Construction and Building Materials 172 (2018) 41–51

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Long term durability properties of concrete modified with metakaolin

and polymer admixture
Adel Al Menhosh a,c, Yu Wang a,⇑, Yan Wang b, Levingshan Augusthus-Nelson a
a
School of Computing, Science & Engineering, University of Salford, Manchester M5 4WT, UK
b
School of Civil Engineering, Chongqing Jiaotong University, Chongqing 400074, China
c
Department of Civil Engineering, University of Basrah, Iraq

h i g h l i g h t s

A review on the principles using metakaolin and polymer to improve concrete properties.

Long-term mechanical and durable properties of concrete using both metakaolin and polymer.

The optimum mix and the complement mechanisms using both metakaolin and polymer.

A guidance for concrete design for the application in severe environmental conditions.

a r t i c l e i n f o a b s t r a c t

Article history: Previous studies show that both metakaolin (MK) and polymer can respectively improve certain mechan-
Received 16 November 2017 ical and durability properties of concrete. Also, recent studies show that a combination of MK and poly-
Received in revised form 13 March 2018 mer further enhances the mechanical properties by complement of each other. However, the knowledge
Accepted 21 March 2018
of the effect on durability, a critical governing factor of concrete for the applications in extreme environ-
ments such as sewage, off-shore and bridge structures, has not been well established yet. This paper
reports on a comprehensive study of the effect of metakaolin as a supplementary cementitious material
Keywords:
together with polymer as admixture on the durability of concrete at relatively old ages. The results con-
Metakaolin
Polymer
firm that replacing Portland cement with 15% metakaolin and an additional 5% polymer (by weight) pro-
High Performance Concrete vide the optimum improvement for Portland cement concrete on both mechanical properties and
Durability durability.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction hydrate (CSH) gel to give high Al uptake and low Ca content in a
new gel formation known as CASH, which has a low Ca/(Al + Si)
Using mineral supplementary cementitious materials (SCM), ratio but a high Al/Ca ratio [40].
such as fly ash (FA), silica fume (SF) and thermally activated kaolin Previous research showed that a 20% replacement of cement
(also known as metakaolin (MK)), as additives has already been using MK resulted in a substantial 50% increase of the compres-
proved effective to improve properties of concrete [26]. MK sive strength of mortar [29], and the concrete using MK additive
requires less energy to produce compared to cement [39,40], displayed a lower water sorptivity compared to that using silica
which, in recent years, has attracted more and more interest in fume [22]. Recently, Pouhet and Cyr [38] studied the pore solution
the use for the SCM [2,41] because of the environmental concern carbonation of MK-based geopolymer and found that the pH
and the decreasing supply capacity of fly ash and silica fume decreased rapidly in the first few days when the normal concrete
[40]. The MK product has predominant alumina (Al2O3) and silica was exposed to natural CO2 conditions. Moreover, a high CO2 con-
(SiO2) composition, which have an active pozzolanic nature [4]. tent or a relatively high environmental temperature led to dura-
The pozzolanic reaction of MK with portlandite (Ca(OH)2) will bility issues when the pH was lower than 10. However, the
result in significant compositional changes of calcium silicate pozzolanic nature of MK increased the pH and kept it above 12
even after one year, indicating a minimum carbonation inside
the concrete. Another study [28] showed that self-compacting
concrete (SCC) with a high MK content (up to 30%) exhibited a sig-
⇑ Corresponding author.
nificant resistance to chloride ion penetration. For acid attack
E-mail address: [email protected] (Y. Wang).

https://doi.org/10.1016/j.conbuildmat.2018.03.215
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
42 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51

resistance, the same SCC with 5 and 10% of MK showed the lowest Table 1
weight loss after 12 weeks immersed in 5% HCl and 5% H2SO4 Typical composition of cement and metakaolin.

