Thermogravimetry analysis, compressive strength and thermalconductivity tests of non-autoclaved aerated Portland cement–flyash–silica fume concrete

Chalermphan Narattha1• Pailyn Thongsanitgarn1

• Arnon Chaipanich1

Received: 7 July 2014 / Accepted: 20 April 2015 / Published online: 22 May 2015

� Akademiai Kiado, Budapest, Hungary 2015

Abstract This paper reports the investigated thermo-

gravimetry analysis, compressive strength and thermal con-

ductivity tests of non-autoclaved aerated Portland cement–fly

ash–silica fume concrete. The mixes were cured in water and air

for 3, 7 and 28 days. Thermogravimetry results showed that

calcium silicate hydrate (C–S–H), ettringite, gehlenite

(C2ASH8), calcium hydroxide [Ca(OH)2] and calcium car-

bonate (CaCO3) phases were detected in all mixes. The com-

pressive strength and thermal conductivity of aerated Portland

cement–fly ash–silica fume concrete increased when compared

with aerated Portland cement–fly ash concrete after 28 days.

The compressive strength and thermal conductivity of aerated

concrete cured in water had higher values than air-cured spe-

cimens. X-ray diffraction and thermogravimetry showed that

Ca(OH)2 decreased with increased silica fume content. This is

due to the increased pozzolanic reaction when compared with

the Portland cement–fly ash mixes, which corresponds to an

increase in compressive strength and thermal conductivity.

Keywords Thermogravimetry � Fly ash � Silica fume �Compressive strength � Thermal conductivity

Introduction

The cement industry is one of the most energy-consuming

industries. The carbon dioxide (CO2) emissions rate is re-

sponsible for approximately 5 % of global man-made CO2

emissions [1]. The production of greenhouse gases by the

cement industry can be reduced by using pozzolanic ma-

terials, such as fly ash and silica fume, as partial replace-

ments for Portland cement (PC) to create a blended cement

[2]. The main reasons for using pozzolanic materials are

not only the cost-effectiveness but also to improve the

properties of concrete [3].

Fly ash (FA) is a by-product of coal combustion in

thermal power plants and contains partially active glass

silica and alumina [4]. FA is widely used in blended ce-

ment and is normally used to replace Portland cement up to

30 % by mass [5–8]. The benefits of using FA in cement

are improved properties such as lower heat of hydration,

increased workability, improved resistance to sulfate at-

tack, lower shrinkage characteristics and reduced expan-

sion in mortar/concrete [9–11]. However, FA has a slow

rate of strength development compared with normal con-

crete [12, 13]. FA requires water and calcium hydroxide

[Ca(OH)2] to react (pozzolanic reaction) to form binding

materials similar to PC [14, 15]. This reaction with

Ca(OH)2 proceeds slowly since Ca(OH)2 is formed from

the hydration reaction of PC.

Silica fume (SF) is a by-product of the ferrosilicon in-

dustry. SF consists of very small spherical particles of non-

crystalline silica (&100 nm) and has a high content of

amorphous silicon dioxide (more than 80 %), characteris-

tics which account for its high pozzolanic reactivity [16].

SF can improve the properties of cement-based materials,

providing increased strength, modulus and ductility

[17, 18]. Korpa et al. [19] found that SF accelerated the

early cement hydration phases by providing large amounts

of reactive siliceous surface, which serves as a site for early

precipitation of hydration products. Kohno et al. [20] re-

ported that SF increased the strength of normal cement.

Other recent studies found that a combination of SF and FA

& Arnon Chaipanich

arnon.chaipanich@cmu.ac.th; arnonchaipanich@gmail.com

1 Advanced Cement-Based Materials Research Unit,

Department of Physics and Materials Science, Faculty of

Science, Chiang Mai University, Chiang Mai 50200,

Thailand

123

J Therm Anal Calorim (2015) 122:11–20

DOI 10.1007/s10973-015-4724-8

improved the properties of concrete better than the addition

of FA alone [21, 22]. Valipour et al. [23] reported that

7.5–10 % SF is optimum amount for cement replacement.

Li et al. [24] also reported that blended cement pastes with

30 % FA and 10 % SF give the optimal mix, which con-

tribute to higher compressive strength than blended cement

pastes with FA.

Curing is an essential procedure of cement samples to

give desirable mechanical properties and endurance. The

procedure provides samples with adequate water for the

hydration reaction and avoids from drying phenomena.

Therefore, 28-day water-cured samples are used for stan-

dard tests. However, in real practice, concrete buildings are

exposed to air after mixing and depositing [25]. Aerated

concrete (or foam concrete [26]) is a cement or lime

mortar, classified as lightweight concrete, in which air

voids are entrapped in the mortar matrix by means of a

suitable aerating agent. Aerated concrete can be non-au-

toclaved (NAAC) or autoclaved (AAC) [27]; the properties

depend on the method and duration of curing. The ad-

vantages of aerated concrete are its lightweight, high de-

gree of thermal insulation and considerable savings in

energy consumption [28]. New options for the application

of aerated concrete in manufacturing processes have arisen

from utilizing a combination of chemical foaming (alu-

minum powder) and air curing. These new aerated con-

cretes as structural lightweight concrete are innovative

building materials which can be used in load-bearing

structures [29].

