash, which increases with increasing

the amount of fly ash and curing temperature, and thus consumes more calcium hydroxide that comes from cement,

43
which also agrees with results in previous research (Liu et al 2005; Termkhajornkit et al 2009)

0.4

Remaining contents of Ca(OH)2(g/g)


4A1- 5˚C
0.3 4A1-20˚C
4A1-40˚C
0.2 4A3- 5˚C
4A3-20˚C
0.1
4A3-40˚C
N- 20˚C
0
0 20 40 60 80

Age(days)

0.4
Remaining contents of Ca(OH) 2(g/g)

2B1- 5˚C
0.3 2B1-20˚C
2B1-40˚C
0.2 2B3- 5˚C
2B3-20˚C
0.1
2B3-40˚C
N- 20˚C
0
0 20 40 60 80

Age(days)

Fig. 2.5 Remaining contents of Ca(OH)2 versus curing age(days).

2.3.3.2 Prediction equation of reaction rate


For the hydration reaction of cement, the reaction rate for each cement minerals calculated by Sagawa et al.

(2004 and 2005) equation can be expressed as Eq. (2.7):

dx 1
 k( ) (2.7)
dt x

Where, x is reaction thickness to inner particle, k is a constant represent the reaction rate, and t is reaction time. The

hydration rate (k) can be derived from the following equation:

 Ea
k  A exp( ) (2.8)
RT

Where, A is experimental constant, Ea is apparent activation energy (J/mol), R is universal gas constant and T is

Kelvin temperature. In addition, an average particle size of fly ash in radius (r) can be expressed as in Eq. (2.9):

44
3
r (2.9)
(   s)

Where, s is the blaine fineness of fly ash (cm2/g) and  is the density of fly ash (g/cm3). The above equation

can be applied in the case of x r however, if x r the following equation can be expressed as follow:

  r  x 3 
Fr (%)  1     100 (2.10)
  r  

2.3.3.3 Reaction of cement in fly ash blended paste


The rate of reaction of cement for fly ash paste sample 4A1 with curing temperature varies as 5℃, 20℃ and 40℃

is given in Fig. 2.6. This data is calculated by the unreacted mineral composite based on the XRD-Rietveld analysis.

Although the cement, which was blended with fly ash, could not react fully as 100%, the reaction rate is almost

85%, containing C2S and C4AF. The figure shows that there is influence of the curing temperature on the cement

reaction rate for different ages; this is obvious when comparing the response for reaction rate of sample 4A1-20℃

with sample 4A1-5℃.

100

90
Reaction rate(%)

80

70
4A1- 5˚C
60 4A1-20˚C
4A1-40˚C
50
0 50 100 150 200

Age(days)

Fig. 2.6 Rate of reaction of cement in fly ash cement mixture.

2.3.3.4 Reaction of fly ash


For the rate of fly ash reaction, the dissolved ratio can be estimated by the insoluble matter contents in glass

phase. The apparent activation energy (Ea) can be calculated by Eq. (8) and was obtained based on a nonlinear

45
regression analysis applied to the experimental results; its obtained values is given Table 2.9. Both two types of fly

ash have the same apparent activation energy (Ea), but the experimental constant (A) is slightly different.

Fig. 2.7 shows the actual and predicted hydration rate (straight line) of fly ash in fly ash cement samples for

different curing conditions and fly ash replacement ratios. In the case of 4A1 and 2B1 samples (10% by mass

replacement ratio of fly ash), the actual and predicted hydrations at the final reaction in paste have the same

tendency for all curing condition cases. However, in the case of 4A3 and 2B3 samples (30% by mass replacement

ratio of fly ash), there is no identical tendency at the final reaction for all curing conditions. It can be inferred from

Fig. 7 that the reaction rate of fly ash depends on the curing temperature, curing age and fly ash replacement ratio.

Table 2.9 Velocity coefficient of reaction rate

Note: Ea = Apparent activation energy, A = Experimental constant

T=5℃ T=5℃
T=20℃ T=20℃
T=40℃ T=40℃
100 Model Simulation 5℃ 4 A1 Model Simulation 5℃ 4 A3
100
Model Simulation 20℃ Model Simulation 20℃
Model Simulation 40℃ Model Simulation 40℃
80 80
Reaction rate(%)

Reaction rate(%)

60 60

40 40

100
80 100
80
20 60
40 60
40
20
Reaction rate(%)

Reaction rate(%)

20
0 20
0
0 50 Age100
(days) 150 200 0 50 Age100
(days) 150 200
0 0
0 50 100 150 200 0 50 100 150 200
Age(days) Age (days)

T=5℃ T=5℃
T=20℃ T=20℃
T=40℃ T=40℃
Model Simulation 5℃ 2 B1 Model Simulation 5℃ 2 B3
100 100
Model Simulation 20℃ Model Simulation 20℃
Model Simulation 40℃ Model Simulation 40℃
80 80
Reaction rate(%)

Reaction rate(%)

60 60

40 40

100
80 100
80
60
40 60
40
Reaction rate(%)

Reaction rate(%)

20 20
0 20 20
0
0 50 Age100
(days) 150 200 0 50 Age100
(days) 150 200

0 0
0 50 100 150 200 0 50 100 150 200
Age (days) Age (days)

Fig. 2.7 Predicted results of fly ash reaction rate versus experimental values.

46
2.3.4 Optimization of fly ash replacement ratio
The above results and discussions clarified that the behavior of fly ash is attributed to the pozzolanic reaction

due to the glass content of acid insoluble. This glass content can be obtained by deducting the amount of crystalline

from the acid insoluble matter in the fly ash. The pozzolanic reaction may continue after the full hydration of

cement, and consequently, remaining pores can be filled by this pozzolanic reaction. Fig. 2.8 shows the calculated

values of reaction rate of fly ash and cement minerals in fly ash blended cement paste (as discussed in section

3.3.2) for which the temperature and relative humidity have been measured for 2 years. It is obvious that the

hydration rate of fly ash and C2S continues, even, after full hydration of C3S at the time of 360 days.

It is known that when the relative dynamic modulus of elasticity was about 60%, the relative length change was

about 1×10-3 due to fact that internal cracking in the concrete sample took place for the rapid freeze and thaw test

according to ASTM C 666 procedure A (Koh et al. 1981). Therefore, to obtain optimization of fly ash replacement

ratio in concrete, it is assumed that the relative dynamic modulus of elasticity for concrete structure deteriorates

until 60% and micro cracks in concrete are about 0.3 % (by volume). An increase of volume by 0.3% is typical to 3

liters, and the amount of fly ash can be calculated as shown in Eqs. (2.11) through (14).

VFA  ( Rwint er  Rspring) 


V  G (2.11)
 100 

Where;

V is increase of volume by 3 liters due to micro cracks, VFA is increase of volume by fly ash reaction, equals to 2.4

which was determined based on the current experiment), R winter is the percentage of fly ash reaction in start of

winter (see Fig. 17), and it is equal to 25.6%, R spring is the percentage of fly ash reaction in start of spring (see Fig.

17), and it is equal to 13.1%, and G is glass phase of fly ash and it can be expressed as [Fw (Fins-Fcry )/(100 f)] .

Hence, Eq. (2.11) can be rewritten as follows,

3 = 2.4(25.6 - 13.1)/100× Fw (Fins-Fcry )/(100 f) (2.12)

Fw  3/[2.4  0.125(Fins-Fcry )/(100 f)] (2.13)

47
Where, Fw is the fly ash content (kg/m3), Fins is insoluble residue of fly ash (%), Fcry is the crystal content in fly ash

(%) and f is density of fly ash (g/cm3), the fly ash substitution ratio F/B can be calculated from Eq. (2.14).

Fw
F/B(%)   100 (2.14)
Cw  Fw

Where, Fw is fly ash content (kg/m3) and Cw is cement unit weight (kg/m3). The fly ash reaction volume,

can be calculated as shown in Eq. (2.15).

Frv  Fw(Fins-Fcry )/(100 f)] (2.15)

In addition, to avoid filling pores in concrete, the maximum fly ash contents can be determined based on a

concrete unit that can be represented as water unit, cement unit, density of cement and the fly ash reaction as

follows:

Ww-(1.06  Cw/c + 2.4  Frv) > 0 (2.16)

Fw < (Ww-1.06  Cw/ρw/ρc)/2.4 (2.17)

Where, Ww is unit water weight (kg/m3), is density of cement (g/cm3), and Frv is fly ash reaction volume.

From the above equations as shown in Eqs. (2.11) through (14), the fly ash replacement ratio determines the

glass content and water cement ratio which are needed to both heal the deteriorated concrete until RDM 60% and to

fill pore for long periods. Fig. 2.8 shows the hydration rate with C3S and C2S fly ash based on the natural

temperature and humidity in internal slab specimens in Asahikawa, Japan. This hydration rate for the mineral

composite was calculated by using reaction rate equation (chapter 2.3.2). From the figure, the hydration rate for

C3S increased until 360 days and was also constant after 360 days of curing age, however, the hydration rate of C2S

and fly ash did not completely hydrated at 720 days.

Fig. 2.9 represents the relationship between the fly ash replacement ratio and water to cement ratio in

deteriorated concrete for 60%, 70% and 80% of relative dynamic modulus of elasticity (RDM), which were derived

from the above equations. This relationship has upper and lower limits for calculation of fly ash replacement ratio

at different water to cement ratio as shown in the figure to ensure proper self-healing performance. Accordingly, it

can be suggested that the practical fly ash replacement ratios in fly ash concrete mix could be 10% to 15% and 11%

to 20% for water to cement ratios (W/C) 0.50 to 0.55 and 0.55 to 0.60, respectively.

48
100
100
100100 35
909090 RDM60%
90
30

Fly ash replacement ratio(%)


808080 RDM70% Upper limit of pore filling
80
707070 RDM80%
70 25
Hy dration rate(%)
Hy dration rate(%)
Hy dration rate(%)
Hy dration rate(%)

606060
60
20
505050
50
404040 15
40 Start of winter(25.6%)
303030
30 10
Start of spring(13.1%)
202020 20
5 Lower limit of self-healing effect
101010 10
C3SC3S C2S
CC3S C2S FA
C2S G
FA
FA GGlass
GFA
000 0 3S C3S C2SC2S FA G
0
0 0 0 0120120120
120 240240240
240 360360360
360 480480480
480 600600600
600 720720720
720 50 51 52 53 54 55 56 57 58 59 60
AA ge(days)
A ge(days)
ge(days)
A ge(days) Water to cement ratio(%)

Fig.
Fig.2.8
2.8Calculating
Calculating reaction rate of
reaction rate of fly
flyash
ashand
andcement
cement Fig. 2.9 Relationship between fly ash replacement
minerals
minerals basedbased on measurement
on measurement of environment
of environment temperature ratio and water cement ratio for different damage
temperature and humidity
and humidity until 2 until
years.2 years. degrees.

2.3.5 Self-healing effects


Figs 2.10 through 12 show the results of compressive strength, bending strength and relative dynamic modulus

of elasticity for fly ash blended cement mortar samples with four fly ash types, replacement ratios vary as 10%,

20%, and 30% by volume and three types of cement. Ten mortar mixtures were prepared with the proportions

summarized in Table 2.8 and section 2.3. The examined mortar samples consist of N sample using OPC without fly

ash replacement, nine samples using combination of high early strength Portland cement and low heat Portland

cement. It can be seen that the investigation of each mortar sample includes four cases which are no cracking case,

after deterioration case, cured in water 20℃ at 1 week case and cured in water 40℃ at 4 weeks case; the results of

each case are shown in Figs. 2.10 through 12. After deterioration, it can be seen from the figures that most of the

investigated samples can heal the compressive strength, bending strength and RDM. However, degree of healing in

mortar samples incorporating fly ash is higher compared to N sample, which is the case of OPC mortar without fly

ash, especially for the case of curing in water 40℃ at 4 weeks. Also, it is evident that the curing condition in water

20℃ at 1 week is not enough to heal the deterioration compared to the curing condition in water 40℃ at 4 weeks,

implying that high temperature in curing deteriorated fly ash cement mortar samples is efficient for self-healing

performance. The fly ash has a pozzolanic reaction that reacts with calcium hydroxide from cement hydration and

produces C-S-H gel, which may heal micro cracks under high temperature curing condition. Consequently, it is

49
expected that the mortar made with cement and fly ash may improve various properties such as compressive

strength, bending strength and RDM, showing self-healing ability. These results agree with the findings in the

previous work (Termkhajornkit et al. et al. 2009).

No cracking After deterioration No cracking After deterioration


Cured in water 20℃ at 1 week Cured in water 40℃ at 4 weeks Cured in water 20℃ at 1 week Cured in water 40℃ at 4 weeks
80 12
70
Compressive strength(MPa)

10

Bending strength(MPa)
60

50 8

40 6
30
4
20
2
10

0 0
N HL FA10 FA20 FA30 FB FC FD N HL FA10 FA20 FA30 FB FC FD
Mixture Designation Mixture Designation

Fig. 2.10 Self-healing effects on compressive Fig. 2.11 Self-healing effects on bending
strength for different mixtures and curing conditions strength for different mixtures and curing
conditions

No cracking After deterioration


Cured in water 20℃ at 1 week Cured in water 40℃ at 4 weeks
140

120

100

80
RDM(%)

60

40

20

0
N HL FA10 FA20 FA30 FB FC FD
Mixture Designation

Fig. 2.12 Self-healing effects on relative dynamic modulus of elasticity


for different mixtures and different curing conditions

Fig. 2.13 shows the relationship between accelerated carbonation time and carbonation depth for specimens

FA30, which include 30% by volume fly ash replacement ratio, considering the four studied cases. It is interesting

to note from the figure that the carbonation depth of sample FA30 in the case of curing in water 20℃ at 1 week

shows almost the same response as the case after deterioration. However, the sample carbonation depth, in the case

that it is cured in water 40℃ at 4 weeks, is less than 1 mm. Also, it is found that the carbonation depth for the case

cured in water 40℃ at 4 weeks is smaller than no cracking case, at any time of accelerated carbonation test,

50
implying that the curing condition of mortar with fly ash would affect the carbonation depth and self-healing. The

fact that curing in water at high temperature would result in continuing the hydration of both unreacted cement and

fly ash, which can fill micro cracks, is the cause of the observed performance in FA30 mortar sample for

carbonation test.

Fig. 2.14 shows the carbonation coefficient with respect to accelerated carbonation test for each investigated case

of the designated mixtures. It can be inferred from the figure that mixtures incorporating fly ash show smaller

carbonation coefficient and better performance compared to the mixtures without fly ash under curing condition of

40℃ at 4 weeks, even after deterioration. The reason due to the fact that incorporating fly ash in investigated

mortar would result in higher resistance to carbonation effect is because of the pozzolanic reaction of fly ash. The

pozzolanic reaction of fly ash may continue after the full hydration of cement, and consequently, induced micro

cracks can be filled or sealed by hydrated products of fly ash resulting in higher resistance to carbonation proceed.

Fig. 2.15 shows the effect of water to cement ratio on the carbonation coefficient in fly ash mortar samples FAX,

FA20 and FAY with different curing conditions. It is found that the fly ash mortar samples with lower water to

cement ratio lead to smaller carbonation coefficient and better self-healing performance for the studied cases. The

fact that smaller water to cement ratio may cause smaller volume of micro cracks and pores in mortar sample,

which is enough to be sealed by hydration products of cement and fly ash, is the reason for this better self-healing

ability and carbonation resistance. On the other hand, larger water to cement ratio may contribute into a large

volume of micro cracks and pores in mortar sample, which requires more hydration products of cement and fly ash

to be filled in order to have more resistance for carbonation test.