solutions, respectively. Contrary to these advantages, it was CEM II/A-LL Cement (BS EN Metakaolin [4]
also found, however, that the MK significantly reduces the work- 197-1:2011 [13])
ability of concrete and thus more water is required to improve Component Values % Standard Values %
the workability [4]. However, the additional water to improve Al2O3 4.19 3–5% 40.18
the workability would lead to durability issues, causing Fe2O3 2.75 2.0–3.5% 1.23
aggregate segregation, excess voids and increased porosity of CaO 65.00 60–70% 2.0
the concrete. SO3 3.19 Less than 3.5% 0.0
MgO 0.86 0.5–1.5% 0.12
Polymers, such as styrene-butadiene rubber (SBR) latex and
Na2O 0.14 Less than 0.75% 0.08
polyvinyl acetate (PVA) emulsion have been commonly used as K2O 0.51 – 0.53
admixtures in concrete practice [6,30]. Polymer admixtures are SiO2 16.19 15–25% 51.52
known to not only increase the workability but also modify the TiO2 – – 2.27
Loss on ignition (L.O.I) – – 2.01
physical properties of cement pastes by reducing macro voids
and improving the bond strength of the polymer cement mortars
to aggregates. For example, the mortar of SBR showed improve-
ment in chloride penetration resistance along with general ionic
permeability. SBR also slightly reduced portlandite content and Table 2
mitigated the carbonation process [43]. The mixtures designed [3].
The nature of the interactions between the polymers and the Concrete Mixtures
Portland and aluminous cements is significantly different. For
MK/Cementitious Polymer/Cementitious Water/
Portland cement both SBR and PVA were found to retard the Binder (MK/C) %* Binder (P/C) % Cementitious
hydration rate to some extent, but SBR appeared to have very Binder ratio
little chemical interaction with ordinary Portland cement W/C
(OPC) while PVA is fully hydrolysed [6]. A recent study on 0 0/2.5/5/7.5 0.35/0.38/
polymer-modified pervious concrete also found that both SBR 10 0.40/0.45/
and PVA polymers retarded the hydration reactions of cement 15 0.50
20
particles and thus improved mechanical resistance and durabil-
30
ity at prolonged curing time, for which PVA showed a better 40
performance, but SBR showed no increase of the concrete stiff-
Composition of Polymer Admixture
ness [21]. For aluminous cement, SBR showed very little effect
on the rate of hydration, but PVA was partially hydrolysed. Component Percentage %

PVA, when added to aluminous cements, produces a so-called SBR 0 20 40 50 60 80 100

macro-defect-free (MDF) matrix of superior strength and frac- PVA 100 80 60 50 40 20 0

ture toughness [6]. *

% by weight, cementitious binder = cement + MK.
A literature study shows that the MK and polymer comple-
ment each other in order to improve the mechanical and dura-
bility properties. A study on Portland cement concrete using
polymer, MK and FA showed a significant effect on the compres- the trade name of MetaStar 501, were used in the study. The mate-
sive strength, the flexural strength and the modulus of elasticity rial compositions are referenced in Table 1.
[31]. However, it is noted that the knowledge of the durability The polymer additive was styrene butadiene rubber (SBR) latex,
properties of concrete modified with polymer and MK have not poly vinyl acetate (PVA) emulsion and their mixtures. Normal
been well established yet [1]. To meet the high-performance sands were used for the fine aggregates while crushed limestone
requirement for sewage and off-shore applications, where the gravels were used for the coarse aggregates with a maximum size
durability of concrete governs the use of concrete, a series of of 10 mm. The particle size distributions of the aggregates [3] fol-
investigations have been conducted on the combined effect of low the requirements in BS 882:1992 [17].
the use of MK and a polymer mixture together on the mechan-
ical properties and durability of the modified concrete. A previ- 2.2. Mixture design
ous publication has reported a study of the conventional
mechanical properties [3]. This paper at first gives a brief review Various combinations of the MK and polymers as listed in
on the major findings in the previous work. After then, it reports Table 2 were tested in a previous study on conventional mechani-
a followed on experimental investigation on the long term dura- cal properties [3]. An optimum proportion was derived based on
bility properties of the optimum mixture identified in the previ- the mechanical properties of the modified concrete. To establish
ous study on mechanical properties. The durability of the a baseline, a mass ratio of 1:1.5:3 for cement:sand:gravel was con-
optimum mix has been compared with other three benchmarks sidered as a control mix.
to understand the effects of the MK, polymer and their
combination.
3. Conventional mechanical properties and the optimum mix

2. Mixtures All the mixes in the Table 2 were tested for their workability
and the mechanical properties at the ages of 7 and 28 days.
2.1. Raw materials Fig. 1 shows that MK significantly reduces the workability of
the modified mixes. The mix of 10% MK/C ratio has a slump value
Portland limestone cement, CEM II/A-LL [13], supplied by much lower than the targeted range of 75 mm to 100 mm even at
Lafarge cement UK LTD under the trade name of Mastercrete, a high W/C ratio of 0.5. Fig. 2 shows the cubic compressive
and a premium metakaolin, produced by IMERYS group under strengths of the MK modified concretes at the age of 7 and 28
 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51 43

160 days for the W/C of 0.45. It shows that the MK/C ratio in the
W/C=0.35 range of 15–20% presents the maximum compressive strength
140
W/C=0.38 at the two ages.
Mean Slump, mm

120
W/C=0.40 Figs. 3–5 show the effect of two polymers and their mixtures
100
W/C=0.45 on their modified concretes. It can be seen that when the polymer
80 W/C=0.50 to cementitious binder ratio (P/C) is more than 5%, the strength of
60 the modified concrete starts to deteriorate considerably. Mean-
40 while, when W/C ratio is more than 0.45, the deterioration on
20
concrete strength accelerates using polymer. Fig. 5 shows that
the polymer mixture of 80% SBR and 20% PVA at the 5% P/C
0
0 10 15 20 and 0.45 W/C presents the highest improvement on concrete
MK/C, % strength. Fig. 6 shows that the polymer mixture of 80% SBR and
20% PVA gives the modified mixture a slump value of about 82
Fig. 1. Slump values of concrete mixes modified with various percentage of mm at 0.45 W/C.
metakaolin and water to cement ratios.
Fig. 7 compares the 28 days compressive strength of mix-
tures using the optimum polymer mixture at 5% P/C and MK
at 15% MK/C separately and both together. Fig. 8 shows the
50
7 days effect of three different curing methods on the compressive
Compressive Strength, MPa