Most of aluminum-added aerated concrete is normally

autoclaved [30–33]. Little is known on the properties and

phase characterization of non-autoclaved aerated concrete

using aluminum especially with FA and SF. Thus, this

work investigated the compressive strength, thermal con-

ductivity and phase characterization by thermal analysis of

non-autoclaved aerated concrete containing FA and SF (air

and water cured) as structural lightweight concrete which

can be used in load-bearing structures.

Experimental

Portland cement type 1 (PC) was used in this study. Fly ash

(FA) was obtained from Mae Moh power plant in Lam-

pang, Thailand, and undensified silica fume (SF), grade

920-U, was produced by Elkem (Norway). The non-auto-

claved aerated PC–FA–SF concrete studied was produced

from cement mortar. River sand, with specific gravity of

2.65 and maximum size of 5 mm, was used as a fine ag-

gregate in concrete, according to ASTM standard C33 [34].

Aluminum powder (Al), a pore-forming agent that reacts

with calcium hydroxide to form hydrogen gas, was sup-

plied by HiMedia Laboratories, India. The reaction

between aluminum with calcium hydroxide to form hy-

drogen gas is shown in the following equation [35].

2Al þ 3Ca OHð Þ2þ6H2O ! 3CaO � Al2O3 � 6H2O þ 3H2

Alumimium powder þ Hydrated lime

! Tricalcium hydrate þ Hydrogen

The superplasticizer (SP) used was obtained from Sika

(Thailand). The chemical composition of materials used in

this research was obtained by means of X-ray fluorescence

and is given in Table 1. FA and SF were used in different

compositions by mass as a partial replacement for PC. The

replacement levels of FA were 20, 25 and 30 %, while SF

was 5 and 10 % by mass of PC. Al was added at 0.2 % by

mass of binders. The mix proportions are summarized in

Table 2. Mixes were designed to have a constant water-to-

binder ratio of 0.475. The flow of mortar at 110 ± 5 mm

was kept constant by the use of a superplasticizer. A fine

aggregate-to-binder ratio of 2.5 was used. Concrete mixes

were mixed and cast in 50 9 50 9 50 mm cubic molds for

compressive strength testing and 125 9 /25 mm cylin-

drical molds for thermal conductivity testing. Concrete

mixes were compacted using a vibration table. The samples

were stored in molds at room temperature for 24 h.

Afterward, the samples were demolded and then cured in

air and water at room temperature for 3, 7 and 28 days.

After the curing process, three specimens were investi-

gated for compressive strength, according to ASTM stan-

dard C109 [36]. In addition, three samples were

investigated for density and three samples were investi-

gated for thermal conductivity in accordance with ASTM

D5930-01 [37]. The microstructure of samples was inves-

tigated using scanning electron microscopy (SEM;

JEOLJSM-840A, Japan). Thermogravimetry (TG) and

Table 1 Chemical composition and physical properties of raw

materials

PC FA SF

Chemical composition/%

SiO2 20.64 45.37 93.543

Al2O3 4.848 20.65 0.556

CaO 63.62 10.43 1.128

Fe2O3 3.17 12.31 0.172

MgO 1.137 2.127 0.747

Na2O 0.51 1.331 0.137

K2O 0.812 1.496 1.047

P2O5 0.32 0.245 0.531

TiO2 0.213 0.517 0.002

SO3 2.753 2.526 1.006

Specific gravity 3.15 2.1 2.2

Average particle size/lm 8.31 5.22 0.1

12 C. Narattha et al.

123

derivative thermogravimetric (DTG) were carried out using

a Mettler–Toledo TGA/SDTA 851e. The samples were

heated from 30 to 1000 �C at a heating rate of 10 �C min-1

under nitrogen atmosphere. The nitrogen flowing rate is

60.0 mL min-1. Moreover, phase identification of the

samples was performed using an X-ray diffractometer

(XRD; Rigaku MiniFlex II, Japan). For XRD and TG

studies, samples were taken from the middle part of the

broken paste cube (&15 mm in size) after compressive

strength test. The samples were immersed in acetone for

3 days to stop hydration reaction after the curing process.

The samples were dried in an oven at 60 �C for 24 h. The

samples were placed in sealed bags. Powder for XRD and

TG studies was prepared from dried sample after grinding

and sifting through a No. 200 mesh sieve. For XRD tests,

room-temperature XRD measurements using CuKa ra-

diation (k = 1.5418 A) operating at 40 kV, 35 mA were

taken. XRD measurements were taken for 2h between 5�and 60� with increment of 0.01� step-1 and scan speed of

3 s step-1. The diffraction peaks were analyzed using

JCPDS software. For SEM test, low vacuum mode was

used. Sample was placed on brass stub with carbon tape.