51
8 3
No cracking No cracking
After deterioration After deterioration
Cured in water 20℃ at 1 week Cured in water 20℃ at 1 week

Carbonation coefficient
6
Carbonation depth(mm)
Cured in water 40℃ at 4 weeks Cured in water 40℃ at 4 weeks
2

1
2

0 0
0 1 2 3 4 N HL FA10 FA20 FA30 FB FC FD

Accelerated carbonation time(week0.5) Mixture Designation

Fig. 2.13 Change of carbonation depth for FA30 sample Fig. 2.14 Self-healing effects on carbonation
after different curing cases. coefficient for different mixtures

3
No cracking
After deterioration
2.5 Cured in water 20℃ at 1 week
Cured in water 40℃ at 4 weeks
Carbonation coefficient

1.5

0.5

0
FAX(W/C 45%) FA20(W/C 55%) FAY(W/C 65%)

Fig. 2.15 Change of carbonation coefficient for fly ash mortar samples
with different water to cement ratio subjected to CO2 for 13 weeks.

Fig. 2.16 the values of carbonation coefficient for two mortar mixes, which are N (OPC mortar mix), FA20 (fly

ash cement mortar mix with 20% by volume replacement ratio). The figure shows the results of carbonation

coefficient for seven cases of each examined mortar mixes, which are no cracking case, after deterioration (RDM

60%) case, cured in water 20℃ at 1 week (RDM 60%) case and cured in water 40℃ at 4 weeks (RDM 60%) case,

after deterioration (RDM 80%) case, cured in water 20℃ at 1 week (RDM 80%) case and cured in water 40℃ at 4

weeks (RDM 80%) case. It can be seen from the figure that fly ash mortar sample FA20 shows smaller carbonation

coefficient under curing condition 40℃ in water at 4 weeks for both deterioration indexes 60% and 80% compared

to OPC mortar mix N for the same curing condition. This behavior can be attributed to the fact that fly ash is a

pozzolanic active material that reacts with calcium hydroxide in cement producing C-S-H gel efficiently under high

temperature curing condition in water, which may fill micro cracks in the fly ash mortar mixtures, and, thus,

improve the self-healing performance and resistance to carbonation.

52
No cracking
After deterioration (RDM80%)
3
Cured in water 20℃ at 1 week (RDM80%)
Cured in water 40℃ at 4 weeks (RDM80%)
After deterioration (RDM60%)
Cured in water 20℃ at 1 week (RDM60%)

Carbonation coefficient
Cured in water 40℃ at 4 weeks (RDM60%)
2

0
N HL FA20

Mixture Designation

Fig. 2.16 Change of carbonation coefficient for different damage degrees


and curing conditions

Various evaluation techniques, which are mercury intrusion porosimetry (MIP), micro-focus X-ray computed

tomography and so on, have been applied for investigating the effect of pore and micro cracks changes on

self-healing performance. A similar approach can be adopted to relate the pore structure and micro cracks due to

damage or self-healing.

Figs. 2.17(a), (b) and (c) show the change in relative pore volume with respect to pore size for three mortar

samples N, HL and FA20, respectively. These results which were obtained by MIP, for four different cases which

are no cracking case, after deterioration case, cured in water 20℃ at 1 week case and cured in water 40℃ at 4

weeks case are given in the figure. It should be noticed that mortar samples HL and FA20 incorporate a

combination of high early strength Portland cement type (H) and low heat Portland cement (L), while mortar

sample N incorporates OPC cement. In addition, mortar sample FA20 incorporates fly ash with 20% replacement

ratio. Pore size distributions were examined; the pore sizes were divided into two ranges, 6 to 50 nm of meso pores

and lager than 50 nm to 10 μm of micro pores. (F. de Larrard 2002 and M.M.Y. Delmi 2004). It is obvious from

Fig. 2.17 that the change in ratio of pore volume for 6 to 50 nm pore size is almost the same between N, HL and

FA20 mortar mixes. While after deterioration, the rate of pore volume for 50 nm to 10 μm pore size for N mortar

sample increased in comparison to HL and FA20 mortar samples, this is due to using of a combination of H and L

cement types. A notable reduction of pore volume rate for 50 nm to 10 μm pore size in fly ash mortar sample FA20
53
was obtained for 40℃ curing condition in water at 4 weeks case. The reason for this performance is that the fly ash

has a pozzolanic ability that produces C-S-H gel, which may fill 50 nm to 10 μm pore size under high temperature

curing condition.

Fig. 2.18 shows change of the number of cracks per unit length obtained by point count test for N, HL and FA10

mortar samples for different curing conditions after 80% deterioration of relative dynamic modulus of elasticity. It

is clear that FA10 sample, in comparison with N and HL samples, experienced greater reduction in the number of

cracks for 40oC curing condition in water at 4 weeks case with respect to the number of cracks for after

deterioration case (RDM 80%). This observation implies that products of fly ash can fill micro cracks under high

temperature curing condition, as an evidence for self-healing phenomena and self-healing effect on micro cracks,

which justifies the results of pore volume change obtained by MIP.

2 2
No cracking No cracking
N HL
After deterioration (RDM80%) After deterioration (RDM80%)
Relative pore volume

1.5 1.5
Relative pore volume

Cured in water 20℃ at 1 week Cured in water 20℃ at 1 week


Cured in water 40℃ at 4 weeks Cured in water 40℃ at 4 weeks

1 1

0.5 0.5

0 0
6-50nm 50nm-10μm 6-50nm 50nm-10μm
Pore diameter Pore diameter

(a) N sample (b) HL sample

2
No cracking
FA20
After deterioration (RDM80%)
Relative pore volume

1.5
Cured in water 20℃ at 1 week

Cured in water 40℃ at 4 weeks


1

0.5

0
6-50nm 50nm-10μm
Pore diameter

(c) FA 20 sample
Fig. 2.17 Change of pore volume of mortar mixtures for different curing conditions.

54
0.5 No cracking
After deterioration (RDM80%)

The number of cracks (number/mm)


Cured in water 20℃ at 1 week
0.4 Cured in water 40℃ at 4 weeks

0.3

0.2

0.1

0
N HL FA10
Mixture Designation

Fig. 2.18 Change of the number of cracks with mortar mixtures


for different curing conditions.

2.4 Conclusions
The purpose of this study is to investigate the hydration process, reaction rate of cement and fly ash in fly ash

blended cement paste, and self-healing ability in mortar incorporating fly ash for long-term performance. First, the

experiment examined the reaction rate of fly ash and consumption of calcium hydroxide by adding calcium

hydroxide reagents to fly ash. Then, reaction rate of fly ash and remaining contents of Ca(OH)2 were measured

based on acceleration reaction test and rate of reaction test in fly ash cement paste taking into consideration

different types of fly ash, fly ash replacement ratios, curing conditions. The performance of self-healing ability of

mortar samples incorporating fly ash was evaluated using ten mixtures with different types of cement, fly ash, fly

ash replacement ratio, water to cement ratio and curing conditions. Mortar samples were deteriorated by

freeze/thaw cycling until 60% and 80% relative dynamic modulus of elasticity to examine their characteristics

properties such as compressive strength, bending strength, accelerated carbonation depth, and pore structure

modification. Based on experiment results, the following conclusions can be derived, which are valuable

information to enhance the existing design codes with long-term performance of self-healing ability of fly ash that

is quite needed to prolong the service life of concrete structures:

(1) From the hydration performance of calcium hydroxide reagent and fly ash, the calculation of reaction rate of

fly ash using pattern 4 was found to have better response to experimental results compared to other patterns.

The reason for this response is due to the assumption of pattern 4, in which the reaction rate is calculated
55
based on the melting ratio for the acid insoluble matter in the glass phase, meaning that there is no

contribution of crystalline phase to reaction rate.

(2) Incorporating fly ash in cement paste would affect the hydration reaction rate and consumption of calcium

hydroxide. In comparison with N sample, which is the case of OPC sample without fly ash replacement, the

remaining contents of calcium hydroxide in fly ash cement samples after 2 days decreased because of its

pozzolanic reaction. For the remaining calcium hydroxide contents at each age, it was found that there is no

significant difference among the types of fly ash.

(3) Fly ash cement samples with fly ash replacement ratio by mass 30%, and with high curing temperature,

showed smaller remaining amount of calcium hydroxide compared to other samples with smaller fly ash

replacement ratio. This implies that large amount of fly ash as 30% replacement ratio and high curing

temperature in water can increase the consumption amount of calcium hydroxide. The reason for this behavior

is due to the pozzolanic reaction of fly ash, which increases with increasing the amount of fly ash and curing

temperature, and thus consumes more calcium hydroxide that comes from cement.

(4) It was found that reaction rate of fly ash depends on curing temperature, curing age and fly ash replacement

ratio.

(5) Based on the practical assumption and to ensure proper self-healing performance, practical fly ash replacement

ratios in fly ash concrete could be 10% to 15% and 11% to 20% for water to cement ratios 0.50 to 0.55 and

0.55 to 0.60, respectively.

(6) The investigated mortar samples can heal the bending strength, compressive strength, crack density and

relative dynamic modulus of elasticity after deterioration. However, degree of recovering in mortar samples

incorporating fly ash was higher compared to OPC mortar sample without fly ash, especially for the case of

curing in water 40℃ at 4 weeks. Also, it was clarified that the curing condition in water 20℃ at 1 week was

not enough to heal the micro cracks compared to the curing condition in water 40℃ at 4 weeks. The fly ash

56
has a pozzolanic reaction that reacts with calcium hydroxide from cement hydration and produces C-S-H gel,

which may heal micro cracks under high temperature curing condition.

(7) It was confirmed that the carbonation depth for the case cured in water 40℃ at 4 weeks is less than 1 mm and

smaller than the no cracking case at any time of accelerated carbonation test for fly ash mortar samples,

concluding that the curing condition of mortar with fly ash would affect the carbonation depth and self-healing

due to continuing the hydration of both unhydrated cement and fly ash. Incorporating fly ash in mortar would

result in higher resistance to carbonation effect is because of the pozzolanic reaction of fly ash. The pozzolanic

reaction of fly ash may continue after the full hydration of cement, and consequently, micro cracks can be

filled by hydrated products of fly ash resulting in higher resistance to carbonation test.

(8) The fly ash mortar samples with lower water to cement ratio resulted in smaller carbonation coefficient and

better self-healing performance for the studied cases. The fact that smaller water to cement ratio may cause

smaller volume of micro cracks and pores in mortar sample, which is enough to be sealed by hydration

products of cement and fly ash, is the reason for this better self-healing ability and carbonation resistance. On

the other hand, larger water to cement ratio may contribute into a large volume of micro cracks and pores in

mortar sample, which requires more hydration products of cement and fly ash to be filled in order to have

more resistance for carbonation test.

(9) The reductions of pore volume rate 50 nm to 10 μm in fly ash mortar sample FA20 were higher than those in

ordinary Portland cement sample N, for 40℃ curing condition in water at 4 weeks case. The reason for this

behaviour is that the fly ash has a pozzolanic ability that produces C-S-H gel, which may fill 50 nm to 10 μm

pore size under high temperature curing condition. All the above results show the potential performance of fly

ash as a pozzolanic material that can improve the self-healing ability.

57
References

Asaga, K.et al., “Effect of curing temperature on the hydration of portland cement compounds.” 9th International
Congress on the Chemistry of Cement, 4, pp. 181-187 (1992)

F. de Larrard, Construire en beton, ENPC Press, Paris (2002).

Hoshino, S., Hirao, H., and Yamada, K., “The application of XRD/Rietveld method to cements with amorphous
admixture as quantitative tool of mineral composition.” Cement Science and Concrete Technology, 59, pp. 14-21
(2005)

Itoh, T., Masaki, K. and Igarashi, H., “Relationship between property of early hydration of cement and aluminate
contents determined by rietveld method.” Cement Science and Concrete Technology, 56, pp. 29-35 (2002)

Izumi, F. and Ikeda, T., “A Rietveld-analysis program RIETAN-98 and its application to Zeolites.” Mater. Sci.
Forum, pp. 321-324 (2000).

JCI. “Technical committee reports on autogenous healing in cementitious materials.” Tokyo, Japan Concrete
Institute (2009)

Liu XY, Yao W, Zheng XF, Wu JP. Experimental study on self-healing performance of concrete. J Build Mater
2005(2).

M.M.Y. Delmi, Etude de l'hydratation et du couplage carbonatation-echanges hydriques dans les mortiers et betons.
PhD Thesis of the University of La Rochelle (2004).

Matsui, J. and Ikabata, T., “Hydration rate and strength development of low heat portland cement with large amount
of pozzolan.” Cement Science and Concrete Technology, 53, pp. 229-236 (1999)

Matsumura T, Katsura O and Yoshino T., Properties of frost damaged concrete and the extimation of the degree of
frost damage, Journal of structural and construction engineering (563), 9-13 (2003)

Ohsawa, S., Sakai, E. and Daimon, M., “Reaction ratio of fly ash in the hydration of fly ash-cement system.”
Cement Science and Concrete Technology, 53, pp. 96-101 (1999)

Saeki, T. and Nagataki, S., “Fundamental study on reaction between fly ash and calcium hydroxide.” JCA
proceedings of cement and concrete, 51, pp. 162-167 (1997).

Saeki, T., Paulo, J.M. Monteiro, “A model to predict the amount of calcium hydroxide in concrete containing
mineral admixtures.” Cement and Concrete Research, 35(10), pp. 1914-1921 (2004).

Sagawa, T. Katsura, O., Sekiguchi, T., and Wakasugi, S., “Hydration rate of portland cement compounds and
strength development in consideration of particel size distribution.” Cement Science and Concrete Technology, 59,
58
pp. 45-52 (2005)

Sagawa, T., Katsura, O., Sekiguchi, T., and Wakasugi, S., “Effect of water to cement ratio and curing condition on
hydration of portland cement compounds.” Cement Science and Concrete Technology, 58, pp. 23-30 (2004)

Scrivener, K.L., Füllmann, T., Gallucci, E., Walenta, G. and Bermejo, E., “Quantitative study of Portland cement
hydration by X-ray Diffraction/Rietveld analysis and independent methods” Cement and Concrete Research, 34(9),
pp. 1541-1547 (2004).

Termkhajornkit, P., Nawa, T., Nakai, M., Saito, T., “Effect of fly ash on autogenous shrinkage.” Cement and
Concrete Research, 35(3), pp. 473-482 (2005)

Termkhajornkit, P., Nawa, T., Yamashiro, Y. and Saito, T., “Self-healing ability of fly ash–cement systems.”,
Cement and Concrete Composites, 31, pp. 195-203 (2009).

Whitfield, P. S. and Mitchell, L. D., “Quantitative Rietveld analysis of the amorphous content in cements and
clinkers.” Journal of Materials Science, 38(21), pp. 4415-4421 (2003).

Yamaguchi, G. et al., “Rate of hydration of cement compounds and portland cement estimated by X-ray diffraction
analysis.” 4th International Symposium on Chemistry of Cement, 1, pp. 496-499 (1960)

Yamamoto, T. and Kanazu, T., “Investigation on the pozzolanic reaction of fly ash-clarification of the mechanism
of pozzolanic reaction and optimization of accelerated chemical test (API method).” Laboratory Rep. No.N04008,
Central Res. Inst. of Electric Power Industry, Civil Engineering Research (2004).

Yoshiro, K. and Toshio H., “On the test method for resistance of concrete to freezing and thawing.” Concrete
Journal, Vol. 19, No. 9, pp. 16~22 (1981),

59
CHAPTER 3 SELF-HEALING EFFECT AND FROST RESISTANCE OF

FLY ASH BLENDED CONCRETE

60
3.1 Overview
As demonstrated in Chapter 2, incorporation of fly ash in cement mortar or paste as self-healing agent had a

positive effect on the self-healing ability after deterioration caused by freeze and thaw cyclic behavior in

accordance with ASTM C 666 procedure A. The self-healing ability was dependent on (1) the level of replacement

ratio, (2) finenesses and (3) deterioration degree (60 or 80 % relative dynamic modulus of elasticity) by

measurements of the accelerated carbonation test, compressive strength, bending strength, relative dynamic

modulus of elasticity and cracks and pore structure observation. In addition to the findings, the effect of kind of fly

ash materials on self-healing performance is insignificant.