28 days
strength of the mixtures using 15% MK/C and varied P/C ratios.
40
All the results have suggested that using 15% MK/C, 5% P/C of a
polymer mixture of 80% SBR and 20% PVA, 0.45 W/C and moist
30
curing gives the modified concrete an optimum mechanical
properties.
20

10 4. Long-term properties of the optimum concrete

0 This paper focuses on the long-term durability properties of

0 10 15 20 30 40 the optimum mixture identified in Section 3, and compares it
MK/C, % with three other representative benchmark mixtures. All the
four mixtures are listed in Table 3. For each data point, three
Fig. 2. Compressive strength of concrete modified with various percentage of samples were tested and their average value is presented as
metakaolin and water to cement ratio of 0.45.
the result.

4.1. Mechanical properties

35
Compressive Strength, MPa

30 4.1.1. Compressive strength

Long-term compressive strengths of concrete cubes [9] of
25
these four mixtures up to 545 days are shown in Fig. 9. It
20 can be seen that mixtures 2 and 3 present a significant
W/C=0.35
15 improvement on the control mix 1 in the long term with about
W/C=0.40 16% increase of the compressive strength at the age of 545
10 W/C=0.45 days. The results indicate that MK has a considerable effect
5 W/C=0.50 on strength increase, but the polymer at 5% P/C has little
influence on this property.
0
0.0 2.5 5.0 7.5
SBR/C, % 4.1.2. Splitting and flexural tensile strength
Fig. 10 shows the splitting and flexural strengths at four ages up
Fig. 3. Compressive strength of the concrete modified using various percentage of to 180 days. The tests were conducted according to BS EN 12390-6:
SBR at 28 days.
(2009) [11] and BS EN 12390-5:(2009) [10], respectively. The
results show that both the splitting and flexural strengths devel-
oped with age. Either using polymer (5% P/C) or MK (15% M/C)
35
increases the tensile strength. The combination of 15% MK/C and
Compressive Strength, MPa

30 5% P/C resulted in the highest splitting and flexural strengths at

25 almost all ages.
20
W/C=0.35 4.1.3. Young’s modulus
15 Fig. 11 shows the modulus of elasticity at the age of 28 days.
W/C=0.40
10 The cast concrete cylinders (150 mm in diameter and 300 mm
W/C=0.45 in height) were at first moist cured for 28 days according to
5
W/C=0.50 [15] Part -121: (1983). It can be seen that using MK has
0 increased the Young’s modulus of elasticity while polymer
0.0 2.5 5.0 7.5
shows the opposite effect. The mix of 5% P/C and 15% MK/C
PVA/C, %
shows a similar result as that of 5% P/C only. This suggests that
Fig. 4. Compressive strength of the concrete modified using various percentage of the influence of MK on modulus in the presence of polymer has
PVA at 28 days. been minimised.
44 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51

40

Compressive Strength, MPa

30

20

10
SBR=0% SBR=100% SBR=80% SBR=60% SBR=50% SBR=40% SBR=20% SBR=0%
PVA= 0% PVA=0% PVA=20% PVA=40% PVA=50% PVA=60% PVA=80% PVA=100%
Combination of SBR and PVA

Fig. 5. Compressive strength of concrete modified with 5% of polymer (various combination between SBR and PVA) and water to cement ratio of 0.45 at 28 days.

160

Compressive Strength, MPa

W/C=0.35
140
W/C=0.38
120 W/C=0.40
Mean Slump, mm

100 W/C=0.45
W/C=0.50
80 Dry curing
Wet curing
60
Moist curing
40

20
P/C %
0
0 2.5 5 7.5 Fig. 8. Effect of curing methods on compressive strength of modified concrete at 28
P/C, % (80% SBR and 20% PVA) days using 0.45 W/C ratio and 15% metakaolin.

Fig. 6. Slump values of concrete mixes modified with various percentage of

polymer and 15% of metakaolin.
Table 3
Mixtures studied in this paper.