Sample was dried using infrared light for 5 min. Then,

sample was coated with gold approximately 20 A thick

using blazer sputtering coater. Micrograph was recorded

with magnifications of 1009 at 15 kV. All SEM, XRD and

TG characterizations were carried out on 28-day samples

as standard test age.

Results and discussion

Compressive strength

The compressive strength results after curing for 3, 7 and

28 days are shown in Fig. 1. It was found that in binary

phase (30FA mix), the compressive strength of aerated

concrete with FA was lower than that of the PC control.

The FA content contributes to slow pozzolanic reaction

rate, which is inadequate for pre-28-day compressive

strength development [12, 13, 38]. The compressive

strength of aerated concrete containing FA can be im-

proved by adding SF as a cement replacement. The

20FA10SF mix had compressive strength values similar to

the PC control mix after 28 days of curing. SF particles are

finer and their pozzolanic reactivity is higher compared

with FA particles. SiO2 in SF reacted with Ca(OH)2 from

hydration reaction to give C–S–H formation to supplement

bonding strength. Furthermore, interior microstructure

pores were filled with SF [39]. Thus, aerated concrete with

both FA and SF has increased compressive strength [21,

22]. Nochaiya et al. [40] reported that 28-day compressive

strength of PC–FA–SF concrete was found to increase by

45 % of compressive strength at 28 days when compared

with PC–FA concrete. Guneyisi et al. [41] also reported

that the ternary blended cement with FA and SF, especially

in 15FA10SF, had slightly higher compressive strength

than binary blended cement with FA. Moreover, water

curing of aerated concrete resulted in higher compressive

strength than air curing, due to quicker and fuller hydration

induced by a sufficient water supply [42]. The major ce-

mentitious compounds in cements, belite (C2S:Ca2SiO4)

and alite (C3S:Ca3SiO5), had completely reacted with wa-

ter to form calcium silicate hydrates (C-S–H) [26]. Aerated

concrete was having compressive strength at 28 days in the

range of 13–23 MPa. These results can be used for mod-

erate-strength concrete and structural concrete [43].

Density

The density of non-autoclaved aerated concrete with FA

(30FA mix) was lower than that of the PC control mix

(Fig. 2). Also, the density of aerated concrete that con-

tained SF replacement was higher than that of the 30FA

mix and tended to increase with increased SF content. This

is because SF improves the pore structure; its small particle

size results in a filler effect, in which the SF particles

bridge the spaces between cement grains and the spaces

between cement grains and aggregate particles. SF also

reacts pozzolanically with Ca(OH)2 to produce a greater

solid volume of the C–S–H phase, leading to an additional

reduction in capillary porosity during hydration [44]. The

Table 2 Mix proportion of non-autoclaved aerated concrete

Mix w/b ratio Proportions/kg m-3 SP/% Al powder/%

PC Water Sand FA SF

PC 0.475 576 288 1440 – – 0.12 0.2

30FA 0.475 403 288 1368 173 – – 0.2

25FA5SF 0.475 403 288 1369 144 30 0.15 0.2

20FA10SF 0.475 403 288 1371 115 60 0.25 0.2

Thermogravimetry analysis, compressive strength and thermal conductivity tests of… 13

123

density of the 20FA10SF mix was higher than that of the

PC control mix. The density of aerated PC–FA–SF con-

crete cured in saturated limewater was higher than that of

air-cured specimens. This is due to hydration of cement can

occur only in water-filled capillary porosities [26]. There-

fore, the specimens under water curing were supplied with

sufficient water, leading to the continuous hydration of

specimens (with less porous structure) than air-cured spe-

cimens. These findings are in agreement with the test re-

sults for compressive strength, which increased with

increased SF content.

Thermal conductivity

The thermal conductivity results of water-cured and air-

cured aerated PC–FA–SF concrete are shown in Fig. 3. The

thermal conductivity of the 30FA mix was lower than the

PC control mix and had the lowest values for both water-

cured and air-cured specimens. The thermal conductivity

was increased when SF was added as a cement replacement

in ternary phase. The thermal conductivity values for

25FA5SF and 20FA10SF mixes were higher than the PC

control mix. Thermal conductivity increased with increas-

ing SF content as a result of the denser microstructure of

the cement paste matrix [38]. Albayrak et al. [45] reported

a decrease in thermal conductivity is relation to the de-

crease in density of autoclaved aerated concrete. Demir-

boga and Gul [46] also reported that a decrease in density

leads to a decrease in thermal conductivity. The thermal

conductivity of 25FA5SF and 20FA10SF mixes and bulk

density increased when replacing SF instead of FA [47].