However, in order to design or develop more durable autogenous healing concrete, the effect of self-healing

ability on fly ash blended concrete has not been fully understood and evaluated. In addition to the problem, when

concrete is exposed to freeze and thaw condition, it is well-known that introduction of micro crack within concrete

sample is difficult, thus new damage technique is needed in order to induce the micro cracks. Filling effect of fly

ash hydrates due to cracking and healing conditions should be valid.

Therefore, the objective of this chapter is to clarify the influence of fly ash blended concrete with and without air

entraing agent on self-healing ability after deterioration caused by frost and thaw cycling test. The effects of

environmental conditions on the self-healing ability in fly ash blended concrete sample incorporating high early

strength Portland cement and low heat Portland cement were examined and the self-healing ability of fly ash

blended concrete with air entraing agent was investigated. In addition, new damage technique was suggested by

application of compression testing. Finally, the visualization of self-healing performance was examined by means

of optical micro scope after cracking and healing condition.

61
3.2 Experimental program
The experiments were designed to achieve the objective of this chapter, with is to evaluate self-healing ability of

concrete sample made with high early strength Portland cement, low heat Portland cement and fly ash as

self-healing agent with and without air entraing agent by measurements of relative dynamic modulus of elasticity,

carbonation depth. Frost resistance and compressive strength in all investigated concrete samples were also tested.

3.2.1 Experimental materials


The chemical and physical properties of experiments used ordinary Portland cement (N), high early strength

Portland cement (H) and low heat Portland cement (L) were tabulated in Chapter 2. Fine and coarse aggregates

characteristic properties, such as surface dried density, absolute dried density, water absorption ratio and coarse

grain ratio were given in Table 3.1. In series 1, three mixtures were tested to evaluate the self-healing ability of

concrete incorporating fly ash and consist of one OPC concrete sample (N) without fly ash, fly ash concrete sample

(FAN) with fly ash replacement ratio as 10 % by volume and another concrete sample (FAHL) with mixing ratio 7

to 3 for high early strength Portland cement to low heat Portland cement with fly ash replacement ratio as 10 % by

volume, as described in Table 3.2.

In series 2, four mixtures, which included two types of normal concrete and two types of air entrained concrete,

were tested to evaluate the self healing ability of concrete samples containing fly ash, concerning fly ash

replacement ratio and air content and curing conditions and concrete mix proportion were shown in Table 4.3.

Before starting test, all of the investigated concrete samples were mixed in accordance with JIS A 1138 and were

designed to ensure the required slump and air contents. The slump was set 18±2 and non air entrained and air

entrained concrete had 1% and 4% of air content. The air content and slump were measured as main fresh

properties in accordance with JIS A 1128 pressure method and JIS A 1101 slump test method. Concrete samples

were casted into 100 X 100 X 400 mm prisms and cylinder dimension. All investigated concrete samples were

sealed and cured in laboratory for 1 day at 20℃, then, placed in water container at 20±3 until testing ages.

62
Table 3.1 Physical properties of used aggregate materials

Surface dried density Absolute dried Water absorption Coarse grain


Types of aggregates 3 3
(g/cm ) density (g/cm ) ratio (%) ratio (%)

Fine aggregate 2.65 2.64 0.42 2.66


Coarse aggregate 2.67 2.62 1.93 6.64

Table 3.2 Mix proportions of concrete samples and fresh characteristic

Unit weight(kg/m3) Air


W/C s/a Slump
Mixture Types of cement Fly Fine Coarse content
Water (mm)
% % N H L ash aggregate aggregate (%)
N 46.8 190 344 - - - 879 998 180 2.3
FAN 55 45.5 190 344 - - 34 834 998 201 1.1
FAHL 45.5 190 - 241 104 34 834 998 203 0.9

Note: s/a=fine aggregate/coarse aggregate, N=ordinary Portland cement, H=high early strength Portland cement

and L=low heat Portland cement.

Table 3.3 Mix proportions of concrete mixtures and fresh characteristic

Unit weight(kg/m3) Slump Air


W/C s/a
Mixture Fly Fine Coarse content
(%) (%) Water Cement Ad1) Ad2)
ash aggregate aggregate (mm) (%)

N 47 151 296 - 955 1084 3.85 - 165 4.0


NA 46 151 296 - 880 1038 3.26 0.067 175 5.5
51.1
F 48 151 296 44.4 930 1065 3.85 - 180 2.6
FA 46 151 296 44.4 836 1038 3.26 0.2893) 190 5.1

Note: s/a=fine aggregate/coarse aggregate, Ad1) = high performance water reducing agent, Ad2) = air entraing agent

and fly ash air entrain agent added to FA mixture.

63
3.2.2 Experimental methods
3.3.2.1 Compressive strength
Compression test of investigated concrete using cylinder specimen (200×100) were examined at 7, 28, 91 and

365 days of curing, in accordance with JIS A 1108.

3.3.2.2 Frost resistance


After 28 and 365 days of curing age, concrete prism sample (dimension 75×75×400 mm) was prepared in order to

evaluate the frost resistance. According to the ASTM C666 procedure A, the freeze and thaw cycling test, which

has been widely used in order to quantify the freezing and thawing durability and which involves the freezing

minimum temperature at -18℃ for 2 hours 30 minutes (freezing) and thawing temperature at +5℃ for 1 hour 30

minutes (thawing), was conducted at the curing age of 28 days. The mass loss, length change and relative dynamic

modulus of elasticity for all investigated samples were measured every 30 cycles until 300 cycles in accordance

with JIS A 1127 procedure. The test was also completed when a sample reached 300 cycles or relative dynamic

modulus of elasticity reached 60% as critical limit from ASTM C666 procedure A. The experimental result was

obtained from an average of three samples. Relative dynamic modulus of elasticity, durability factor and length

change can be calculated from the results of rapid freeze and thaw testing of all investigated samples, which are

expressed as follows:

fn2
Pn  ( ) 100 (3.1)
f02

where;

Pn is the relative dynamic modulus of elasticity, after c cycles of freezing and thawing, percent; f 0 is the

fundamental transverse frequency at 0 cycles of freezing and thawing, and f n is the fundamental transverse

frequency after n cycles of freezing and thawing.

P N
DF  (3.2)
M

where DF is the durability factor of the test specimen; P is the relative dynamic modulus of elasticity at N

cycles (%); N is the number of cycles at which P reaches the specified minimum value less than 60% or the

relative dynamic modulus of elasticity after 300 cycles, and M is the relative dynamic modulus of elasticity after

64
300 cycles.

l 2  l1
Lc  (3.3)
Lg

where Lc is the length change ratio of the test specimens after c cycles of freezing and thawing (%); l1 is the

length comparator reading at 0 cycles(mm); l 2 is the length comparator reading at after c cycles(mm), and Lg is

the effective gage length between the innermost ends of the gage studs as shown in the mold diagram.

W0  Wn
Wn  (3.4)
W0

where Wn is the weight change ratio of the test specimens after c cycles of freezing and thawing (%), and W0 is

the weight at 0 cycles(g).

3.3.2.3 Pore structure investigation


After 28 and 365 days of curing age, all investigated concrete samples were cut into 5 mm of cube size, and then

replaced by acetone to stop the hydration reaction. After 3 days, freeze-drying was dried for 24 h, prior to

measurement. Then, pore size distribution was determined with mercury intrusion porosimetery

(AutoporeMaster33). The surface tension and density of mercury were 0.480 N/m and 13.546 g/ml and the contact

angle was 140º.

3.3.2.4 Accelerated carbonation test


The hardened concrete beam sample (100×100×400 mm) were cured for 28 days at 20±3°C and RH 60±5%

before the start of testing. Accelerated carbonation exposure was carried out in a controlled environmental chamber

at 20°C, with 5.0% CO2 and 60% RH. After the curing age of 1, 4, 13 and 26 weeks, the carbonation depth was

measured by the phenolphthalein test according to JIS A 1153 procedures. The carbonation coefficient was

calculated from the results of carbonation depth of investigated samples, by the following equation:

xk t (3.5)
0.5
where x is carbonation depth (mm); k is the carbonation coefficient (mm/weeks ) and t is time (week)

65
3.3.2.5 New damage technique by means of compression test
Four cylindrical samples with dimensions 100Ø ×200 mm were prepared for each mix. After casting, all of the

concrete samples were placed in water at 20℃ for 365 days. After the designated curing period, compressive

strength of all of the concrete samples was measured and calculated from load and cross section. Then, the relative

dynamic modulus of elasticity in new concrete sample decreased based on the calculated maximum compressive

load (85% of the maximum compressive load) until 80 to 90% of relative dynamic modulus of elasticity. After

deterioration, the damaged concrete sample was stored in water at 20℃ for 50 days and resonant frequency

repetitively was measured.

3.3.2.6 Damage degree


After 4 weeks of curing age in water, rapid frost and thaw testing was used in order to damage into concrete

samples with different damage and healing conditions in which the damage condition is for 30 cycles by taking into

consideration of natural cold climate and healing curing is in water at 40℃ for 2 weeks. The healing conditions

under high temperature is to investigate the effect of healing potential on self-healing efficiency in fly ash blended

concrete sample. In addition, in order to simulate self-healing ability under natural climate, repeated damage and

healing conditions was accepted to be considered an equivalent maturity reported by Tomokazu et al (2006), who

investigated the accelerated self-healing curing condition due to consideration of natural climate conditions in

summer and they were revealed that the equivalent maturity is 210°. From the findings, similar approach was

adopted to relate the self-healing and damage of cracks within concrete sample under natural environmental

conditions and repeated for three cycles, which assume damage and healing for 3 year.

3.3.2.7 Image analysis


According to Matsumura T. et al (2003), image analysis was also performed in order to observe the micro crack in

concrete sample. The surface of the in all investigated samples was ground and polished to obtain an acceptable

smooth surface for microscopical observation directly through microscope image and induced micro crack in

concrete samples was coated with light ink material. Then, image analysis was examined and the magnification of

Microscope used was 100 to 175.


66
3.3 Result and discussion
3.3.1 Self-healing ability for Non-AE concrete
3.3.1.1 Compressive strength
Fig. 4.1 shows the results of compressive strength for three types of concrete samples, which consist of N

samples without fly ash (N), one samples using with fly ash (FAN) and another sample using combination of high

early strength Portland cement and low heat Portland cement, in water after 1, 28 and 91 days of curing . It can be

confirmed from the figure that the there is no visible different in the compressive strength for all investigated

concrete samples after 7 days of curing age. However, a notable increase was observed in fly ash blended concrete

samples (FAN and FAHL) until 28 to 19 days of curing ages in comparison to N concrete sample, implying that in

addition of fly ash may lead an increase in compressive strength due to pozzolanic behavior in fly ash blended

concrete samples.

60
Compressive strength (MPa)

50

40

30 91 days

20 28 days

7 days
10

0
N FAN FAHL
Mixture Designation

Fig. 4.1 Result of compressive strength for different


mixtures and curing ages.

3.3.1.2 Variation of carbonation coefficient


The relative dynamic modulus of elasticity and carbonation coefficient for each studied case of the designated

concrete samples is given in Fig. 3.2 (a) and (b). It can be seen that the investigation of each concrete samples

include four cases which are No cracking case, after deterioration case, cured in water 40℃ at 2 weeks case and

repetition case until 30 cycles and healing in water 40℃ at 3 days (Tomosawa et al 2006).

After deterioration, it can be seen from the figure (see Fig. 3.2 (a)) that relative dynamic modulus of elasticity in

concrete samples incorporating fly ash, which include 10% by volume fly ash replacement ratio and combination of

67
high early strength Portland cement and low heat Portland cement with 10% by volume fly ash replacement ratio, is

lower compared to N sample, which is the case of OPC without fly ash. The relative dynamic modulus of elasticity

of each concrete sample was 83 % for N, 65% for FA and 65% FAHL after deterioration. However, in the case of

water curing (healing) in water 40℃ for 2 weeks, FA and FAHL fly ash blended concrete samples increase in

comparison to N concrete sample about twice (relative dynamic modulus of elasticity)

Meanwhile, it was confirmed that, in the case of repetition of deterioration and healing condition, N and FAN

concrete samples can improve resistance to freeze and thawing cyclic behavior. The tendency of carbonation

coefficient, which was used based on accelerated carbonation test, was also constant the results of relative dynamic

modulus of elasticity.

No cracking No cracking
after deterioration after deterioration
after healing after healing
repetition of deterioration and healing repetition of deterioration and healing

120 8
Relative dynamic modulus of elasticity

100 120 120


Relative dynamic modulus of elasticity
Relative dynamic modulus of elasticity

Carbonation coefficient

100 100
80 6 80
80 60 60
40 40
(%)

60 20 4 20
0 0
N FA FAHL N FA
(%)
(%)

40
Mixture Designation Mixture Designation
2
20

0
0
N FAN FAHL
N FA FAHL
Mixture Designation
Mixture Designation

(a) Change of relative dynamic modulus of elasticity (b) Accelerated carbonation test results

Fig. 3.2 Self-healing effects on carbonation coefficient and relative dynamic modulus
of elasticity for different concrete samples and different curing conditions

3.3.1.3 Self-healing effect on the relative dynamic modulus of elasticity under


accelerated environmental conditions
The Fig. 3.2 (a) shows the results of relative dynamic modulus of elasticity under repetition of deterioration and

healing. After deterioration (1 cycle), N concrete sample, which is the case of OPC sample without fly ash,

decreased compared with FAN and FAHL samples. Also, it is observed that there is no significant difference in N

concrete sample between deterioration and water curing. In comparison to N concrete sample, after deterioration (1
68
cycle), FAN sample which is the case of fly ash replacement 10 % decreased until 65%, and significant relative

dynamic modulus of elasticity difference did not observed. Meanwhile, it should be noticed that FAHL sample

which is the combination of high early strength Portland cement type and low heat Portland cement with fly ash

decreased significantly after deterioration (3 cycles). There is no clear explanation for this phenomenon. A possible

hypothesis in this observation would be due to the fact that no or little self healing ability of fly ash occurs, since

accelerated environmental conditions applied is extreme condition to heal the damaged concrete sample. In addition,

in order to improve self-healing performance of fly ash blended concrete, an addition of air entraing agent during

concrete mixing is highly advisable, thus next series is to investigate the effect of fly ash blended concrete with air

entraing agent on self-healing ability.

N FAN FAHL
120
Relative dynamic modulus of elasticity

100

80

60
(%)

40

20 D H D H D H

0
0 3 6 9 12 15 18 21 24
Age(days)

Fig. 3.3 Self-healing effects on change of relative dynamic modulus of elasticity for different concrete samples and

repetition of deterioration and healing (D; deterioration, H; healing)

69
3.3.2 Self-healing ability of fly ash blended air entrained concrete
3.3.2.1 Compressive strength development and pore structure investigation
In Fig. 3.4 shows the results of compressive strength test for four types of concrete samples at 3, 7, 28, 91 and

365 days of curing. The compressive strength for N and NA concrete samples showed significant compressive

strength increase after 7 days of curing age, it can be seen that a slight increase at 28 days of curing age, but the

compressive strength of concrete samples produced with fly ash as fine aggregate (F and FA) were significantly

increased after 28 days, in particular, F sample compressive strength increase was remarkable. Therefore, in the

comparison to normal concrete (N and NA), fly ash addition concrete samples as fine aggregate exhibited

significant compressive strength increase, which can be related to pore structure in concrete samples.