Mixtures MK/C%* P/C%* Water to cement ratio W/C

4.2. Durability 1 (Control) 0 0 0.45
2 15 0 0.45
4.2.1. Drying shrinkage 3 15 5 0.45
4 0 5 0.45
Drying shrinkage is an important characteristic of concrete
*
[23], which affects the long-term mechanical properties and % by the weight of cementitious binder (cement + MK).
durability properties of structures [14,24]. Concrete prisms
(100  100  400 mm) were cast and moist cured for 7 days.
Thereafter they were stored in open atmosphere and their MK and polymer respectively reduce the drying shrinkage.
dimensional changes along the length of the prisms were However, the optimum mix using both of them shows the lowest
monitored and recorded up to 365 days. Fig. 12 shows that both drying shrinkage at all ages.
Compressive Strength, MPa

50

40

30

20

10

0
P/C= 0% P/C= 0% P/C=5% P/C= 5% P/C= 5% P/C=5% P/C=5%
MK/C= 0% MK/C= 0% MK/C=15% MK/C= 15% MK/C= 15% MK/C=15% MK/C=15%
W/C=0.35 W/C=0.45 W/C=0.35 W/C= 0.38 W/C= 0.40 W/C=0.45 W/C=0.50
Mixing proportions
Fig. 7. Compressive strength of concrete modified with various combination of polymer, metakaolin and water to cement ratio at 28 days.
 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51 45

40000

Modulas of Elasticity, MPa

30000

20000

10000

0
P/C=0% P/C=0% P/C=5% P/C=5%
MK/C=0% MK/C=15% MK/C=15% MK/C=0%
Mix Proportion %

Fig. 11. Young’s modulus of elasticity at 28 days.

Fig. 9. Relationships between the compressive strength and the age of concrete.

Fig. 12. Drying shrinkage at age up to 365 days.

(a) Splitting tensile test

100 ± 6mm

Plastic sheet

Sealing material
2 ± 1mm
50 ± 3mm

Specimen
Pan

Specimen
supports Water

Fig. 13. Standard test method for measurement of rate of absorption of water by
hydraulic-cement concretes [7].

mass of a specimen due to water absorption when only one surface

of the specimen is exposed to water, as shown in Fig. 13. Concrete
(b) Flexural tensile test cylinders (100 mm diameter and 50 mm height) of all the mixes
were cast and moist cured for 28 days. ASTM C1585-04 [7] recom-
Fig. 10. Tensile strength at age up to 180 days.
mends conducting the investigation at the age of 28 days. How-
ever, reaction of MK with hydrated cement product continues
4.2.2. Rate of water absorption over time and changes the internal microstructure [2,25]. In order
Measurement of the rate of absorption of water was made to understand the long-term reaction of MK and changes in the
according to ASTM C1585-04 [7]. This test method determines microstructure, which influences the rate of water absorption, half
the rate of absorption of water by measuring the increase in the of the specimens were dry cured for a further 28 days. After the
46 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51

(a) The water adsorption at age 28 days

Fig. 15. Carbonation depth against the time of exposure.

4.2.3. Carbonation test

Concrete cylinders (100 mm in diameter and 200 mm in
height) were made and moist cured for 7 days. After then hav-
ing the two end surfaces coated using epoxy resin, they were
stored openly exposed to atmosphere. At the ages of 21, 28,
56, 90, 120 and 180 days, the cylinders were split in half along
the diameter to examine the depth of carbonation in the radial
direction using 1% phenolphthalein [15,36,18]. The carbonation
depths were measured at six different locations in the direction
of the height of the specimens [35]. Fig. 15 shows the average
values of carbonation depth. It can be seen that both MK and
polymer helped to decrease the carbonation rate. The optimum
mixture of 5% P/C and 15% MK/C shows the lowest carbonation
rate, at approximately half of that of the control mixture at
180 days.

(b) Water absorption comparison at ages 56 days 4.2.4. Chloride penetration

Similar cylindrical specimens as those used for the carbona-
Fig. 14. Water absorption tests.
tion test were prepared and moist cured for 28 days. Thereafter,
they were immersed in a 3% NaCl solution in order to simulate
a chloride environment. The chloride penetration depth was
curing, the specimens were treated for 3 days at a temperature of monitored up to 180 days following a similar method as that
50 degrees Celsius and relative humidity of 80%. One circular sur-
face was immersed in water to a depth of 1 to 3 mm only, such that
water ingress of unsaturated concrete was dominated by capillary
suction during initial contact with water. The rate of water absorp-
tion I in the unit of mm is defined by Eq. (1) [7]:

mt
I¼ ð1Þ
AD

where mt is the change of weight with time ðtÞ; A is the cross-

sectional area of the cylinder and D is the density of water. For
the purpose of this test, the temperature dependence of the den-
sity of water is neglected and a value of 0.001 g/mm3 is used.
Fig. 14(a) and (b) show the water absorption test results for
the concrete specimens cured for 28 days and the comparison
of control and combined MK and polymer samples between 28
and 56 days, respectively. It can be seen that the combination
of MK and polymer significantly reduces the water absorption
in along run, which can be attributed to a significant reduction
in the capillary pores because of the pozzolanic reaction of MK
with the cement hydrated products and the hydrophobic effect
of polymer. Fig. 16. Chloride penetration depth against time of exposure.
 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51 47