Aluminum added in aerated concrete resulted in higher

porosities and lower thermal conductivity due to large

volume of air voids [43]. Moreover, thermal conductivity

values of water-cured specimens were higher compared

26

24

2220

18

16

14

12

10

8

6

4

20

26

24

2220

18

16

14

12

10

8

6

4

20

26

24

2220

18

16

14

12

10

8

6

4

20

Com

pres

sive

str

engt

h/M

Pa

Com

pres

sive

str

engt

h/M

Pa

Com

pres

sive

str

engt

h/M

Pa

30FA 25FA5SF 20FA10SF PC 30FA 25FA5SF 20FA10SF PC

30FA 25FA5SF 20FA10SF PC

Mixes Mixes

Mixes

WaterAir

WaterAir

WaterAir

(a) (b)

(c)

Fig. 1 Compressive strength of aerated PC–FA–SF concrete under water cured and air cured at a 3 days, b 7 days and c 28 days

14 C. Narattha et al.

123

with air-cured specimens. Water curing provided sufficient

moisture to maximize the hydration reaction which gives a

denser matrix resulting in a higher conductivity than air-

cured specimens [48].

Microstructure

The microstructure of aerated PC–FA–SF pastes can be

seen from scanning electron microscope (SEM) micro-

graphs of PC mixes cured for 28 days (Fig. 4). Water-

cured PC specimens (Fig. 4a) are noticeably denser than

air-cured specimens (Fig. 4b) and have smaller pore

sizes. This indicates that water curing aided hydration of

PC mixes by maintaining sufficiently moist conditions

[48]. This was because water is the most important factor

in the hydration reaction. Figure 5 shows SEM micro-

graphs of aerated PC–FA–SF pastes of 30FA mix and

25FA5SF mix after curing for 28 days. In general, larger

pore size and more pores can be seen in air-cured

samples than water-cured samples. This relates to the

higher density of water-cured samples compared to air-

cured samples. Some remaining unreacted FA particles

(&50–100 lm) can be detected (spherical shapes) in

Fig. 5, which indicates that some of the FA has already

reacted with Ca(OH)2, while the remaining of larger

([50 lm) unreacted FA particles represent an incomplete

reaction of FA. This confirms that fly ash is a slow-

reacting material. In contrast, silica fume particles would

have already reacted to form hydration products, as de-

tected by the DTG curve [20]. Water cured and air cured

of 25FA5SF pastes microstructure appear to be denser

than 30FA pastes.

XRD

The X-ray diffraction (XRD) patterns of aerated PC–FA–

SF pastes cured for 28 days are shown in Fig. 6. In all

mixes cured in water and air (Fig. 6a, b), peak of calcium

2000

1900

1800

1700

1600

1500

1400

1300

1200

Den

sity

/kg

m–3

2000

1900

1800

1700

1600

1500

1400

1300

1200

Den

sity

/kg

m–3

2000

1900

1800

1700

1600

1500

1400

1300

1200

Den

sity

/kg

m–3

30FA 25FA5SF 20FA10SF PCMixes

30FA 25FA5SF 20FA10SF PCMixes

30FA 25FA5SF 20FA10SF PCMixes

WaterAir

WaterAir

WaterAir

(a) (b)

(c)

Fig. 2 Density of aerated PC–FA–SF concrete under saturated limewater curing and air curing at a 3 days, b 7 days and c 28 days

Thermogravimetry analysis, compressive strength and thermal conductivity tests of… 15

123

hydroxide [Ca(OH)2] phase can be seen. In addition, the

peaks of belite (C2S:Ca2SiO4) and alite (C3S:Ca3SiO5)

phases (unhydrated products from Portland cement) were

detected. This study found that the intensity of Ca(OH)2 in

ternary aerated cement pastes was lower than in binary

aerated cement paste that incorporated only FA. The in-

tensity of Ca(OH)2 decreased with increasing SF content.

This indicated that SiO2 reacted with Ca(OH)2. Moreover,

the reaction between CO2 and Ca(OH)2 to form CaCO3 in

carbonation reaction when samples are exposed to the air is

another reason in the decreased Ca(OH)2 [49]. Thus,

Ca(OH)2 may also be partly reduced due to the reaction of

Ca(OH)2 with CO2 when samples are exposed briefly in air

during the curing and drying process.