According to the pore structure investigation by measurement of MIP, in Fig. 3.5 shows pore size distribution

results for four investigated concrete samples with different curing ages at 28 and 365 days. In comparison to

normal concrete samples N and NA in Figs. 3.5 (a) and (b), there is major difference from the pore size distribution

results. It should be note that fly ash concrete samples F and FA were shifted to lower pore size region (left side)

when curing age increase, while normal concrete samples (N and NA) were not noticeable, hence, the fly ash

replacement as fine aggregate could be improved compressive strength due to the pozzolanic effect and attributed

to densification of the matrix for the observation between compressive and pore structure investigation, which is

similar with those observation reported by previous research (R. Siddique 2003).

70

60
Compressive strength(MPa)

50

40

30

20
N NA
10 F FA

0
0 100 200 300 400
Curing age (days)

Fig. 3.4 Compressive strength development of for types of concrete sample at 3, 7, 28, 91 and 365 days of curing

70
0.20

Log differential intrusion (cc/g)


0.20

Log differential intrusion (cc/g)


(a) N 4W (b) NA 4W
0.15 N 1Y 0.15 NA 1Y

0.10 0.10

0.05 0.05

0.00 0.00
1 10 100 1000 1 10 100 1000
Pore size diameter(nm) Pore size diameter(nm)

0.20 0.20
Log differential intrusion (cc/g)

Log differential intrusion (cc/g)


(c) F 4W (d) FA 4W
F 1Y FA 1Y
0.15 0.15

0.10 0.10

0.05 0.05

0.00 0.00
1 10 100 1000 1 10 100 1000
Pore size diameter(nm) Pore size diameter(nm)

Fig. 3.5 Change of pore volume of different concrete samples for different curing conditions

3.3.2.2 Frost resistance characteristics


In Fig. 3.6 presented the results of freeze thaw testing for different curing ages at 28 and 365 days according to

the ASTM C 666 procedure A. It has been well-known that the relative dynamic modulus of elasticity decrease as

internal micro cracks due to ice expansion within concrete increases, and therefore, the increase of length change

approximately 9% can be occurred due to their ice expansion (Power 1945).

The relative dynamic modulus of elasticity is an evaluation index to observe in external damaged concrete. At

age of 28 days, as given in Fig. 3a, in the case F concrete sample in terms of the relative dynamic modulus of

elasticity reaches 60% at 42 cycles. However, in the case of FA concrete sample was similar to NA concrete sample.

At curing age of 365 days, the values of relative dynamic modulus of elasticity of all of the concrete samples were

rapidly decreased below 30 cycles, in addition, the variation of relative dynamic modulus of elasticity for N, NA

and FA concrete samples has similar trend. Therefore, fly ash addition as fine aggregate in concrete samples was

found to be similar to two types of normal concrete sample with respect to the variation of relative dynamic

modulus of elasticity in this series.

71
NA

Relative dyna
20 F
FA
0
0 50 100 150 200 250 300
Number of cycles

120 120

Relative dynamic modulus of elasticity(%)


Relative dynamic modulus of elasticity(%)
after curing 28 days after curing 365 days
100 100

80 80

60 60

N N
40 40
NA NA
20 F 20 F
FA FA
0 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300
Number of cycles Number of cycles

Fig. 3.6120Change of relative dynamic modulus of elasticity for different concrete samples with curing ages.
Relative dynamic modulus of elasticity(%)

after curing 365 days


100

80
3.3.2.3 Fly ash AE concrete self healing effect on the variation of relative dynamic
60
modulus of elasticity
N
40
NA
Fig. 4.7 represents the results of relative dynamic modulus of elasticity for fly ash concrete samples which
20 F
FA
include 10 % 0by mass fly ash replacement ratio with (F) or without air entrained agent (FA) and normal concrete
0 50 100 150 200 250 300

samples with (N) or withoutNumber of cycles


air entrained agent (NA) with different curing ages at 28 days (top) and 365 days

(bottom), considering the three studies, cases which consist of No cracking case (initial value), deterioration case

(micro-cracking) and healing case in water curing at 40℃ for curing age of 28days. In the case of curing ages at 28

days, after deterioration, it can be seen from the figure that most of the investigated concrete samples decreased

until 80 to 90 of relative dynamic modulus of elasticity, after healing case, they were healed almost until No

cracking case (initial value) in all concrete samples. Meanwhile, it can be observer after curing ages at 28 days that

most of the investigated concrete samples were partially recovered after healing case. This observation for different

curing age of 28 and 365 days is due to the fact that self healing effect can be decreased with increasing curing ages

due to decrease of un-reacted cement and fly ash in investigated concrete samples.

72
40

Relative dyna
No cracking
20 After deterioration
After healing
0
N NA F FA

120 120
(b) after 365 days of curing age

Relative dynamic modulus of elasticity(%)


Relative dynamic modulus of elasticity(%)
(a) after 28 days of curing age
105
98 100 99
100 100
91 92
87
85 86
80 75 80 81
80 80
72 71 73

60 60

40 40

No cracking No cracking
20 After deterioration 20
After deterioration
After healing After healing
0 0
N NA F FA N NA F FA

Fig. 3.7 Self-healing


120 effect on change of relative dynamic modulus of elasticity for different concrete samples with
(b) after 365 days of curing age
Relative dynamic modulus of elasticity(%)

100
different curing ages (a) after water
92 curing at 28 days (left figure) and (b) 365 days (right figure).
87
75 80 81
80
72 71 73

60

3.3.2.4 Self-healing effect on change of carbonation coefficient


40

Fig. 3.8 shows the results of accelerated carbonation testing for four concrete samples after water curing at 28
No cracking
20
After deterioration
After healing
and 91 days (before
0 deterioration), including four cases which consist of No cracking case, after deterioration case,
N NA F FA

after healing case and repetition of deterioration and healing case. The carbonation coefficient in all investigated

concrete samples was calculated by measured carbonation depth value until 26 weeks. It can be confirmed from the

figure in both of two curing ages at 28 and 91 days that significant decrease of F and FA concrete samples, which

involve 10 % by volume of fly ash replacement ratio, observed after deterioration case in comparison to N and NA

concrete sample, which are normal concrete samples without addition of fly ash. From the findings, incorporating

fly ash into fine aggregate exhibited high self-healing ability after water curing at 40℃ for 28 days, resulting in the

prevention of carbonation process after deterioration.

No cracking after deterioration after healing repetition of deterioration and healing

5 5
(a) after 28 days of curing age (b) after 91 days of curing age
4 4
Carbonation coefficient
Carbonation coefficient

3 3

2 2
5
4グラフ タイトル
3
2 1 1
1
0
0N NA F 0 FA
N F NA FA N F NA FA
Mixture Designation Mixture Designation

73
Fig. 3.8 Self-healing effect on change of carbonation coefficient by means of accelerated carbonation testing for

different concrete samples with different curing age (a) after water curing at 28 days (left side) and (b) 91 days

(right side).

3.3.2.5 Self-healing effect on relative dynamic modulus of elasticity under accelerated


environmental conditions
Fig. 3.9 shows the change of relative dynamic modulus of elasticity for different concrete samples and repetition

of deterioration and healing. Two types of curing age at 28 and 91 days were performed to investigate the effect of

curing age on self-healing ability of concrete samples. It can be seen from the figure, in both curing ages, most of

concrete samples gradually decreased after repetition of deterioration and


N healing.
NA Especially,
F F sample
FA which
120
Relative dynamic modulus of elasticity

involve 10% by mass fly ash replacement ratio lastly decreased(a)even


after 28 dayswater
after of curingcuring
age at 40℃ for 28 days.
100

According to the relative dynamic modulus of elasticity data80measured from figure 3.7 (a), it can be seen that F

sample had high self-healing potential after water curing.60However, although high self-healing performance
(%)

40
remains to heal, for non-AE concrete (F sample), it could not recover the cracks due to the less air content. Hence,
20
D H D H D H
suitable AE agent addition could be significant in order to develop the self-healing concrete and ensure the
0
0 3 6 9 12 15 18 21 24
deterioration which is caused by freeze and thaw cyclic behavior. Age(days)

N NA F FA N NA F FA
120 120
Relative dynamic modulus of elasticity
Relative dynamic modulus of elasticity

(a) after 28 days of curing age (b) after 91 days of curing age
100 100

80 80
(%)

60 60
(%)

40 40

20 20
D H D H D H D H D H D H
0 0
0 3 6 9 12 15 18 21 24 0 3 6 9 12 15 18 21 24
Age(days) Age(days)

Fig. 3.9 Self-healing effects on change of relative dynamic modulus of elasticity for different concrete samples and
N NA F FA
120
repetition of deterioration and healing (the symbols indicate D; deterioration and H; healing)
ative dynamic modulus of elasticity

(b) after 91 days of curing age


100

80
74
(%)

60

40

20
D H D H D H
3.3.3 New damage technique and visualization of micro crack
As mentioned above, concrete sample with suitable air content was found to be difficult damage by respect to

rapid freeze and thaw testing due to the higher frost resistance. Therefore, new damage technique is quite needed to

induce micro cracks, which are similar to freeze and thaw cycling testing, thus, new damage technique for air

entrained concrete sample was investigated in this chapter. Compressive strength of all investigated concrete

sample after curing age of 365 days was firstly tested, then, the concrete sample was induced stress about ten times

until 80% of compressive strength. Finally, damage degree due to induction of crack within concrete sample was

measured in terms of resonant frequency, and relative dynamic modulus of elasticity was estimated from the

response of resonant frequency after deterioration.

Fig. 3.10 shows change of relative dynamic modulus of elasticity for four types of concrete samples, which

involve two non-AE concrete samples (N and F) and AE concrete samples (NA and FA) by means of repeated

cyclic loading. It can be seen from the figure that the target relative dynamic modulus of elasticity (80 to 90%) in

all concrete samples reaches after 5 to 8 cycles. The relative dynamic modulus of elasticity of FA concrete sample

decreased gradually in comparison to the three concrete samples (N, NA and F), which is presumably attributed to

the densification due to addition of fly ash. Also, in the case of water curing in 20℃ after deterioration, the relative

dynamic modulus of elasticity with different tendency is visible. It can be observed that the self-healing concrete

samples (F and FA), which incorporate 15% by volume of fly ash, quickly recovered when compared to two types

of normal concrete (N and NA). Fig. 3.11 represents the appearance of micro cracking for AE concrete sample

caused by different two types of technique that involve the freeze / thaw cycling and repeated cycling loading. This

implies that the micro crack caused by repeated cycling loading is very similar to the freeze / thaw cycling method

and present of visible micro crack on the damaged concrete is remarkably observed.

From the observation above, the repeated cyclic loading technique is an effective way for making the crack

within the concrete than rapid freeze and thaw test, and this method can apply air entrained concrete sample having

suitable amount of air content.

75
Relative dyanmic modulus of elasticity

Cycles

Curing age (days)

Cycles

Curing age (days)

Fig. 3.10 Change of relative dynamic modulus of elasticity for different concrete samples after repeated cyclic

loading

Freeze and thawing cycling Repeated cyclic loading

Fig. 3.11 Visual features for AE concrete sample caused by different crack introductions (left is the freeze and

thawing cycling method and right is the developed repeated cyclic loading method)

76
3.4 Conclusions
Based on the experimental results, the following conclusions can be derived;

(1) The experimental results show that significant increase in fly ash blended concrete sample can be observed,

when curing ages increase. This result can be explained by the fact that F and FA sample tends to shift to

lower pore size region due to the densification of pozzolanic ability at long-term curing age for the

compression test and pore structure investigation, is shown in series 1 and 2.

(2) It was observed that self-healing effect is related to curing age before deterioration, when curing age increases,

for long-term curing age case, the healing ability decreases due to decrease of un-reacted cement and fly ash in

investigated concrete samples based on measurement of relative dynamic modulus of elasticity.

(3) It was found that although fly ash had self-healing potential, air entraining agent addition in the fly ash

concrete sample is advisable because of the poor frost resistance due to the less air content in fly ash blended

concrete sample.

(4) It was confirmed that relative dynamic modulus of elasticity decrease with increasing the number of repeated

cycling loading based on compression test. In comparison to normal concrete (N and NA), relative dynamic

modulus of elasticity of fly ash blended concrete sample, which involves 15% replacement ratio by mass fly

ash, decrease slowly. This can be explained by densification of fly ash addition. In addition to its findings, the

repeated cycling loading is a useful method in order to give damage in AE-concrete sample and a promising

way to study for self-healing ability using cementitious composite.

References

Siddique R.. Effect of fine aggregate replacement with Class F fly ash on the mechanical properties of concrete,
Cement and Concrete Research 33, pp. 539 – 547 (2003)

Sasaki Tomokazu, Hama Yukio, Katsura Osamu and Taniguchi Madoka, Evaluation of Natural Condition
influence on Deterioration and Self-Healing of Concrete, Architectural Institute of Japan, Hokkaido, (79), 29-34
(2006)
77
Matsumura T, Katsura O and Yoshino T., Properties of frost damaged concrete and the extimation of the degree of
frost damage, Journal of structural and construction engineering (563), 9-13 (2003)

Powers, T. C., A Working Hypothesis for Further Studies of Frost Resistance of Concrete, Journal of American
Concrete Institute, Vol.16, No.4, pp.245-272 (1945)

78
CHAPTER 4 SELF-HEALING ABILITY OF GROUND GRANULATED

BLAST BURNACE SLAG BLDED SYSTEM AFTER FROST DAMAGE

79
4.1 Overview
Durability of concrete is very important in order to prolong their service life time in concrete industry, therefore

many engineers has been performed and designed to make the more durable concrete. It is well-known that

properties such as carbonation, chloride ingress and frost resistance may affect the durability of concrete. Especially,

among the durability properties, the frost resistance of concrete structure is of great importance in cold region.

However, due to the reduction of the frost durability as freezing and thawing cycles, micro cracks within concrete

structure occur and therefore carbonation and chloride ingress is very rapid. Hence, for the extension of concrete

structure, the technique of cracks caused by frost damage should be needed and addressed.

Recently, many researchers have been attempted to investigate about self healing effect. They were examined the

autonomous healing performance (self-healing) for blended cementitious systems by incorporating various mineral

admixtures, for example, expansive agent, fly ash, ground granulated blast furnace slag and nano-clay. (S. Qian et

al 2009, Sisomphon K. 2010, Pipat T. et al 2009, Ahn TH and KIshi T 2010, Mustafa S. and Victor C. Li 2008).

Pipat et al (2009) has shown by fly ash hydration characteristic and some mechanical tests, that the addition of fly

ash has positive effect for the self healing ability and would enable to heal cracks due to the use of pozzolanic

behavior, which reacts with calcium hydroxide and water, then, generated C-S-H gel.

In addition, among those mineral admixtures, for self-healing ability related to blast furnace slag, S. Qian et al

(2009) reported that high power/binder and low water to binder ratio have a positive effect on the self-healing

ability due to the fact that un-hydrated cement and un-reacted cementitious materials increase.

Tittelboom et al. (2012) investigated the effect of mix composition on the extent of autogenous cracks healing of

mortar sample incorporating latent hydraulic or pozzolanic binder materials such as fly ash, ground granulated blast

furnace slag. They were reported that addition of blast furnace slag can affect the self-healing ability due to latent

hydraulic behavior. In addition, induced cracks in mortar sample can be healed due to CaCO3 precipitation.

According to the above literatures, however, there is lack of information about fundamental research regarding

BFS replacement level and method on self healing effect. Therefore, the objective of this chapter is to investigate

the self healing effect of BFS blended mortar system after frost damage.

80
4.2 Experimental program
4.2.1 Experimental materials
Portland cement (density 3.17 g/cm3) and three types BFS (blain fineness: 3000, 4000 and 8000 cm2/g) were

used. Chemical compositions of cement and BFS are displayed in Table 1 and 2. As for mix proportion, 13 types

BFS blended cement mortar, which with BFS replacement ratio by mass 20% and 40% as partial replacement of

sand and with BFS replacement ratio by mass 45% or 75% as partial replacement of Portland cement, were

prepared. Mortar mix proportion is given in Table 2. Air amount in all specimens were set below 2.0% because this

study was for Non-AE mortar. During the mixing process, a standard mortar mixer was used to make specimens

according to designed mortar mixtures. In order to evaluate fresh characteristic after mixing, flow and air content

tests were conducted according to JIS R 5201 and JIS A 1128. All of the investigated mortar specimens

(160×40×40 mm prism) were placed in laboratory for 24 hours. After hardening, each of mortar specimens was

cured in water 20℃ until testing ages.