0.14

0.12 365 days 270 days 180 days

Avarage Weight Loss, mg

0.1

0.08

0.06

0.04

0.02

P/C=0% Mk/C= 0%

P/C=5% Mk/C= 15%

P/C=0% Mk/C= 15%

P/C=5% Mk/C= 15%

P/C=5% Mk/C= 0%
P/C=0% Mk/C= 15%
P/C=0% Mk/C= 0%

P/C=5% Mk/C= 0%

P/C=0% Mk/C= 0%

P/C=0% Mk/C= 15%

P/C=5% Mk/C= 15%

P/C=5% Mk/C= 0%
Curing 1 Curing 2 Curing 3

Mix proportions %
Fig. 17. Weight loss of the reinforcements after exposure to three conditions: Curing 1 – open atmospheric condition; Curing 2 – immersed into 20% NaCl; Curing
3 – alternated condition between the atmospheric and immersed into 20% NaCl.

used in the carbonation test. Chloride penetration depth was

identified using a solution containing 0.1% sodium fluorescein
and 0.1 N silver nitrate solution sprayed on the two surfaces
exposed by splitting through the specimens along the diameter
[5,33]. The penetration depth was measured in the same way as
that for carbonation. Fig. 16 shows a similar trend as that of
the carbonation test. Both polymer and MK decreased the chlo-
ride penetration rate considerably. The optimum mixture of 5%
P/C and 15% MK/C demonstrated the best resistance to chloride
penetration. After 90 days of immersion in 5% H2SO4 and 5% HCl 90 days

4.2.5. Corrosion weight loss

Reinforced concrete cubes (100  100  100 mm) were cast
with a 60 mm long carbon steel rebar of diameter 16 mm
positioned in each cube and parallel to a surface at a depth
of 25 mm from that surface. Before casting, the carbon steel
rebar was thoroughly cleaned and weighed to confirm its ini-
tial weight as that described by Parande et al. [37]. These cast
reinforced specimens were moist cured for 28 days [37]. To
simulate the real world situation where concrete is subjected After 180 days of immersion in 5%HCL
to various service conditions from normal atmosphere to sub-
merged under saline environments over time, the samples P/C=0% P/C=0% P/C=5% P/C=5%
were divided into three groups and each group was exposed MK/C=0% MK/C=15% MK/C=15% MK/C=0%
to different conditions. One group was exposed to an open
atmospheric environment, another group was immersed in a
20% NaCl solution, and the last group was alternately put in
these two environmental conditions for 7 days each and up
to 38 cycles in 365 days. On the time after 180, 270 and
365 days, the concrete specimens were split open using com-
pressive machine and subjected to visual observation at first.
Thereafter the steel reinforcements were pulled out from the
concrete and their surfaces were carefully cleaned thoroughly After 180 days of immersion in 5% H2SO4
using steel wire brush to get rid of all the concrete remains
and the rusts of corrosion. Later the cleaned steel rebars were Fig. 18. Concrete specimens after exposure to acidic solutions (H2SO4 and HCl).
48 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51

Fig. 19. The percentage of weight change of the concrete sample with time for various alkaline and acidic solutions.

weighed again to work out their weight loss due to corrosion recorded at 7, 14, 28, 56, 90 and 180 days. Fig. 19 show that
[19,37]. Fig. 17 shows that the weight losses of all samples the mixtures using either MK or polymer had less weight
under all three conditions are obvious, particularly, for the increase when exposed to the alkaline and salty solutions, and
samples of the control mix of 0% P/C and MK/C. However, less weight loss when exposed to acidic solutions, compared
the samples of the optimum mix of 5% P/C and 15% MK/C to the control mix with no modification using MK and polymer.
has the lowest weight loss. It can be also noticed that the It can be noticed, however, that when exposed to alkaline and
weight loss became significant after 270 days. Particularly, the salty solutions all mixtures had a steep weight increase in the
alternated exposure to atmospheric condition and immersion first 56 days, which indicates that the cured concrete underwent
in 20% NaCl demonstrates a critical influence. It can be con- further chemical reactions with infiltrated salt and alkali ions in
cluded that an alternating environmental condition accelerate an early stage. The MK modified mixtures have a significantly
the corrosion rate in concrete. reduced weight change in all the tests. It effectively confirms
the pozzolanic reactions between MK and cement hydration
4.2.6. Chemical resistance products, which result more hydration gel products with a for-
The chemical resistance was inspected by immersing cubic mation of CASH and NASH [28] to help the resistance to chem-
specimens (100  100  100 mm), after 28 days moist curing, in ical attack. The optimum mixture of 5% P/C and 15% MK/C
four different chemical solutions for 180 days. These solutions presents the least weight change in all the cases. It also can
were 20% sodium hydroxide (NaOH), 5% sodium chloride (NaCl),
5% sulphuric acid (H2SO4) and 5% hydrochloric acid (HCl). They
were selected to simulate various environmental conditions [8]. Top cap
Outlet tube
In practice, special cements (for example sulphur resistance
cement), which are very expensive, have been used for the applica-
tion in severe environments. This experimental investigation aims
to help understand how effective using ordinary MK and polymer Seal Specimen Steel casing
modified Portland cement to replace the special cements to meet
these special requirements.
Fig. 18 shows the appearance of the samples modified with
5% P/C and 15% MK/C after 90 days exposed to HCl and Inlet tube Steel bed
H2SO4, and all the samples after 180 days immersed in the
acidic solutions. The weight changes of the specimens were Fig. 20. Schematic diagram of a test specimen in PERL-200 apparatus.
 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51 49