0.60

0.55

0.50

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

The

rmal

con

duct

ivity

/W m

–1 K

–1

0.60

0.55

0.50

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

The

rmal

con

duct

ivity

/W m

–1 K

–1

0.60

0.55

0.50

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

The

rmal

con

duct

ivity

/W m

–1 K

–1

30FA 25FA5SF 20FA10SF PCMixes

30FA 25FA5SF 20FA10SF PCMixes

30FA 25FA5SF 20FA10SF PCMixes

WaterAir

WaterAir

WaterAir

(a) (b)

(c)

Fig. 3 Thermal conductivity of aerated PC–FA–SF concrete under saturated limewater curing and air curing at a 3 days, b 7 days and c 28 days

Fig. 4 SEM micrograph of PC mix cured at 28 days: a water cured and b air cured

16 C. Narattha et al.

123

Thermogravimetry

Thermogravimetry (TG) results of non-autoclaved aerated

PC–FA–SF pastes cured for 28 days are shown in Figs. 7

and 8. The results were plotted as derivative thermo-

gravimetric (DTG) to observe each phase, as shown in

Figs. 9 and 10. The main mass loss transition of all pastes

occurred in three temperature ranges. The first main mass

loss up to &350 �C was through dehydration of several

hydrate phases, such as calcium silicate hydrate (C–S–H),

ettringite and gehlenite (C2ASH8). The second main mass

loss from &350–500 �C was due to dehydroxylation of

calcium hydroxide [Ca(OH)2] [50]. The third main mass

loss from &500–1,000 �C occurred from decarbonation of

calcium carbonate (CaCO3) [51–53]. DTG results of all

pastes after 28 days showed the following detected phases:

ettringite at &79–83 �C, C–S–H at &108–119 �C,

C2ASH8 at &156–165 �C and Ca(OH)2 at &457–465 �C.

CaCO3 presented two dominant peaks at &680–713 �Cand &871–893 �C for water-cured pastes. For air-cured

pastes, the detected phases from DTG results are ettringite

at &70–93 �C, C–S–H at &108–121 �C, C2ASH8 at

&157–168 �C and Ca(OH)2 at &456–467 �C. CaCO3 also

presented two dominant peaks at &699–748 �C and

&879–905 �C. The first CaCO3 peak occurred due to de-

composition of ordinary calcite, and the second peak is

likely to be shifted due to difference in surface area, par-

ticle size, partial pressure of CO2 and impurities such as

magnesium which affected the decomposition of calcite

[54]. The higher intensity of the CaCO3 peak of the 30FA

Fig. 5 SEM micrograph of

non-autoclaved aerated PC–FA–

SF pastes: a 30FA mix water

cured, b 30FA mix air cured,

c 25FA5SF mix water cured,

d 25FA5SF mix air cured

10 20 30 40 50 60

2θ/°10 20 30 40 50 60

2θ/°

Ca(OH)2 C3S C2S

20FA10SF

25FA5SF

30FA

PC

20FA10SF

25FA5SF

30FA

PC

Inte

nsity

/a.u

.

Inte

nsity

/a.u

.

Ca(OH)2 C3S C2S(a) (b)

Fig. 6 XRD pattern of aerated cement pastes cured at 28 days: a water cured and b air cured

Thermogravimetry analysis, compressive strength and thermal conductivity tests of… 17

123

mix may be because the more FA content in the mix re-

sulted in greater porosity; thus, atmospheric CO2 could

diffuse more easily than in other mixes. Pandey and

Sharma [55] reported total porosity increased with the in-

crease in the addition of fly ash. The peak intensity of C–S–

H curves of mixes that contained SF was higher than of

mixes that contained only FA. The C–S–H curves in-

creased with increased SF content due to the higher poz-

zolanic reaction rate when SF was added. This result

agreed with the increase in compressive strength of aerated

concrete.

The mass loss of non-autoclaved aerated PC–FA–SF

pastes cured for 28 days is shown in Table 3. The mass

loss of C–S–H ? ettringite ? C2ASH8 phases for all

pastes increased with increased SF content. The mass loss

for PC–FA–SF pastes was higher than for PC–FA pastes.

This was because SF improved the pozzolanic activity of

aerated cement pastes. The mass loss of the Ca(OH)2 phase

of pastes tended to decrease when SF was added as a ce-

ment replacement because Ca(OH)2 was consumed by FA

and further by SF in the pozzolanic reaction. In all cases,

when comparing water- and air-cured specimens, the mass

loss of C–S–H ? ettringite ? C2ASH8 and Ca(OH)2 for

water-cured mixes was higher than for air-cured mixes.

The mass loss of CaCO3 for air-cured mixes was higher

than for water-cured mixes because air-cured samples were

directly exposed to carbonation during curing while water-

cured samples were not exposed during curing but only

during the drying process. Moreover, there are more in-

terconnected channels and large pores size in air-cured

samples which resulted to more diffusion of CO2 which

increase carbonation [56]. The mass loss of CaCO3 for

30FA air-cured mixes had the highest value. This was

because part of Ca(OH)2 phase being used by the car-

bonation reaction with atmospheric CO2 (exposure during

storage), forming CaCO3 [57, 58].