Table 3.1 Chemical composition of cement

Mineral composition (%) Cement

C3S 6.57
C2S 33.86
C3A 15.19
C4AF 0.90
3 3.17
Density(g/cm )
2 3390
Blain fineness(cm /g)

Table 3.2 Chemical composition of different types of BFS


Major elements
BFS3000 BFS4000 BFS8000
as oxide (wt.%)
MgO 6.57 6.21 5.35
SiO2 33.86 34.12 33.67
Al2O3 15.19 16.07 16.24
Fe2O3 0.9 0.38 0.59
CaO 40.89 42.06 42.14
3 2.91 2.91 2.91
Density(g/cm )
2
Blain fineness(cm /g) 3180 4030 7300

81
Table 3.3 Mix proportions of mortar and fresh characteristic

Water/binder Water Cement BFS Sand Flowability Air content


Mixture 3 3 3 3
ratio (kg/m ) (kg/m ) (kg/m ) (kg/m ) (mm) (%)
N 0.55 342 621 0 1242 261 X 256 1.6
3S20 0.46 342 621 124 1118 221 X 220 0.9
3S45 0.38 342 621 280 963 218 X 221 5.0
3S45C 0.55 342 342 280 1242 257 X 261 1.8
3S70C 0.55 342 186 435 1242 274 X 276 2.1
4S20 0.46 342 621 124 1118 186 X 189 0.9
4S45 0.38 342 621 280 963 186 X 189 1.1
4S45C 0.55 342 342 280 1242 238 X 238 1.4
4S70C 0.55 342 186 435 1242 263 X 256 1.2
8S20 0.46 342 621 124 1118 208 X 214 1.7
8S45 0.38 342 621 280 963 155 X 154 0.9
8S45C 0.55 342 342 280 1242 221 X 218 0.6
8S70C 0.55 342 186 435 1242 223 X 226 0.3

4.2.2 Experimental methods


4.2.2.1 Frost resistance
After curing age of 4 days, hardened mortar sample (dimension 40×40×160 mm) was used in order to

evaluate the frost resistance. According to the ASTM C666 procedure A, the freeze and thaw cycling test, which

has been widely used in order to quantify the freezing and thawing durability and which involves the freezing

minimum temperature at -18℃ for 2 hours 30 minutes (freezing) and thawing temperature at +5℃ for 1 hour 30

minutes (thawing), was conducted at the curing age of 28 days. The mass loss, length change and relative dynamic

modulus of elasticity for all investigated samples were measured every 30 cycles until 300 cycles in accordance

with JIS A 1127 procedure. The test was also completed when a sample reached 300 cycles or relative dynamic

modulus of elasticity reached 60% as critical limit from ASTM C666 procedure A. The experimental result was

obtained from an average of three samples. Relative dynamic modulus of elasticity, durability factor and length

change can be calculated from the results of rapid freeze and thaw testing of all investigated samples, which are

expressed as follows:

fn2
Pn  ( ) 100 (4.1)
f02

where;

Pn is the relative dynamic modulus of elasticity, after c cycles of freezing and thawing, percent; f 0 is the

fundamental transverse frequency at 0 cycles of freezing and thawing, and f n is the fundamental transverse

frequency after n cycles of freezing and thawing.

82
P N
DF  (4.2)
M

where DF is the durability factor of the test specimen; P is the relative dynamic modulus of elasticity at N

cycles (%); N is the number of cycles at which P reaches the specified minimum value less than 60% or the

relative dynamic modulus of elasticity after 300 cycles, and M is the relative dynamic modulus of elasticity after

300 cycles.

l 2  l1
Lc  (4.3)
Lg

where Lc is the length change ratio of the test specimens after c cycles of freezing and thawing (%); l1 is the

length comparator reading at 0 cycles(mm); l 2 is the length comparator reading at after c cycles(mm), and Lg is

the effective gage length between the innermost ends of the gage studs as shown in the mold diagram.

W0  Wn
Wn  (4.4)
W0

where Wn is the weight change ratio of the test specimens after c cycles of freezing and thawing (%), and W0 is

the weight at 0 cycles(g).

4.2.2.2 Accelerated carbonation test


The hardened mortar sample (100×100×400 mm) were cured for 28 days at 20±3°C and RH 60±5% before the

start of testing. Accelerated carbonation exposure was carried out in a controlled environmental chamber at 20°C,

with 5.0% CO2 and 60% RH. After the curing age of 1, 4, 13 and 26 weeks, the carbonation depth was measured by

the phenolphthalein test according to JIS A 1153 procedures. The carbonation coefficient was calculated from the

results of carbonation depth of all investigated samples, by the following equation:

xk t (4.4)
0.5
where x is carbonation depth (mm); k is the carbonation coefficient (mm/weeks ) and t is time (week)

4.2.2.3 Self healing evaluation


After 28 days, in order to investigate the self healing ability of BFS, freeze and thaw cycling testing, according to

the JIS A 1148 method, was used to give frost damage into mortar specimens until 60% of relative dynamic
83
modulus of elasticity. After target value in all mortar samples reached, the damaged mortar specimens that caused

freeze and thaw cycling testing were cured with two types of healing curing conditions in water at 20℃ for 1 week,

that was standard curing conditions, and in water 40℃ for 4 weeks. The curing condition at 40℃ was to

investigate self healing potential of BFS. For assessing the performance of self healing in all mortar samples, after

each condition, compressive strength and bending strength for cracked and un-cracked mortar samples were

conducted according to JIS A 5201. Further, accelerated carbonation test (phenolphthalein method) was performed

to calculate the carbonation depth and coefficient before and after deterioration and healing curing conditions, in

accordance with JIS A 1153, and relative dynamic modulus of elasticity was also measured according to JIS A

1127.

84
4.3 Results and discussions
4.3.1 Frost resistance
Relative dynamic modulus of elasticity result with different fineness of BFS (blain fineness: 3000, 4000 and

8000 cm2/g) is shown In Fig. 4.1 and 4.2. In general, relative dynamic modulus of elasticity tends to decrease when

number of freeze-thaw cycles increases. In cases of mortar samples made by 45% and 75% BFS replacement for

cement, as can be seen in Figs. 4.1 (a) and (b), the relative dynamic modulus of elasticity for all mortar samples

was decreased before 55 cycles, in addition, mortar samples incorporating BFS were improved slightly for frost

resistance. Mortar samples incorporating BFS, however, were gradually decreased compared normal mortar sample.

In Fig. 4.2 (b), relative dynamic modulus of elasticity of mortar samples replaced by BFS (45 wt. %) does not

decrease as number of freeze-thaw cycles increase. From these observations, it is indicated that an increase in sand

replacement ratio of BFS showed significant better frost resistance performance compared to normal mortar sample

and optimum replacement of BFS appeared to be 45 wt. % (see Fig. 3.3).

120 120
Relative dynamic modulus of elasticity (%)

Relative dynamic modulus of elasticity (%)

(a) 45% BFS replacement for cement (b) 70% BFS replacement for cement
100 100

80 80

60 60

N N
40 40 3S70C
3S45C
4S45C 4S70C
20 20
8S45C 8S70C

0 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300

Number of freeze- thaw cycles (cy.) Number of freeze- thaw cycles (cy.)

Fig. 4.1 Change of relative dynamic modulus of elasticity for different mortar mixtures with or without BFS during

freeze and thaw cycling.


Relative dynamic modulus of elasticity (%)

120 120
Relative dynamic modulus of elasticity (%)

(a) 20% BFS replacement for sand (b) 45 % BFS replacement for sand
100 100

80 80

60 60

N N
40 40
3S20 3S45
4S20 20 4S45
20
8S20 8S45
0 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300

Number of freeze- thaw cycles (cy.) Number of freeze- thaw cycles (cy.)

Fig. 4.2 The change of relative dynamic modulus of elasticity for different mortar mixtures
85
120
Blain fineness
100 3000 cm2/g

Drability factor
80 4000 cm2/g
8000 cm2/g
60

40

20

0
45% 70% 20% 45%
BFS replacement for cement BFS replacement for sand

Fig. 4.3 The change of durability factor of all mortar samples

4.3.2 Compressive and bending strength


Fig. 4.4 shows the results of the compressive strength of all mortar samples after 28days of curing age.

Compressive strength and bending strength of N sample (normal mortar) were 40.5 and 6.9 MPa. For BFS

replacement for sand, increasing the BFS replacement ratio (20% and 45%) and blain fineness (3000, 4000 and

8000 cm2/g) had a high compressive strength (see Fig. 4.4. (a)). In addition, for BFS replacement for cement,

compressive strength of 8S45C and 8S70C mortar samples were increased compared to the normal mortar sample

(see Fig. 4. (b)). Figure 4.5 shows the results of the bending strength of all mortar samples. For BFS replacement

for sand, all the mortar samples produced by BFS showed higher bending strength compared to normal mortar

sample, however, in case of BFS replacement for cement, except 4S70C and 8S70C mortar samples, bending

strength in all mortar samples decreased slightly. It can be observed that addition of BFS replacement for sand had

beneficial for compressive and bending strength. This is probably due to the fact that dense pore structure occurs in

investigated mortar samples.

86
80 80

(a) (b)

Compressive strength (MPa)


Compressive strength (MPa)
60 60

40 40

20 20

0 0
3S20 3S45 4S20 4S45 8S20 8S45 3S45C 3S70C 4S45C 4S70C 8S45C 8S70C

3000 4000 8000 3000 4000 8000

Fig.4.4 Compressive strength development for different mortar mixtures; (a) is BFS replacement for sand and (b) is

BFS replacement for cement.


10 10
(a) (b)
Bending strength (MPa)

8
Bending strength (MPa)
8

6 6

4 4

2 2

0 0
3S20 3S45 4S20 4S45 8S20 8S45 3S45C 3S70C 4S45C 4S70C 8S45C 8S70C
3000 4000 8000 3000 4000 8000

Fig. 4.5 Bending strength development for different mortar mixtures; (a) is BFS replacement for sand and (b) is

BFS replacement for cement.

4.3.3 Self healing ability


4.3.3.1 Self-healing effect on compressive strength after deterioration
Fig. 4.6 shows the compressive strength results. After deterioration, the compressive strength of most mortar

samples was decreased by frost damage. In case of some mortar samples (3S45, 4S20 and 8S20), significant

reduction of compressive strength was observed after deterioration. The experimental compressive strength data is

re-plotted in Fig. 4.7. From the figure, under the same replacement ratio levels, high fineness of BFS had low self

healing effect (see Fig.4.6a). In addition, 3S45C and 4S45C mortar samples were healed after healing curing

conditions. It is believed that the hydration rate of BFS is fast as high fineness of BFS increases, so that it can cause

high compressive strength for initial period, leading to the low self healing ability. Meanwhile, there is no obvious

difference in healing curing conditions in this study.

87
No cracking After deterioration
Cured in water 20℃ Cured in water 40℃
70

60
Compressive strength (MPa) 50

40

30

20

10

0
3S20 3S45 4S20 8S20 3S45C 3S70C 4S45C 4S70C 8S45C 8S70C
N Sand replacement Cement replacement

Fig. 4.6 Self healing ability on compressive strength test results for different curing conditions after deterioration

No cracking After deterioration No cracking After deterioration


Cured in water 20℃ Cured in water 40℃ Cured in water 20℃ Cured in water 40℃
70 70
(a) (b)
60 60
Compressive strength (MPa)

Compressive strength (MPa)

50 50

40 40

30 30

20 20

10 10

0 0
3S45C 4S45C 8S45C 3S20 4S20 8S20

Fig.4.7. Compressive strength test results under the same BFS replacement levels: (a) is BFS replacement for

cement and (b) is BFS replacement for cement for sand.

4.3.3.2 Self-healing effect on bending strength after deterioration


From the results of compressive strength test in Figs. 4.8 and 4.9, we could not find out the major differences

about healing curing conditions, from the results of bending strength test, we can observe the differences about

healing curing conditions. Under the same BFS replacement levels, it can be suggested that micro cracks caused by

freeze and thaw cycling test did not heal in water 20℃ for 7 days. However, high healing efficiency of mortar

samples incorporating BFS were observed in high temperature at water 40℃ (see Fig. 4.9a) in comparison to water

20℃. Based on the observation above, it was confirmed that, when BFS was replaced, curing condition would

affect the self healing ability for bending strength test, especially for 3S45C, 4S45C and 8S45C mortar samples.

88
No cracking After deterioration
Cured in water 20℃ Cured in water 40℃
9

Bending strength (MPa) 8


7
6
5
4
3
2
1
0
3S20 3S45 4S20 8S20 3S45C 3S70C 4S45C 4S70C 8S45C 8S70C
N Sand replacement Cement replacement

Fig. 4.8 Bending strength test results.

No cracking After deterioration No cracking After deterioration


Cured in water 20℃ Cured in water 40℃ Cured in water 20℃ Cured in water 40℃
10 10
(a) (b)
Bending strength (MPa)
Bending strength (MPa)

8 8

6 6

4 4

2 2

0 0
3S45C 4S45C 8S45C 3S20 4S20 8S20

Fig. 4.9 Bending strength test results under the same BFS replacement levels: (a) is BFS replacement for cement

and (b) is BFS replacement for cement for sand.

4.3.3.3 Self-healing effect on accelerated carbonation test results


Fig. 4.10 shows the results from accelerated carbonation test of N and 3S70C mortar samples until 13 weeks.

The test value was obtained from an average of two mortar samples. The variation of carbonation depth after

different stages between no cracking, after deterioration and two types of healing curing conditions is observed. At

initial period (no cracking), carbonation depth of normal mortar sample (N) was twice lower than that of BFS

blended mortar sample (3S70C), which is due to the fact that calcium hydroxide was consumed by hydration

reaction of BFS.

89
Meanwhile, after healing curing conditions, the change of carbonation depth for normal mortar sample (N) did

not observe and the 3S70C mortar sample, however, can decrease the carbonation depth. It would be possible that

addition of BFS appeared to be attributed to the improved carbonation performance after healing curing conditions.

In order to evaluate self healing ability, the performance can be calculated from experimental data in terms of

carbonation coefficient changes (after healing conditions / deterioration). Generally, it would be well-known that,

when deterioration would occur, the carbonation coefficient tends to increase with decreasing pH in cementitious

composites. Afterwards, it would be possible that micro cracks caused by frost damage would heal due to the

production of calcium carbonate which is the main component of self healing ability.

Fig. 4.13 shows the accelerated carbonation test results under four stages (no cracking, after deterioration and

after two types of sealing conditions). After healing conditions, all of the mortar samples decreased due to the

external water supply, when compared to deterioration stage. Due to increasing the BFS Blaine fineness with 3000,

4000 and 8000 cm2/g, the low self healing ability could be observed after two types of healing conditions,

especially for 3S45C, 4S45C and 8S45C mortar samples. In addition, 3S45C mortar sample which was

incorporated by cement with 45 wt. % was remarkable decreased in water 40℃. These observations revealed that

incorporating low fineness of BFS for cement has high self healing ability due to the slow hydration and un-reacted

materials in terms of calculated self healing ability value.

Carbonation coefficient Carbonation coefficient


20 20
No cracking y = 1.7089x No cracking y = 3.1708x
18 18
Carbonation depth(mm)

y = 1.9847x
Carbonation depth(mm)

After deterioration After deterioration y = 3.4821x


16 Cured in water 20℃
16 Cured in water 20℃ y = 2.3469x
y = 1.6806x
14 Cured in water 40℃ y = 1.1795x 14 Cured in water 40℃ y = 2.8556x
12 12
10 10
8 8
6 6
4 4
2 N 2 3S70C
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Accelerated carbonation time(week 0.5 ) Accelerated carbonation time(week 0.5 )

Fig. 4.10 Carbonation depths with increasing accelerated carbonation time in N and 3S70 mortar samples.