Screw serrated bolt

Pressure gauge
Screw threaded rod

Sealing Ring

Water inlet under pressure

Fig. 22. The depth penetration of water under pressure test arrangement (BS EN
12390-8:2009 [12]).

where, Q is flow rate (cc/s), k is the permeability, L is the length of

flow (cm), l is the viscosity, A is the cross-sectional area of flow
(cm2), P1 is the upstream pressure, and P2 is the downstream pres-
(a) Concrete sure. All the concrete and mortar samples were tested using this
apparatus. Fig. 21(a) and (b) illustrate the results obtained from
concrete and mortar, respectively. Clear evidence can be seen that
the water flow rate significantly reduced for concrete modified by
MK and polymer. The mortar sample of the optimum mix of 5% P/
C and 15% MK/C shows the lowest the permeability.

4.2.8. Depth of penetration of water under pressure

The water penetration test [12], the most commonly used test
to evaluate the permeability of concrete, was conducted as well.
In this test, water was applied on one face of the 150  150 
150 mm concrete cube specimens under a pressure of 0.5 MPa (5
bars), as shown in Fig. 22. This pressure was maintained constant
for a period of 72 h. After the completion of the test, the specimens
were taken out and split open into two halves. The water penetra-
tion front profile in concrete was then marked (Fig. 23(a)) and the
maximum depth of water penetration front in specimens was
recorded and considered as an indicator of the water penetration
[20]. This test was conducted at 28, 56 and 90 days after moist cur-
ing for 28 days. As can be seen in Fig. 23(b), the depth of penetra-
(b) Mortar
tion of water significantly reduced in the specimen of the optimum
Fig. 21. Relationships between permeability and time of curing (PERL-200 device). mix of 4% P/C and 15% MK/C compared with the rest of the sam-
ples. Furthermore, there is a development in water permeability
of the modified concrete with increased age of the specimens com-
pared with the control sample. This can be attributed to the contin-
be seen that all these curves present a flatting trend after 90
uous reaction of MK with the hydrated cement product, producing
days, indicating a long-term durability, in which the optimum
a less permeable matrix with time.
mixture shows the best.

4.2.7. Water flow rate of concrete 4.2.9. Gas penetration of concrete under pressure
In this study, both mortar and concrete specimens, of the In this test, the similar experimental setup and samples were
dimension of 2.54 cm in diameter and 2.54 cm in length, were used as in the test for depth of penetration of water under pres-
tested. The specimens were moist cured for 28 days and tested sure. However, the CO2 was applied from the top instead from
at ages 28, 56 and 90 days. The PERL-200 permeameter, provided the bottom with a pressure of 0.4 MPa (4 bars) for 8 h. These
by Core Lab Instruments, was used for the test. Similar to the arrangements were implemented to maintain the constant pres-
test apparatus used by Kameche et al. [27] and Li et al. [32], sure in order to simulate the concrete under sewerage conditions.
incorporating a digital pressure transducer and a calibrated After the test, the specimens were taken out and split open into
visual flow (measurement) cell, it uses the valves to control a two halves. The penetration of carbonation was determined by
flow system to enable the determination of flow rate of water treating a freshly broken surface with 1% phenolphthalein. The
through a one-inch diameter core sample plug. The water flow region of Ca(OH)2 is coloured pink while the carbonated areas is
rate through the sample is determined by measuring the time uncoloured. The gas penetration profile on the concrete surface
required for the water (10 cm3) to pass between the calibrations was then marked and the maximum depth of gas penetration in
marks. Fig. 20 shows the schematic diagram of the experimental specimens was recorded and considered as an indicator of the
setup. The permeability of the sample can be determined in gas penetration. Fig. 24 compares the gas penetration depths of
terms of the Darcy’s Law: the four mixtures. It can be seen that using MK and polymer
respectively have exhibited reductions in the gas penetration.
k  AðP 1  P2 Þ The optimum mix of 5% P/C and 15% MK/C shows the best result
Q¼ ð2Þ
lL for the gas penetration resistance.
50 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51

P/C= 5% P/C= 5% P/C = 0% P/C = 0%

MK/C = 0% MK/C = 15% MK/C = 15% MK/C = 0%

(a) Water front profiles

(b) Depth of water penetration vs curing time

Fig. 23. Water penetration under 0.5 MPa (5 bars) pressure.