100

95

90

85

80

75

70

Mas

s/%

0 200 400 600 800 1000

Temperature/°C

20FA10SF25FA5SF30FA

Fig. 7 TG curves of aerated PC–FA–SF pastes water curing at

28 days

100

95

90

85

80

75

70

Mas

s/%

0 200 400 600 800 1000

Temperature/°C

20FA10SF25FA5SF30FA

Fig. 8 TG curves of aerated PC–FA–SF pastes air curing at 28 days

0 200 400 600 800 1000

Temperature/°C

20FA10SF25FA5SF30FA

0.0000

–0.0002

–0.0004

–0.0006

–0.0008

–0.0010

–0.0012

–0.0014

–0.0016

Der

ivat

ive

mas

s lo

ss/°

C–1

Fig. 9 DTG curves of aerated PC–FA–SF pastes water curing at

28 days

0 200 400 600 800 1000

Temperature/°C

20FA10SF25FA5SF30FA

0.0000

–0.0002

–0.0004

–0.0006

–0.0008

–0.0010

–0.0012

–0.0014

–0.0016

Der

ivat

ive

mas

s lo

ss/°

C–1

Fig. 10 DTG curves of aerated PC–FA–SF pastes air curing at

28 days

18 C. Narattha et al.

123

The compressive strength and density of non-auto-

claved aerated PC–FA–SF concrete were

&1765–1869 kg m-3 and &13–23 MPa, respectively.

These results of products were sufficient to ACI Com-

mittee 213 standard [43] for use as moderate-strength

concrete and structural concrete. Thus, aluminum-added

non-autoclaved aerated PC–FA–SF concrete can be con-

veniently cast into any desirable size using less energy

than autoclaved method. This study focussed on the me-

chanical and thermal properties of non-autoclaved aerated

PC–FA–SF concrete. Durability properties such as freeze–

thaw of non-autoclaved aerated PC–FA–SF concrete are

recommended for future studies.

Conclusions

Aluminum-added aerated concrete was cast and tested

(density &1765–1869 kg m-3) having compressive

strength at 28 days in the range of 13–23 MPa. These

results can be used for moderate-strength concrete and

structural concrete. Non-autoclaved aerated PC–FA–SF

concrete has considerable advantages such as reducing the

cost of products, accommodating the production process

and low energy consumption compared with autoclaved

aerated concrete. The compressive strength and thermal

conductivity decrease with the use of binary FA cement

but were found to increase when SF was incorporated

with FA for both water- and air-cured samples. This is

due to the increase in the hydrated product (C–S–H) and

less Ca(OH)2 in the ternary (PC–FA–SF) mixes as de-

tected by TG and XRD, but more hydration products were

formed in water-cured samples. The pozzolanic reaction

was thus affected by the air curing method where TG

results showed that C–S–H, ettringite and C2ASH8 in

water-cured samples were higher than in air-cured sam-

ples, while Ca(OH)2 and CaCO3 in water-cured samples

were lower than in air-cured samples.

Acknowledgements The authors gratefully acknowledged the Of-

fice of the Higher Education Commission (OHEC), the Thailand

Research Fund (TRF), Department of Physics and Materials Science,

Faculty of Science, Chiang Mai University, and the Graduate School

of Chiang Mai University.

References

1. World Business Council for Sustainable Development. Cement

Technology Roadmap 2009, Carbon Emissions reductions up to

2050: OECD/IEA and the World Business Council for Sustain-

able Development; 2009.

2. Damtoft J, Lukasik J, Herfort D, Sorrentino D, Gartner E. Sus-

tainable development and climate change initiatives. Cem Concr

Res. 2008;38(2):115–27.

3. Nehdi M, Mindess S, Aitcin PC. Use of ground lime stone in

concrete: a new look. Bull Res J. 1995;43(4):245.

4. Zeng Q, Li K. Reaction and microstructure of cement–fly ash

system. Mater Struct. 2014. doi:10.1617/s11527-014-0266-y.

5. Ben Haha M, De Weerdt K, Lothenbach B. Quantification of

degree of reaction of fly ash. Cem Concr Res. 2010;40:1620–9.

6. Papadakis VJ. Effect of fly ash on Portland cement systems. Part

I: low calcium fly ash. Cem Concr Res. 1999;29:1727–36.

7. Lam L, Wong YL, Poon CS. Degree of hydration and gel/space

ratio of high-volume fly ash/cement systems. Cem Concr Res.

2000;30:747–56.

8. Han SH, Kim JK, Park YD. Prediction of compressive strength of

fly ash concrete. Cem Concr Res. 2003;33:965–71.

9. Willinska I, Pacewska B. Calorimeter and thermal analysis

studies on the influence of waste aluminosilicate catalyst on the

hydration of fly ash–cement paste. J Therm Anal Calorim. 2014;

116:689–97.

10. Elsayed AA. Influence of silica fume, fly ash, super pozz, and

high slag cement on water permeability and strength of concrete.

Concr Res Lett. 2012;3(4):528–40.

11. Alasali MM, Malhotra VM. Role of concrete incorporating high

volume of fly ash in controlling expansion due to alkali–aggre-

gate reaction. ACI Mater J. 1991;88(2):159–63.