90
10 No cracking 140 After deterioration
After deterioration Cured in water 20℃
Cured in water 20℃ 120
8 Cured in water 40℃

Carbonation coefficient
Cured in water 40℃

(healing/deterioration)
Self-healing ability, %
100

6 80

60
4
40
2
20

0 0
N 3S45C 3S70C N 3S45C 3S70C

10 No cracking 140 After deterioration


After deterioration Cured in water 20℃
Cured in water 20℃ 120
8 Cured in water 40℃
Carbonation coefficient

Cured in water 40℃

(healing/deterioration)
Self-healing ability, %
100

6 80

60
4
40
2
20

0 0
4S45C 4S70C 4S45C 4S70C

10 No cracking 140 After deterioration


After deterioration Cured in water 20℃
Cured in water 20℃ 120
8 Cured in water 40℃
Carbonation coefficient

Cured in water 40℃


(healing/deterioration)
Self-healing ability, %

100

6 80

60
4
40
2
20

0 0
8S45C 8S70C 8S45C 8S70C

Fig. 4.13 Self healing ability on carbonation coefficient test results

91
4.4 Conclusions
This chapter purpose was to investigate the effect of BFS replacement ratio and method on the self healing effect

BFS blended mortar system after frost damage caused by rapid freeze and thaw testing. Freeze and thawing test

was used in order to damage in all investigated mortar samples, and the level of damage was ranged 58 to 65 of

relative dynamic modulus of elasticity for all mortar samples. Various mechanical properties on compression,

bending and accelerated carbonation testing were performed for assessing the self healing ability. Further, the self

healing ability was quantitatively expressed in terms of carbonation coefficient.

(1) From the freeze and thaw cycling test, it is revealed that sand replacement for BFS has higher frost resistance

performance than cement replacement and optimum replacement ratio (sand replacement) could be 45 wt. %.

Durability factor would increase with increasing the fineness for most BFS blended mortar samples, except for

20% replacement ratio for sand mortar sample.

(2) Based on results of compressive and bending strength test, it is confirmed addition of BFS had positive effect

on compressive and bending strength, especially for sand replacement and increasing fineness.

(3) After deterioration (micro cracking) that introduced by freeze and thaw cycling test, it is conducted that low

BFS fineness had higher self healing ability for compressive, bending strength and accelerated carbonation test

due to the slow hydration and un-reacted materials and carbonation coefficient that calculated from

experimental carbonation depth appeared to be good indicator to evaluate the self healing ability in mortar

samples.

92
References.

T. Ishida, Y. Luan, T. Sagawa and T. Nawa, Modeling of early age behavior of blast furnace slag concrete based on
micro-physical properties, Cement and Concrete Research, Volume 41, pp. 1357-1367 (2011)
Ahn TH and Kishi T., Crack self-healing behavior of cementitious composites incorporating various mineral
admixtures, J Adv Concr Technol , Vol.8, No.2, pp.171–86 (2010)

Mustafa S. and Victor C. Li, Durability of mechanically loaded engineered cementitious composites under highly
alkaline environments, Cement & Concrete Composites 30 pp. 72–81 (2008)

Pipat T., Toyoharu N., Yoichi Y. and Toshiki S., Self-healing ability of fly ash–cement systems, Cement & Concrete
Composites 31 pp. 195–203 (2009)

S. Qian, J. Zhou, M.R. de Rooij, E. Schlangen, G. Ye and K. van Breugel, Self-healing behavior of strain hardening
cementitious composites incorporating local waste materials, Cement and Concrete Composites, Volume 31, Issue
9, October, pp. 613-621 (2009)

Sisomphon K, Copuroglu O., Some characteristics of a self-healing mortar incorporating calcium sulfo-aluminate
based agent. In: Proceedings of the 2nd international conference on durability of concrete structures, Sapporo,
Japan; pp. 157–64 (2010)

Kim Van Tittelboom, Elke Gruyaert, Hubert Rahier and Nele De Belie, Influence of mix composition on the extent
of autogenous crack healing by continued hydration or calcium carbonate formation, Construction and Building

Materials 37, 349–359 (2012)

93
CHAPTER 5 COMPARISON OF SELF-HEALING ABILITY OF

CONCRETE INCORPORATING FLY ASH AND GROUND GRALUATED

BLAST FURNANCE SLAG

94
5.1 Overview
As demonstrated in Chapter 2 to 4, the types of mineral admixture, healing curing condition and damage degree

have an important effect on self-healing ability in blended mortar and concrete mixture with different amount of air

content with respect to various mechanical properties such as compressive strength, bending strength, accelerated

carbonation and rapid freeze and thaw testing.

Recently, several researchers reported the self-healing index between damage and healing behavior by

measurements of mechanical properties with compressive strength, bending strength and crack investigation about

difference between damage and self-healing responses. Among them, Jaroenratanapirom and Sahamitmongkol

(2011) were quantitatively evaluated self-healing closing ability with respect to difference between initial line and

the line at each crack-closing period after damage and healing. Tittelboom et al (2012) were also conducted a

similar approach with the crack self-healing closing method. In addition, there is a different approach. V. Wiktor,

H.M. Jonkers (2011) examined crack-healing quantification for bacteria-based mortar with various lengths and

widths due to difference between the initial crack width and width measured at time. Different approach was

observed. Emily N. Herbert and Victor C. Li (2013) monitored the change of resonant frequency recovery when

exposed to the environmental conditions.

According to previous researches, since there are many examples of self-healing index in cementitious based

materials, however, there is little information in the literature on the self-healing ability of these materials by

taking into consideration of carbonation process. In addition, Although the effect of fly ash and blast furnace

slag on self-healing ability in blended mortar and concrete sample was examined in chapters 2 to 4, the quantitative

investigation of self-healing ability due to damage and healing by respect to carbonation coefficient has not been

fully understood. Therefore, this chapter focuses on self-healing an index to evaluate the self-healing efficiency

with respect to carbonation coefficient in this chapter and designs a more durable self-healing concrete

incorporating fly ash and blast furnace slag. Finally, new durability factor in cementitious based concrete was

suggested to be reflected an original durability.

95
5.1.1 Self-healing efficiency
Deterioration of concrete structure, that is inevitable when exposed to natural climate, has become a major

concern of infrastructure and building designers, and it is especially dependent on various environmental factors for

temperature and humidity. Due to the damaged concrete structure, cracks occur in the concrete, and then

carbonation process is faster due to the fact that chemical reaction between carbon dioxide from the air and the

hydration products of cement in concrete can causes a reduction in the alkalinity of concrete, leading to the

reduction of service life time of concrete structure due to the steel corrosion (Ho DWS, Lewis RK 1987; O. Burkan

Isgor and A. Ghani Razaqpur 2004). In addition to the problems, when cracks introduces, the cost and amount of

labor required for diagnosis and repair work should be needed in order to restore the original performance (Min Wu

et al 2012; Na et al 2012).

Based on the observation above, it is revealed that carbonation process plays a major role in the long-term

durability, thus, an index of self-healing ability by taking into consideration of difference between damage and

healing cases should be considered and thus suggested in this chapter.

In advance, the carbonation coefficient was calculated from the results of carbonation depth of all investigated

samples, by the following equation:

xk t (5.1)

where x is carbonation depth (mm); k is the carbonation coefficient (mm/weeks0.5) and t is time (week)

Then, self-healing ability (ES) was estimated from the carbonation coefficient for difference between damage

and healing according to following equation;

PS
ES  (5.2)
DI

where ES is a index of self-healing; PS is self-healing potential(D-S); DI is the damage degree (D-I); I is initial

carbonation coefficient before damage; D is carbonation coefficient after damage and S is carbonation coefficient

after healing

96
(3.87) ES (1.85) ES
3 (1.00) 3 (1.00)
Self-healing region Self-healing region

Original potential (PS)


Original potential (PS)

(0.75) (0.75)
2 2

(0.50) F (0.50)
FA F
FA
1 1
(0.25) (0.25)
NA NA N
N
0 0
0 1 2 3 0 1 2 3
Damage index (DI) Damage index (DI)

(a) 4 weeks (b) 13 weeks

Fig. 5.1 The relationship between damage indexes (DI), original/potential ability (PS) and self-healing ability (ES)

Fig. 5.1 represents the results of self-healing ability in all investigated concrete samples, which include 10 % by

volume fly ash replacement ratio with (F) or without air entrained agent (FA) and normal concrete samples with (N)

or without air entrained agent (NA) with different curing ages at 28 and 56 days. Before discussion, self-healing

ability (ES) is strongly related to original/potential (PS), which plays an influential factor affecting the self-healing

ability due to remaining un-reacted cement in the cement matrix. This behavior can be attributed to the fact that the

hydration of un-reacted cement may proceed for a long time when space with pores and micro cracks may be

within the concrete, especially under high temperature and humidity.

If ES value is much higher than 1, the carbonation coefficient can be healed than initial value because of original

/potential ability (PS) and self-healing effect. It can be confirmed from the figure that the ES value of self-healing

ability for concrete sample made with fly ash (FA) and air entraing agent is higher than that of other concrete

samples. In addition, the ES value in all investigated concrete samples would decrease with increasing curing ages.

The fact that un-reacted cement and fly ash would consume, which proceeds hydration, is reason for the decreasing

ES value in all investigated concrete samples. Moreover, it is confirmed that carbonation coefficient as an index of

healing efficiency is available to quantitatively evaluate the self-healing ability in all investigated concrete samples.

However, for the quantitative investigation of self-healing ability, some problems can be observed about evaluation
97
and damage methods. One is that the accelerated carbonation test is difficult to apply for concrete structure.

Secondly, the damage degree with respect to rapid freeze and thawing testing is also difficult to correctly induce

micro cracks in concrete sample.

Moreover, there is a lack of information about a comparison research on the use of fly ash and ground granulated

blast furnace slag blended concrete with air entraing agent under same damage condition. Therefore, the main

objective is (1) to investigate the comparison to fly ash and ground granulated blast furnace slag as self-healing

agent in AE concrete mixtures, (2) develop the durability evaluation equation (frost and thaw durability)

considering the self-healing, (3) explore sustainability of self-healing performance until 1 year, and (4) examine

self-healing effect of fly ash ground granulated blast furnace slag blended concrete under same damage degree.

5.2 Experimental program


5.2.1 Experimental materials
Ordinary Portland cement (OPC, density: 3.16 g/cm3, blain fineness: 3250 cm2/g), fly ash and (FA, density: 2.15

g/cm3, blain fineness: 3890 cm2/g) ground granulated blast furnace slag (BFS, density: 2.91 g/cm3, blain fineness:

4030 cm2/g) were used in AE concrete mixture. Chemical and physical properties of OPC, FA and BFS and

physical properties of used aggregate materials are tabulated in Chapter 2. In the mix proportions of concrete as

listed in Table 5.1 (series 1), water to binder ratio in all concrete sample were kept at 0.55 and 0.48 with the

addition BFS and FA. The BFS was replaced fine aggregate by mass of replacement ratio 15%, and cement by mass

of replacement ratios 15 and 45%. The FA was incorporated fine aggregate by mass of replacement 15%. In

addition, in series 2, in order to develop more superior self-healing blended concrete, three mixtures, which involve

15% by mass of each fly ash and BFS replacement ratio as fine aggregate, were prepared.

A general pan mixer was used in producing all AE concrete mixtures. The mineral materials (OPC, FA and BFS)

were added in the mixer for 1 min, after that, water included HRWR (high-range water reducing agent), AEWR (air

entraining water reducing agent), AE (air entraining) agent and AA (antifoaming agent) was added to improve to

workability of concrete mixtures into dry mixtures and mixed for 2 to 5 minutes. In addition, concrete mixture

incorporating fly ash, air entraining agent for fly ash was also added. After desired fresh characteristics enhanced

98
(target slump and air content were 180mm and 4 to 6%), the mixed AE concrete mixtures were cured for 1 day and

then placed in water before required testing ages.

Table 5.1 Mix proportions of concrete mixture

Unit weight(kg/m 3 ) Air


Mix W/C s/a Slump
Series Mixture Fine Coarse content
No. (%) (%) W Cement BFS FA HRWR AE agent (mm)
aggregate aggregate (%)

1 Series1 NAH 55 45 174 316 - - 832 993 0.002 3) 195 4.3

2 NAL 48 43 180 375 - - 764 994 0.002 3) 195 4.0

3 SA15C 48 43 177 313 55 - 769 999 0.0012 1) 205 4.2

4 SA45C 55 45 163 163 133 - 849 1014 0.0012 1) 205 6.5


250mℓ
5 SA15S 55 43 172 313 47 - 778 1011 /B100k 0.0012 1) 195 5.9
g
6 FA15S 55 43 169 307 - 46 778 1011 0.018 2) 180 5.5

7 Series2 NA 55 43 174 316 - - 824 1000 0.002 3) 190 3.9

8 FA15S 55 43 169 307 - 46 778 1011 0.018 2) 180 4.4


1)
0.0012
9 FASA30 55 43 171 310 47 47 738 961 200 4.2
0.018 2)

5.2.2 Experimental methods


5.2.2.1 Mechanical properties
Before evaluating the self-healing effect, compression test of investigated concrete using cylinder specimen

(Ø 100×200mm) were examined at 3, 28, 91 and 365 days in series 1 and at 42 days (6 weeks) in series 2, in

accordance with JIS A 1108 and relative dynamic modulus of elasticity was also measured according to JIS A 1127.

5.2.2.2 Frost resistance


Rapid freeze and thaw testing were performed using concrete specimen (400×75×75mm prism) according to

ASTM C 666 A. To evaluate the effect of curing ages on frost resistance of concrete specimen, all of the specimens

were cured with different curing ages for 2, 13 weeks and 1 year in series 1, and at 42 days (6weeks) in series 2.

. According to the ASTM C666 procedure A, the freeze and thaw cycling test, which has been widely used in order

99
to quantify the freezing and thawing durability and which involves the freezing minimum temperature at -18℃ for

2 hours 30 minutes (freezing) and thawing temperature at +5℃ for 1 hour 30 minutes (thawing), was conducted at

the curing age of 28 days. The mass loss, length change and relative dynamic modulus of elasticity for all

investigated samples were measured every 30 cycles until 300 cycles in accordance with JIS A 1127 procedure. The

test was also completed when a sample reached 300 cycles or relative dynamic modulus of elasticity reached 60%

as critical limit from ASTM C666 procedure A. The experimental result was obtained from an average of three

samples. Relative dynamic modulus of elasticity, durability factor and length change can be calculated from the

results of rapid freeze and thaw testing of all investigated samples, which are expressed as follows:

fn2
Pn  ( ) 100 (3.1)
f02

where;

Pn is the relative dynamic modulus of elasticity, after c cycles of freezing and thawing, percent; f 0 is the

fundamental transverse frequency at 0 cycles of freezing and thawing, and f n is the fundamental transverse

frequency after n cycles of freezing and thawing.

P N
DF  (3.2)
M

where DF is the durability factor of the test specimen; P is the relative dynamic modulus of elasticity at N

cycles (%); N is the number of cycles at which P reaches the specified minimum value less than 60% or the

relative dynamic modulus of elasticity after 300 cycles, and M is the relative dynamic modulus of elasticity after

300 cycles.

l 2  l1
Lc  (3.3)
Lg

where Lc is the length change ratio of the test specimens after c cycles of freezing and thawing (%); l1 is the

length comparator reading at 0 cycles(mm); l 2 is the length comparator reading at after c cycles(mm), and Lg is

the effective gage length between the innermost ends of the gage studs as shown in the mold diagram.