5. Conclusions

This paper has reported an experimental study on long-term

durability properties of concrete modified with MK and polymer.
Various long-term durability tests were carried out to understand
the behaviour of modified concrete subjected to an extremely
harsh environment. The following conclusions can be drawn based
on the experimental results:

Metakaolin will accelerate the setting time of cement pastes but

reduce the workability of concrete. However, polymer has an
inverse influence on the two properties.

The bi-polymer of composition 80% SBR and 20% PVA shows an
optimised result when it works together with MK.

The addition of 5% optimised bi-polymer and 15% cement
replacement by metakaolin generates an optimised concrete
mixture for long-term mechanical properties.

All the long-term durability experimental investigations
suggest that both MK and polymer improve the properties.
Fig. 24. Gas penetration depth vs time for conventional and modified concrete. Their combination presents a complement to each other.
 A. Al Menhosh et al. / Construction and Building Materials 172 (2018) 41–51 51

The optimum mix base on mechanical properties also demon- [22] E. Guneyisi, M. Gesoglu, S. Karaoglu, K. Mermerdas, Strength, permeability and
shrinkage cracking of silica fume and metakaolin concretes, Constr. Build.
strates a great enhance on durability properties compared to
Mater. 34 (2012) 120–130.
using MK or polymer only. [23] E. Guneyisi, M. Gesoğlu, K. Mermerdasß, Improving strength, drying shrinkage,
and pore structure of concrete using metakaolin, Mater. Struct. 41 (5) (2008)
Conflict of interest 937–949.
[24] M.M. Hossain, M.R. Karim, M. Hasan, M.K. Hossain, M.F.M. Zain, Durability of
mortar and concrete made up of pozzolans as a partial replacement of cement:
The authors declare that they have no conflict of interest. a review, Constr. Build. Mater. 116 (2016) 128–140, https://doi.org/10.1016/
j.conbuildmat.2016.04.147.
[25] J.M. Justice, L.H. Kennison, B.J. Mohr, S.L. Beckwith, L.E. McCormick, B. Wiggins,
Acknowledgment Z.Z. Zhang, K.E. Kurtis, Comparison of two metakaolins and a silica fume used
as supplementary cementitious materials. Paper Presented at the Proc.
This work was funded by the Iraqi Ministry of Higher Education Seventh International Symposium on Utilization of high-strength/High
Performance Concrete, Washington D.C, 2005.
and Scientific Research Scholarship Program. [26] E. Kamseu, M. Cannio, E.A. Obonyo, F. Tobias, M.C. Bignozzi, V.M. Sglavo, C.
Leonelli, Metakaolin-based inorganic polymer composite: effects of fine
References aggregate composition and structure on porosity evolution, microstructure
and mechanical properties, Cem. Concr. Compos. 53 (2014) 258–269.
[27] Z.A. Kameche, F. Ghomari, M. Choinska, A. Khelidj, Assessment of liquid
[1] S.F.U. Ahmed, Mechanical and durability properties of mortars modified with
water and gas permeabilities of partially saturated ordinary concrete, Constr.
combined polymer and supplementary cementitious materials, J. Mater. Civil
Build. Mater. 65 (2014) 551–565, https://doi.org/10.1016/j.conbuildmat.
Eng. 23 (9) (2011) 1311–1319.
2014.04.137.
[2] S. Aiswarya, G. Prince Arulraj, C. Dilip, A review on use of metakaolin in
[28] V. Kannan, K. Ganesan, Chloride and chemical resistance of self compacting
concrete, IRACST – Eng. Sci. Technol. (2013).
concrete containing rice husk ash and metakaolin, Constr. Build. Mater. 51
[3] A. Al Menhosh, Y. Wang, Y. Wang, The mechanical properties of the concrete
(2014) 225–234, https://doi.org/10.1016/j.conbuildmat.2013.10.050.
using metakaolin additive and polymer admixture, J. Eng. 2016 (2016) 1–6.
[29] J.M. Khatib, E.M. Negim, E. Gjonbalaj, High volume metakaolin as cement
[4] J. Ambroise, S. Maximilien, J. Pera, Properties of metakaolin blended cements,
replacement in mortar, World J. Chem. 7 (1) (2012) 7–10, https://doi.org/
Adv. Cem. Based Mater. 1 (4) (1994) 161–168.
10.5829/idosi.wjc.2012.7.1.251.
[5] C. Andrade, M. Castellote, C. Alonso, C. González, Relation between
[30] B.B. Konar, A. Das, P.K. Gupta, M. Saha, Physicochemical characteristics of
colourimetric chloride penetration depth and charge passed in migration
styrene-butadiene latex- modified mortar composite vis-à-vis preferential
tests of the type of standard ASTM C1202–91, Cem. Concr. Res. 29 (3) (1999)