12. Carette G, Bilodeau A, Chevrier RL, Malhotra VM. Mechanical

properties of concrete incorporating high volume fly ash from

sources in the US. ACI Mater J. 1993;90(6):535–44.

13. Naik TR, Singh SS, Ramme BW. Mechanical properties and

durability of concrete made with blend fly ash. ACI Mater J.

1998;95(4):454–62.

14. Baert G, Horste S, Schutter G, Belie N. Reactivity of fly ash in

cement paste studied by means of thermogravimetry and

isothermal calorimetry. J Therm Anal Calorim. 2008;94:485–92.

15. Vessalas K, Thomas PS, Ray AS, Guerbois JP, Joyce P, Haggman

J. Pozzolanic reactivity of the supplementary cementitious ma-

terials pitchstone fines by thermogravimetric analysis. J Therm

Anal Calorim. 2009;97:71–6.

16. Malhotra VM, Metha PK. Pozzolanic and cementitious materials,

advances in concrete technology. London: Gordon and Breach;

1996.

17. Chung DDL. Review improving cement-based materials by using

silica fume. J Mater Sci. 2002;37:673–82.

18. Rong ZD, Sun W, Xiao HJ, Wang W. Effect of silica fume and

fly ash on hydration and microstructure evolution of cement

based composite at low water–binder ratio. Constr Build Mater.

2014;51:446–50.

Table 3 Mass loss of non-autoclaved aerated PC–FA–SF pastes cured at 28 days

Mixes Mass loss/%

C–S–H ? ettringite ? C2ASH8 Ca(OH)2 CaCO3

Water cured Air cured Water cured Air cured Water cured Air cured

30FA 13.2679 9.9143 4.8148 4.8040 5.3794 7.2777

25FA5SF 15.0997 12.8886 4.8060 4.3238 5.1594 6.0018

20FA10SF 16.1158 15.3355 4.1288 3.7519 5.4410 5.9646

Thermogravimetry analysis, compressive strength and thermal conductivity tests of… 19

123

19. Korpa A, Kowald T, Trettin R. Hydration behavior, structure and

morphology of hydration phases in advanced cement-based sys-

tems containing micro and nanoscale pozzolanic additives. Cem

Concr Res. 2008;38:955–65.

20. Kohno K, Aihara F, Ohna K. Relative durability properties and

strength of mortars containing finely ground silica and silica

fume. In: Malhotra VM, editor. Proceedings of the third congress

on fly ash, silica fume, slag and national pozzolans in concrete.

ACI SP 114. vol 1; 1989. p. 815.

21. Chaipanich A, Nochaiya T. Thermal analysis and microstructure

of Portland cement–fly ash–silica fume paste. J Therm Anal

Calorim. 2010;99:487–93.

22. Yazici H. The effect of silica fume and high-volume class c fly

ash on mechanical properties, chloride penetration and freeze–

thaw resistance of self-compact concrete. Constr Build Mater.

2008;22:456–62.

23. Valipour M, Pargar F, Shekarchi M, Khani S. Comparing the

natural pozzolan, zeolite, to metakaolin and silica fume in term of

their effect on the durability characteristics of concrete: a

laboratory study. Constr Build Mater. 2013;41:879–88.

24. Li Y, Sun D, Wu X, Wang A, Xu W, Deng M. Drying shrinkage

and compressive strength of blended cement pastes with fly ash

and silica fume. Adv Mater Res. 2012;535–537:1735–8.

25. Lai WL, Tsang WF. Characterization of pore systems of air/

water-cured concrete using ground penetrating radar (GPR)

through continuous water injection. Constr Build Mater. 2008;22:

250–6.

26. Neville AM. Properties of concrete. London: Longman; 1995.

27. Narayanan N, Ramamurthy K. Structure and properties of aerated

concrete: a review. Cem Concr Compos. 2000;22:321–9.

28. Baspinar MS, Demir I, Kahraman E, Gorhan G. Utilization of fly

ash together with silica fume in autoclaved aerated concrete

production. J Civ Eng. 2014;18(1):47–52.

29. Just A, Middendorf B. Microstructure of high-strength foam

concrete. Mater Charact. 2009;60:741–8.

30. Wongkeo W, Thongsanitgarn P, Chaipanich A. Compressive

strength of binary and ternary of blended cement mortars con-

taining fly ash and silica fume under autoclave curing. Adv Mater

Res. 2012;343:316–21.

31. Wongkeo W, Thongsanitgarn P, Chindaprasirt P, Chaipanich A.

Thermogravimetry of ternary cement blends, effect of different

curing methods. J Therm Anal Calorim. 2013;113:1079–90.

32. Holt E, Raivio P. Use of gasification residues in aerated auto-

claved concrete. Cem Concr Res. 2005;35:796–802.

33. Kunchariyakun K, Asavapisit S, Sombatsompop K. Properties of

autoclaved aerated concrete incorporating rice husk ash as partial

replacement for fine aggregate. Cem Concr Compos. 2015;55:

11–6.