W0  Wn
Wn  (3.4)
W0

where Wn is the weight change ratio of the test specimens after c cycles of freezing and thawing (%), and W0 is

100
the weight at 0 cycles(g).

5.2.2.3 Damage method


After water curing 20℃ for 4, 13 weeks and 1 year in series 1, and after 42 days (6weeks) in series 2,

continuous cyclic loading method, which is an application of compression test, was conducted in order to damage

in AE concrete specimen, based on compression test using cylinder specimen (Ø 100×200mm). Target relative

dynamic modulus of elasticity was set at 85%. The concrete specimen was cracked until maximum compressive

strength 80% and 10 times, as already expressed in Chapter 5.

5.2.3.4 Self-healing evaluation


To evaluate self-healing ability, RDM value was measured and divided into three cases, which are initial case,

deterioration case and healing curing case In the healing curing case, the damaged specimens were cured with three

curing conditions (self-healing curing), that are in water at 20℃ for 1 week, 40℃ for 4 weeks and no water

service condition. Then, RDM value was measured after each curing conditions.

101
5.3 Results and discussion
5.3.1 Frost resistance
Durability factor for different concrete mixtures with different curing ages at 2 weeks and 1 year is shown in Fig.

5.2 that, in the case of curing age at 2 weeks, SA45C, SA15S, FA15S and SA15C concrete samples incorporating

FA and BFS is slightly higher than that of normal concrete sample. In addition, under the same replacement ratio,

FA15S concrete sample is slightly decreased compared to SA15C concrete sample. It can be remarkably seen that

durability factor of investigated concrete samples incorporating FA and BFS is significantly reached about 80 to 90

in the case of curing age at 1 year, in comparison to the normal concrete samples (NAH and NAL), implying that

addition of FA and BFS exhibits good freezing and thawing resistance with increasing curing age.

cured for 2 weeks cured for 1 year


100
86 87 86
90 81
80 73 73
Durability factor

70 62 63
57 60
60 54
50 48
40
30
20
10
0
NAH NAL SA15C SA45C SA15S FA15S
No admixture Slag Fly ash
Mixture Designation

Fig. 5.2 Freeze and thaw testing results for six investigated concrete samples with different water curing ages 2

weeks and 1 year.

5.3.2 Self-healing effect on the change of relative dynamic modulus of elasticity


Fig. 5.3 shows the change of relative dynamic modulus of elasticity (RDM) for four types of concrete samples

with and without mineral admixtures (fly ash and ground granulated blast furnace slage) with two types of curing

ages at 4 weeks and 1 year. After deterioration (micro cracking), the RDM values in all concrete samples were

between 85 to 93%. In the case of curing age at 4 weeks, it can be observed from the figure that, after deterioration,

102
RDM value in all investigated samples is significantly increased until 1 week in both of self-healing curing

conditions which are in water 40℃ case and sealing cases.

However, after water curing for 2 weeks, little increase in most of concrete samples can be observed, except for

FA15S concrete sample. This observation implies that significant increase of RDM value exhibits in water curing

condition at early curing ages. For comparison with two types curing cases, a sealed curing condition has little

influence on self-healing ability compared to water curing condition. It is well-known that self-healing behavior

varies depending on the presence of water, but this trend is different in current study. In addition, it can be seen that

FA15 concrete sample, which involves 10% by mass of fly ash replacement ratio, is continuously increased

compared to the other concrete samples. This observation agrees with the previous findings obtained, which is

previously discussed in Chapter 2.

As for curing age at 1 year, significant recovery did not observe and there are no significant differences for most

concrete samples with two types of curing conditions. In addition to the findings, it can be inferred that, when the

curing age increase (initial period), self-healing ability would decrease due to hydration process of cement, FA and

BFS, resulting in a decrease of self-healing ability, which is similar to previous research, as demonstrated in chapter

3.

On the other hand, self-healing ability can be strongly related to the damage degree obtained. In this study, in

order to investigate fly ash and blast furnace slag on self-healing efficiency in this study, all investigated concrete

samples were damaged by means of cyclic loading method, but damage degree using the method does not same.

Thus, all investigated concrete samples should be tested damage degree is needed to investigate under the same

damage degree and to accurately evaluate the self-healing ability.

103
110 110 110
NAH SA45C SA15C
105 105 105

100 100 100

RDM(%)

RDM(%)
RDM(%)

95 95 95

90 90 90

85 85 85

80 80 80
I D 1W 2W 3W 4W I D 1W 2W 3W 4W I D 1W 2W 3W 4W

110 110 110

105
FA15S 105
SA15S 105
NAL
100 100 100
RDM(%)

RDM(%)
RDM(%)

95 95 95

90 90 90

85 85 85

80 80 80
I D 1W 2W 3W 4W I D 1W 2W 3W 4W I D 1W 2W 3W 4W

water curing at 20℃ for 1 W water curing at 40℃ for 4 W sealed curing at 40℃ for 4 W
(a) Water curing at 20℃ for 4 weeks
110
105
RDM(%)

100
95
90
85
80
I D 1W 2W 3W 4W
110 110 110
NAH SA45C SA15C
105 105 105

100 100 100


RDM(%)

RDM(%)

RDM(%)

95 95 95

90 90 90

85 85 85

80 80 80
I D 2W 4W I D 2W 4W I D 2W 4W

110 110 110


FA15S 105
SA15S NAL
105 105

100 100 100


RDM(%)

RDM(%)

RDM(%)

95 95 95

90 90 90

85 85 85

80 80 80
I D 2W 4W I D 2W 4W
I D 2W 4W

water curing at 20℃ for 1 W water curing at 40℃ for 4 W sealed curing at 40℃ for 4 W
(c) Water curing at 20℃ for 1 year
110
105
RDM(%)

95 100
Fig. 5.3 Change of relative
85 dynamic modulus of elasticity for different concrete samples with different water
90
80
I D 1W 2W 3W 4W

curing ages at 20℃ for 4 weeks and 1 year (blue color is water curing at 20℃ for 1 week, red color is water

curing at 40℃, and green is a sealed curing at 20℃ for 4 weeks)

104
5.3.3 New durability factor considering self-healing effect
In order to access potential self-healing, an evaluation index was developed and calculated in Chapter 4 from

experimental data based on carbonation coefficient by measurement of accelerated carbonation. However, this

evaluation index did not consider the deterioration resistance of original durability. In addition, as mentioned above,

carbonation coefficient as self-healing index might not apply in concrete structure. On the other hand, non

destructive technique for evaluating difference between damage degree and self-healing ability is required to keep

the concrete structure. Thus, resonance frequency was used. Relative dynamic modulus of elasticity was calculated

from the obtained results of resonance frequency before or after damage and healing, as demonstrated in Fig. 5.4.

Besides, new durability factor in concrete sample, which considers self-healing, may be also estimated.

SI  ES  DF (5.3)

where SI is the durability considering self-healing effect, ES is the self-healing effect, DF is the durability factor

obtained by measurement of rapid freeze and thaw testing


Relative dynamic modulus of elasticity(%)

Initial property
After deterioration
After self-healing
I
Di
S
Ps
D

Fig. 5.4 Schematic illustration of concept for new self-healing system

Figure 5.5 represents the durability factor values by taking into consideration of the self-healing effect in all

investigated concrete samples with different initial curing age of 4, 13 and 1 years in water curing case at 40℃ and

sealed condition case at 20℃, which are with two types of control concrete of water to cement ratio (0.48 and 0.55),

with fly ash replacement ratio as 15% by mass, and with BFS replacement ratio as 15, 45% by mass as cement and

fine aggregate. The durability factor was calculated from Eq. (5.3)

105
After deterioration, it can be seen from the figures that most of the investigated samples would decrease the

self-healing index (ES), implying that new durability factor (SI) would decrease with increasing the curing age due

to the fact that un-reacted cement can be consumed with hydration.

In the case of high temperature in water at 40℃, new durability factor of concrete sample, which includes 15%
(a) Curing age of 4 weeks in water at 40℃ (b) Curing age of 4 weeks at 20℃
by mass
100fly ash replacement ratio is higher in comparison to other
ES=1.20 ES=1.66
100 investigated concrete samples. This is probably
New durability factor(SI)

New durability factor(SI)


ES=1.05
80 80
due to the fact that micro cracks can be filled by hydrated products of fly
ES=1.08 ash ES=0.88
ES=0.89 due to the pozzolanic reaction,
ES=0.95 ES=0.92
60 60
ES=0.78
resulting inES=0.94
higher durability factor. After initial curing age of 1 year, in comparison withES=0.75
ES=0.82
NAH and NAL concrete
40 40
samples as normal concrete, fly ash and slag blended concrete samples seemed to have high durability factor under
20 20

high temperature
0 curing condition. 0
NAH SA45C SA15S FA15S SA15C NAL NAH SA45C SA15S FA15S SA15C NAL
Mixture designation Mixture designation
(a) Curing age of 4 weeks in water at 40℃ (b) Curing age of 4 weeks at 20℃
100 100
100 ES=1.20 ES=1.66 100
(a)Curing age of 13 weeks in water at 40℃ (c)
factor(SI)

New durability factor(SI)

(c) ES=1.05 (d) Curing age of 13 weeks at 20℃


durabilityfactor(SI)

New durability factor(SI)

80 80
80 ES=1.08 80
ES=1.16 ES=0.89 ES=0.88
ES=0.95 ES=0.92
60 60
60 ES=0.94 60 ES=0.78
ES=0.75
Newdurability

ES=0.57 ES=0.82
40 ES=0.51 ES=0.54 40 ES=0.53
40 ES=0.50 40
ES=0.57 ES=0.39 ES=0.41 ES=0.37
20 20 ES=0.45
ES=0.21
New

20 20
0 0
0 NAH SA45C SA15S FA15S SA15C NAL 0 NAH SA45C SA15S FA15S SA15C NAL
NAH SA45C Mixture
SA15S designation
FA15S SA15C NAL NAH SA45C SA15S FA15S SA15C NAL
Mixture designation
Mixture designation Mixture designation

100 100
100 100
(c) Curing age of 13 weeks in water at 40℃ (d) Curing age of 13 weeks at 20℃
factor(SI)

factor(SI)

80 (e) Curing age of 1 year in water at 40℃ 80 (f) Curing age of 1 year at 20℃
factor(SI)
factor(SI)

80 ES=1.16 80
60 (b) 60 (d)
durability

60
durability

60 ES=0.57
ES=0.51 Es=0.75 ES=0.54 Es=0.79 Es=0.70
durability

Es=0.61 Es=0.62 ES=0.53


durability

40 Es=0.71 40
ES=0.50 ES=0.39 Es=0.60
40 ES=0.57 Es=0.48 40 ES=0.45 ES=0.41 ES=0.37
Es=0.42 Es=0.45 ES=0.21
20 Es=0.53
New

New

20 Es=0.44
New
New

20 20
0 0
NAH SA45C SA15S FA15S SA15C NAL 0 NAH SA45C SA15S FA15S SA15C NAL
0
NAH SA45C Mixture
SA15Sdesignation
FA15S SA15C NAL NAH SA45C SA15Sdesignation
Mixture FA15S SA15C NAL
Mixture designation Mixture designation
100 100

Fig. 5.4 (e) durability


New Curing age of 1 yearfor
factor in water at 40℃
different concrete (f) Curing
samples with different age ofconditions
curing 1 year at 20℃
after initial curing age
New durability factor(SI)
New durability factor(SI)

80 80

60
of 4 weeks and 1 60
year
Es=0.75 Es=0.71 Es=0.61 Es=0.62 Es=0.79 Es=0.70
Es=0.48 Es=0.60
40 40
Es=0.42 Es=0.45 Es=0.53
Es=0.44
20 20

0 106 0
NAH SA45C SA15S FA15S SA15C NAL NAH SA45C SA15S FA15S SA15C NAL
Mixture designation Mixture designation
5.3.4 Self-healing ability comparison
The second objective of this study was to compare the fly ash and BFS self-healing ability as healing agent. For

this purpose, with 15% by mass of fly ash BFS replacement ratio, the SI value considering self-healing effect is

therefore presented in Fig. 5.5. It is confirmed from the figure that fly ash concrete sample is higher than that of

BFS concrete sample, SA15C concrete sample is also constant after 13 weeks (initial period), suggesting that fly

ash concrete sample lead to better self-healing performance compared to the BFS concrete sample under the same

replacement ratio in this study.

160

140 4 weeks
Durability factor including

13 weeks
120
self-healing (SI )

1 year
100

80

60

40

20

0
SA15S FA15S
Mixture Designation

Fig. 5.5 Durability factor with self-healing effect different water curing ages at 20℃ for 4, 13 weeks and 1 year

5.3.5 Self-healing effect of fly ash and slag blended composites


Fig. 5.5 shows the results of compressive strength for different concrete mixtures, which are normal concrete

without admixture (NA), 15% replacement ratio by mass fly ash (FA15C) and 15% replacement ratio by mass fly

ash with BFS (FASA30). It can be seen that, in all concrete samples, there was no significant differences

observed in the compression test (44.5MPa for NA, 45.1MPa for FA15S and 42.7 for FA30).

Fig. 5.6 shows the results of freeze and thaw testing for different concrete mixtures after water curing age of 6

weeks. The durability factor in all investigated concrete samples was not significantly different (97.7 for NA, 99.9

for FA15S and 97.6 for FA30), indicating that the samples have higher frost resistance for the freeze and thaw

cyclic testing.

107
60 100
97.7 99.0 97.6
Compressive strength(MPa)

50 45.1
44.5
42.7 90

Durability factor
40

30 80

20
70
10

0 60
NA FA15S FASA30 NA FA15S FASA30
Mixture Designation Mixture Designation

Fig. 5.5 Compressive strength result for three Fig. 5.6 Durability factor result for three
investigated concrete samples investigated concrete samples

Figs. 5.7 and 5.8 show the change of relative dynamic modulus of elasticity for three concrete samples (NA,

FA15S and FASA30) under the same deterioration degree, and calculated durability factor (SI). The target value

was 85% of relative dynamic modulus of elasticity. After deterioration, in case of normal concrete sample without

admixture, the relative dynamic modulus of elasticity increased about 88% and then the little increase was observed

after water curing age of 5 weeks. Meanwhile, FA15S and FASA30 samples, which incorporate fly ash and BFS,

the relative dynamic modulus of elasticity tend to increase continuously even after 1 week. In addition, it is found

that significant increase in FASA30 sample was observed. Above observation, it can be confirmed that the

combination of fly ash BFS has excellent self-healing ability compared with normal and fly ash concrete samples.
Relative dynamic modulus of elasticity (%)

102 140
100
120
98
Durability factor(SI)

96 100
94
80
92
90 60
88
86 40
OPC
NA
84 the same deterioration
FA
FA15S 20
82 FG
FASA30
80 0
No After 1 week 2 weeks 3 weeks 4 weeks 5 weeks
NA FA15S FASA30
cracking deterioration
Mixture Designation

Fig. 5.7 Self-healing effect on the change of relative Fig. 5.8 Calculated durability factor based on self-healing
dynamic modulus of elasticity for different concrete effect

108
5.4 Conclusions
Based on the experimental results, the following conclusions can be derived;

(1) It was found that frost resistance depends on water curing ages at 2 weeks and 1 year. Increasing curing age,

except for NAL concrete sample, freeze and thaw durability would increase in most concrete samples. In

addition, it was confirmed that, in comparison with N concrete sample, concrete samples the incorporation of

fly ash and BFS tend to be better frost resistance in the case of curing age at 1 year in series 1.

(2) Durability factor considering self-healing effect was developed and was calculated from experimental data

based on relative dynamic modulus of elasticity. From the calculated durability factor, it was observed that the

durability factor can be decreased with decreasing self-healing effect in all concrete samples in series 1.

(3) Under the same admixture replacement ratio, concrete sample incorporating fly ash may lead to a better

self-healing ability in comparison with BFS concrete sample in series 1.