interactions, J. Macromol. Sci., A 48 (9) (2011) 757–765, https://doi.org/
417–421.
10.1080/10601325.2011.596072.
[6] K.M. Atkins, R.N. Edmonds, A.J. Majumdar, The hydration of portland and
[31] S.C. Kou, C.S. Poon, A novel polymer concrete made with recycled glass
aluminous cements with added polymer dispersions, J. Mater. Sci. 26 (1991)
aggregates, fly ash and metakaolin, Constr. Build. Mater. 41 (2013) 146–151.
2372–2378.
[32] X. Li, Q. Xu, S. Chen, An experimental and numerical study on water
[7] ASTM Standards C1585 – 04, Measurement of Rate of Absorption of Water by
permeability of concrete, Constr. Build. Mater. 105 (2016) 503–510, https://
Hydraulic Cement Concrete, U. S, 2004.
doi.org/10.1016/j.conbuildmat.2015.12.184.
[8] M.A. Beulah, M.C. Prahallada, Effect of replacement of cement by metakalion
[33] E. Meck, V. Sirivivatnanon, Field indicator of chloride penetration depth, Cem.
on the properties of high performance concrete subjected to hydrochloric acid
Concr. Res. 33 (8) (2003) 1113–1117.
attack, Int. J. Eng. Res. Appl. 2 (2012).
[35] M. Otieno, H. Beushausen, M. Alexander, Effect of chemical composition of slag
[9] BS EN 12390-3, Testing Hardened Concrete: Compressive Strength of Test
on chloride penetration resistance of concrete, Cem. Concr. Compos. 46 (2014)
Specimens, BSI, London, 2009.
56–64, https://doi.org/10.1016/j.cemconcomp.2013.11.003.
[10] BS EN 12390-5, Testing Hardened Concrete: Flexural Strength of Test
[36] V.G. Papadakis, Effect of supplementary cementing materials on concrete
Specimens, BSI, London, 2009.
resistance against carbonation and chloride ingress, Cem. Concr. Res. 30 (2000)
[11] BS EN 12390-6, Testing Hardened Concrete: Tensile Splitting Strength of Test
291–299.
Specimens, BSI, London, 2009.
[37] A. Parande, B. Babu, M. Karthik, K. Kumaar, N. Palaniswamy, Study on strength
[12] BS EN 12390 – 8, Depth of Penetration of Water Under Pressure, BSI, London,
and corrosion performance for steel embedded in metakaolin blended
2009.
concrete/mortar, Constr. Build. Mater. 22 (3) (2008) 127–134, https://doi.
[13] BS EN 197-1, Cement: Composition, Specifications and Conformity Criteria for
org/10.1016/j.conbuildmat.2006.10.003.
Common Cements, British Standards Institution, London, 2011.
[38] R. Pouhet, M. Cyr, Carbonation in the pore solution of metakaolin-based
[14] BS ISO 1920-8, Testing of Concrete. Determination of the Drying Shrinkage of
geopolymer, Cem. Concr. Res. 88 (2016) 227–235.
Concrete for Samples Prepared in the Field or in the Laboratory, BSI, London,
[39] A.M. Rashad, Metakaolin as cementitious material: history, scours, production
2009.
and composition – a comprehensive overview, Constr. Build. Mater. 41 (2013)
[15] BS 1881 Part-121, Method for Determination of Static Modulus of Elasticity in
303–318.
Compression, BSI, London, 1983.
[40] A. Souri, H. Kazemi-Kamyab, R. Snellings, R. Naghizadeh, F. Golestani-Fard, K.
[17] BS 882, Specification for Aggregates from Natural Sources for Concrete, BSI,
Scrivener, Pozzolanic activity of mechanochemically and thermally activated
London, 1992.
kaolins in cement, Cem. Concr. Res. 77 (2015) 47–59.
[18] C.F. Chang, J.W. Chen, The experimental investigation of concrete carbonation
[41] K. Srinivasu, M.L.N. Krishna Sai, N. Venkata Sairam Kumar, A review on use of
depth, Cem. Concr. Res. 36 (9) (2006) 1760–1767, https://doi.org/10.1016/j.
metakaolin in cement mortar and concrete. Paper Presented at the
cemconres.2004.07.025.
International journal of Innovative Research in Science, Engineering and
[19] D. Chung, Corrosion control of steel-reinforced concrete, J. Mater. Eng.
Technology, 2014.
Perform. 9 (5) (2000) 585–588.
[43] Z. Yang, X. Shi, A.T. Creighton, M.M. Peterson, Effect of styrene–butadiene
[20] P. Dinakar, P.K. Sahoo, G. Sriram, Effect of metakaolin content on the properties
rubber latex on the chloride permeability and microstructure of Portland
of high strength concrete, Int. J. Concr. Struct. Mater. 7 (3) (2013) 215–223,
cement mortar, Constr. Build. Mater. 23 (6) (2009) 2283–2290, https://doi.org/
https://doi.org/10.1007/s40069-013-0045-0.
10.1016/j.conbuildmat.2008.11.011.
[21] F. Giustozzi, Polymer-modified pervious concrete for durable and sustainable
transportation infrastructures, Constr. Build. Mater. 111 (2016) 502–512,
https://doi.org/10.1016/j.conbuildmat.2016.02.136.