34. ASTM C33-02. Standard specification for concrete aggregates.

American Society for Testing and Materials; 2002.

35. Hamad AJ. Materials, production, properties and application of

aerated lightweight concrete: review. Int J Mater Sci Eng. 2014;

2:152–7.

36. ASTM C109. Standard test method for compressive strength of

hydraulic cement mortars (Using 2-in. or [50-mm] cube speci-

mens). American Society for Testing and Materials; 2002.

37. ASTM D 5930-01. Standard test method for thermal conductivity

of plastics by means of transient line source technique. American

Society for Testing and Materials; 2002.

38. Wongkeo W, Chaipanich A. Compressive, microstructure and

thermal analysis of autoclave and air cured structural lightweight

concrete made with coal bottom ash and silica fume. Mater Sci

Eng A. 2010;567:3676–84.

39. Shih JY, Chang TP, Hsiao TC. Effect of nanosilica on charac-

terization of Portland cement composite. Mater Sci Eng A.

2006;424:266–74.

40. Nochaiya T, Wongkeo W, Chaipanich A. Utilization of fly ash

with silica fume and properties of Portland cement–fly ash–silica

fume concrete. Fuel. 2010;89:768–74.

41. Guneyisi E, Gesoglu M, Booya E, Mermerdas K. Strength and

permeability properties of self-compacting concrete with cold

bonded fly ash lightweight aggregate. Constr Build Mater.

2015;74:17–24.

42. Yuanming S, Jueshi Q, Jinxiang L, Bo W, Yingchum W. Effect of

conditions on the hydration and performance of CFBC ash cemen-

titious system. Mater Sci Ed. 2014. doi:10.1007/s11595-014-0873-z.

43. Mindess S, Young JF, Darwin D. Concrete. 2nd ed. London:

Pearson Education; 2003.

44. Bentz DP, Stutzman PE. Evolution of porosity and calcium hy-

droxide in laboratory concretes containing silica fume. Cem

Concr Res. 1994;24:1044–50.

45. Albayrak M, Yorukoglu A, Karahan S, Atlihan S, Aruntas HY,

Girgin I. Influence of zeolite additive on properties of autoclaved

aerated concrete. Build Environ. 2007;42:3161–5.

46. Demirboga R, Gul R. The effect of expanded perlite aggregate,

silica fume, fly ash on thermal conductivity of lightweight con-

crete. Cem Concr Res. 2003;33:723–7.

47. Demirboga R. Influence of mineral admixtures on thermal con-

ductivity and compressive strength of mortar. Energy Build.

2003;35:189–92.

48. Zongjin L. Advanced concrete technology. New York: Wiley; 2011.

49. Galan I, Andrade C, Castellote M. Thermogravimetrical analysis

for monitoring carbonation of cementitious materials, uptake of

CO2 and depending in C–S–H knowledge. J Therm Anal Calorim.

2012;110:309–19.

50. Hamidi M, Kacimi L, Cyr M, Clastres P. Evaluation and im-

provement of pozzolanic activity of andesite for its use in eco-

efficient cement. Constr Build Mater. 2013;47:1268–77.

51. Alarcon Ruiz L, Platret G, Massieub E, Ehrlacher A. The use of

thermal analysis in assessing the effect of temperature on a ce-

ment pastes. Cem Concr Res. 2005;35:609–13.

52. Neves A Jr, ToledoFilho RD. Early stage hydration of high initial

strength Portland cement Part I. Thermogravimetric analysis on

calcine mass basis. J Therm Anal Calorim. 2012;108:725–31.

53. Bai J, Chaipanich A, Kinuthia JM, Farrel MO, Sabir BB, Wild S,

et al. Compressive strength and hydration of wastepaper sludge

ash–ground granulated blast furnace slag blended pastes. Cem

Concr Res. 2003;33:1189–202.

54. Mitchell LD, Margeson JC. The effects of solvents on C–S–H as

determined by thermal analysis. J Therm Anal Calorim. 2006;

86(3):591–4.

55. Pandey SP, Sharma RL. The influence of mineral additives on the

strength and porosity of OPC mortar. Cem Concr Res. 2000;30:19–23.

56. Lo Y, Lee HM. Curing on carbonation of concrete using a phe-

nolphthalein indicator and fourier-transform infrared spec-

troscopy. Build Environ. 2002;37:507–14.

57. Antoni M, Rossen J, Martirena F, Scrivener K. Cement substi-

tution by a combination of metakaolin and limestone. Cem Concr

Res. 2012;42:1579–89.

58. Dweck J, Ferreira da Silva PF, Buchler PM, Cartledge FK. Study

by thermogravimetry of the evolution of ettringite phase during

type II Portland cement hydration. J Therm Anal Calorim. 2002;

69:179–86.

20 C. Narattha et al.

123