(4) In series 2, three types of concrete samples were prepared and it was found that concrete samples combination

of fly ash and BFS has excellent self-healing ability compared with normal concrete and fly ash concrete

samples.

109
Reference

Emily N. Herbert and Victor C. Li, Self-Healing of Microcracks in Engineered Cementitious Composites (ECC)
Under a Natural Environment, Materials, 6, 2831-2845 (2013)

Ho DWS, Lewis RK. Carbonation of concrete and its prediction. Cement Concr Res, 17:489–504 (1987)

Kim Van Tittelboom, Nele De Belie,Willem De Muynck and Willy Verstraete, Use of bacteria to repair cracks in
concrete, Cement and Concrete Research 40 pp. 157–166 (2010)

Min Wu, Björn Johannesson and Mette Geiker, A review: Self-healing in cementitious materials and engineered
cementitious composite as a self-healing material, construction and Building Materials 28 pp. 571–583 (2012)

O. Burkan Isgor and A. Ghani Razaqpur, Finite element modeling of coupled heat transfer, moisture transport and
carbonation processes, Cement & Concrete Composites 26 57–73 (2004)

Seung Hyun Na, Yukio Hama, Madoka Tanigchi, Takahiro Sagawa and Mohamed Zakaria: Experimental
investigation on reaction rate and self-healing ability in fly Ash blended cement mixtures, J. Adv. Concr. Technol.,
Vol.10, No.7, pp.207-218 (2012)

Virginie Wiktor and Henk M. Jonkers, Quantification of crack-healing in novel bacteria-based self-healing concrete,
Cement & Concrete Composites 33, 763–770 (2011)

110
CHAPTER 6 CONCLUSIONS AND FUTURE WORK

111
6.1 Overview
To date, various techniques such as high strength concrete, which is designed with low water to cement ratio

(low W/C), and air-entraing agent (AE agent) addition have been widely used in order to control frost damage of

concrete structure in concrete industry. However, some disadvantages these techniques have reported by many

researchers, who were referred that, in the case of high strength concrete, cracks can be occurred by considerable

autogenous, drying shrinkage and brittle fracture due to high cement content and low W/C (I. Maruyama and A.

Teramoto 2013). In addition to the findings, it is well-known that the high strength concrete has an excellent frost

resistance because of AE agent addition which is suitable during mixing process. However, it has been reported that

frost resistance in the high durable concrete can decrease with increasing exposure ages (Yukio H. et al 2002). The

reason is due to the fact that cracks can occur during summer, leading to poor frost resistance (Yoshimichi A. et al

2007). Above observation, cracks due to the drying shrinkage and poor frost resistance are a considerably big

problem in concrete industry and new technique is needed to addresses the occurred cracks in the concrete

structure.

The objective of present research is to investigate the effect of self-healing performance on cement based system

incorporating fly ash and ground granulated blast furnace slag. The conclusions that can be drawn from the findings

of this research are given in following Sections.

6.2 Conclusions
6.2.1 Experimental investigation of reaction rate and self-healing ability in fly ash
blended cement mixtures
To quantify the effect of reaction rate on fly ash, the hydration rate of fly ash and consumption of calcium

hydroxide in fly ash paste containing calcium hydroxide reagents were examined at different curing ages and

curing temperatures, five types of fly ash blended cement paste, each of which with 10% by mass fly ash

replacement ratio, were prepared for the acceleration test at 80ºC. Four fly ash cement pastes, with two types of fly

ash and with fly ash replacement ratios by mass 10% and 30%, were tested to measure the rate of reaction of fly ash

in the mixtures. TG-DTA, XRD-Rietveld analysis and ignition loss were measured and the selective dissolution

method was used.

112
For determination of fly ash reaction rate,based on the hydration characteristics of fly ash and calcium hydroxide

reagent, the determination of reaction rate for fly ash using pattern 4, which was obtained from the melting ratio for

the acid insoluble matter in the grass phase, appeared to have better response to experimental results when

compared with other pattern.

For prediction of fly ash reaction rate, comparison of predicted and experimental results of fly ash reaction rate

was performed based on hydration model. It can be inferred that fly ash blended paste, which involves 10% of fly

ash replacement ratio, has a good performance in comparison to 30% replacement ratio by mass of fly ash. This

model is useful to understand the fly ash reaction rate.

To optimize fly ash replacement ratio, we assumed that when the relative dynamic modulus of elasticity was

about 60%, the relative length change was about 1×10-3 (Koh et al 1981). This is due to the fact that internal

cracking in the concrete sample occurs for the rapid freeze and thaw testing. From this observation, we assumed

that micro cracks in concrete are about 0.3% by volume due to decrease of relative dynamic modulus of elasticity

until 60% and suggested optimization equation of fly ash replacement ratio. It was found that the practical fly ash

replacement ratio in fly ash blended concrete could be 10% to 15% and 11% to 20% for W/C 0.50 to 0.55 and 0.55

to 0.60, respectively.

For self-healing effect on various mechanical tests including compressive, bending strength, accelerated

carbonation, the fly ash blended mortar samples can recover the compressive/bending strength and carbonation

coefficient after deterioration, which is introduced in rapid freeze and thaw testing. It was clarified that curing

conditions 20ºC at 1 week was not enough to heal the micro cracks in comparison to the curing conditions in water

at 40C for 4 weeks, implying that the fly ash has a pozzolanic behavior that reacts with calcium hydroxide from

cement and produces C-S-H gel, which may seal micro cracks in high temperature curing conditions.

By investigating self-healing effect on pore structure and crack density, from the mercury intrusion porosimetry

and point count test, we evaluated the self-healing ability before and after deterioration. From the measurement, it

was confirmed that, after water curing at 40ºC for 4 weeks, the reductions of pore volume rate 50 nm to 10

micrometers in fly ash mortar sample FA20 were higher than those in normal mortar sample without fly ash. The

113
similar tendency was also observed the point count test results, which implies that fly ash hydration products can

seal micro cracks under high temperature curing conditions, as an evidence for self-healing phenomena.

6.2.2 Self-healing effect and frost resistance of fly ash blended concrete
This chapter was divided into three series, which are Non-AE concrete / AE concrete self-healing performance

and the variation of self-healing ability of two types of concrete after exposure. Self-healing ability for fly ash

blended Non-AE concrete was experimentally verified. In the case of Non-AE concrete with fly ash, slight increase

was observed from the results of accelerated carbonation, relative dynamic modulus of elasticity. In addition, under

accelerated environmental conditions, in which include repetition case until 30 cycles and healing in water 40C at

3days, no/little self-healing ability of fly ash concrete sample was observed, this is due to the fact that assumed

accelerated environmental conditions could be extreme conditions to healing the damage concrete sample. This

problem will be able to solve AE agent addition which is suitable.

For self-healing ability and frost resistance of fly ash blended AE concrete, it was found that AE concrete has a

better frost resistance compared to non-AE concrete. it is interesting that high frost resistance of AE concrete was

decreased with increasing curing ages at 28 and 365 days.

In addition, at curing age of 28 days, the recovery of fly ash blended concrete (F and FA) was observed

compared to the normal concrete samples (N and NA). From these findings, it was inferred that, when curing age

increases, self-healing efficiency in all investigated concrete was decreased, because of decrease of un-reacted

cement and fly ash with respect to relative dynamic modulus of elasticity.

In chapter 2 to 4, since limitation of self-healing research was observed, we investigated new damage technique

and accelerated curing conditions. The new damage technique was developed based on compression test and it is

possible to make damage within AE-concrete concrete. The technique is an effective way to study in self-healing

field

6.2.3 Self-healing ability of ground granulated blast furnace slag (BFS) blended
system after frost damage
Thirteen mortar samples were prepared and the effects of BFS on the self-healing efficiency and frost resistance
114
of blended mortar mixtures at the different replacement ratio, method, curing conditions were investigated by

measurement of rapid freeze and thaw, accelerated carbonation, relative dynamic modulus of elasticity,

compressive and bending strength. For damage degree, all of the mortar samples deteriorated until 60% of relative

dynamic modulus of elasticity and then were stored in different curing cases, which involve in water at 20ºC for 1

week case and 40ºC for 4 weeks case.

For the effects of BFS addition on compressive/bending strength development and frost resistance before

deterioration, in the case of BFS replacement for sand, the mortar samples exhibited high frost resistance

performance when compared to mortar sample which was replaced to cement. In addition, it was found that 45% of

BFS replacement ratio for sand has a positive effect on freeze and thaw cyclic performance compared with 20% of

BFS replacement ratio for sand. This tendency agrees with compressive and bending strength results. It is found

that addition of BFS for sand has positive effect on compressive and bending strength, compared to normal mortar

samples without BFS, especially for sand replacement and increasing BFS finenesses. Above this observation, it

was revealed that BFS replacement ratio and method would affect the compressive/bending strength development

and freeze and thaw resistance. In addition, the replacement of BFS is advisable to make more durable mortar

(concrete). For BFS self-healing effect on changes of relative dynamic modulus of elasticity, accelerated

carbonation test, compressive and bending strength after deterioration. In the case of 3000 g/cm3 of BFS fineness,

compressive/bending strength was recovered than initial value after water curing at 40C for 4 weeks. The strength

development of BFS blended mortar samples, which incorporate 8000 cm2/g of BFS fineness, was partially healed.

6.2.4 Comparison of self-healing ability of concrete mixture incorporating fly ash


and ground graduated blast furnace slag
By taking into consideration the self-healing ability, durability factor can be calculated from experimental data

based on relative dynamic modulus of elasticity. Based on the obtained equation, the durability factor could be

decreased with decrease of self-healing efficiency. In addition, concrete sample, which incorporates fly ash, was

found to be better self-healing ability compared with BFS concrete sample under the same replacement ratio. More

durable concrete and higher self-healing concrete was developed by combination of fly ash and BFS.
115
6.3 Future work
With incorporation of two types of materials which include fly ash and BFS, we investigated the self-healing

ability to fill the micro cracks with respect to the relative dynamic modulus of elasticity, accelerated carbonation

test, mercury intrusion porosimetry, ultrasonic velocity and torrent permeability. Before that, we introduced micro

cracks into the mortar and concrete sample in terms of rapid freeze and thaw testing and compression test.

However, in the case of micro cracks introduced by measurement of the compression, it is possible that the

cracks may be different the frost damage due to freezing and thawing cyclic behaviour. When exposed to the

freezing and thawing cyclic conditions, the lower frost resistance of concrete structure could be occurred to the

scalling in the surface and micro cracking in the internal part. This is different things with the freezing and thawing

cyclic damage.

In addition, quantifying the artificial micro crack (width and depth) by means of compression test is limited.

This is a problem which we should address in the future work.

References
I. Maruyama and A. Teramoto, Temperature dependence of autogenous shrinkage of silica fume cement pastes
with a very low water–binder ratio, Cement and Concrete Research 50, pp. 41–50 (2013)

Yukio H., Eisuke H., Osamu S. and Tomosawa F., Effect of curing condition before freezing and thawing test to
frost resistance of high strength concrete and high fluidity concrete, cement science and concrete technology, No.56,
pp. 425-430 (2002) (in Japanese)

Yoshimichi A., Fumiaki M., Sumio S. and Yukio H., Nano-structure changes of C-S-H in hardened cement paste
during drying at 50ºC, Journal of Advanced Concrete Technology, Vol.5, No.3, pp. 313-323 (2007)

116
ACKNOWLEDGEMENTS

I would never have been able to finish my dissertation without the guidance of my laboratory members, help

from friends, and support from my family.

I would like to express the deepest gratitude to my advisor, Dr. Yukio Hama, for his excellent guidance, caring,

patience, and providing me with an excellent atmosphere for doing research.

I would also like to thank, Dr. Takahiro Sagawa, Dr. Madoka Taniguchi, Dr. Yoshihiko Kishimoto, and Dr.

Osamu Katsura for guiding my research for the past several years and helping me to develop my background in

concrete engineering, environmental technology and cement chemistry.

Special thanks goes to Dr, Mitsuo Mizoguchi and D. Noriyuki Sugata, who were willing to participate in my

final defense committee at the last moment.

I would like to thank Dr. Wenyan Zhang, who as a good friend and was always willing to help and give her best

suggestions. It would have been a lonely lab without her.

Many thank to Tohru Nakamura, and other undergraduate and Master’s degree students in the laboratory of

Building Materials and Construction. My research would not have been possible without their helps.

Finally, I would also like to thank my parents, older brother and his family. They were always supporting me and

encouraging me with their best wishes.

117
LIST OF PUBLICATIONS AND CONFERENCE PRESENTATION

Technical papers:
1. Seung Hyun Na, Yukio Hama, Madoka Taniguchi, Osamu Katsura, Takahiro Sagawa and Mohamed Zakaria,
Experimental Investigation on Reaction Rate and Self-healing Ability in Fly Ash Blended Cement Mixtures,
Journal of Advanced Concrete Technology, pp. 240-253, 2012. (Published)

Proceedings and International Conferences:


1. SeungHyun Na, Yukio Hama, Madoka Taniguchi, Osamu Katsura, Takahiro Sagawa and Mohamed Zakaria,
Influence of deterioration degree on self-healing ability of concrete incorporating fly ash and ground granulated
blast furnace slag, Joint Seminar on Environmental Science and Disaster Mitigation Research 2013, 8 March 2013,
Muroran Institute of Technology, Muroran, Japan. (Oral English presentation by poster prize winners)

2. SeungHyun Na, Yukio Hama, Madoka Taniguchi and Takahiro Sagawa, A comparison with self-healing effect of
concrete using fly ash and ground granulated blast furnace slag, Proceedings of Japan Concrete Institute, Vol. 34(1),
1402-1407, 2012. (Oral Japanese presentation)

3. SeungHyun Na, Yukio Hama, Madoka Taniguchi, Osamu Katsura, Takahiro Sagawa and Mohamed Zakaria,
Effects of continual deterioration and the same damage degree on self-healing concrete, Proceeding of 6th
International Symposium between Korea, China and Japan on Performance Improvement of Concrete for Long Life
Span Structure, 24-25 Aug., Cheonju, Korea, 2012. (Oral English presentation)

4. SeungHyun Na, Musaka Osamu, Yoshihiko Kishimoto, Mohamed Zakaria, and Yukio Hama, Consideration of
Evaluation Index for Self-Healing Effect of Concrete, Joint Seminar on Environmental Science and Disaster
Mitigation Research 2012, 9 March 2012, Muroran Institute of Technology, Muroran, Japan. (Poster award)

5. SeungHyun Na, Yukio Hama, Madoka Taniguchi, Osamu Katsura and Takahiro Sagawa, Evaluation of
self-healing effect of concrete using fly ash and ground granulated blast furnace slag, Proceedings of the 3rd
International Conference on Self-Healing Materials, Bath, UK. 27-29 June 2011. (Poster presentation)

6. Madoka Taniguchi, Yukio Hama, Osamu Katsura, Takahiro Sagawa, Seung Hyun Na, self healing of
frost-damaged concrete incorporation fly ash, Proceedings of the 3rd International Conference on Self-Healing
Materials, Bath, UK., 27-29 June 2011. (Oral English presentation)

7. SeungHyun Na, Yukio Hama, Madoka Taniguchi, Osamu Katsura and Takahiro Sagawa, Evaluation of
Self-healing effect of concrete using fly ash and GGBFS considering frost resistance, Proceeding of 5th
International Symposium between Korea, China and Japan on Performance Improvement of Concrete for Long Life
118
Span Structure, 9-10 Aug., Asahikawa, Japan, 2011. (Oral English presentation)

8. SeungHyun Na, Masaru Shibuya, Yukio Hama, Self-Healing of Concrete using Fly-ash and Ground Granulated
Blast Furnace Slag, Joint Seminar on Environmental Science and Disaster Mitigation Research, JSED, 2011, 9
March 2011, Muroran Institute of Technology, Muroran, Japan. (Poster presentation)

119