CONCRETE CARBONATION IN ARID CLIMATE

H. AI-Khaiat*, Kuwait University, Kuwait

M. N. Haque, Kuwait University, Kuwait
N. Fattuhi, England

29th Conference on OUR WORLD IN CONCRETE & STRUCTURES: 25 - 26 August 2004,

Singapore
 
Article Online Id: 100029019 

The online version of this article can be found at:

http://cipremier.com/100029019

This article is brought to you with the support of 

Singapore Concrete Institute 

www.scinst.org.sg 

All Rights reserved for CI‐Premier PTE LTD 

You are not Allowed to re‐distribute or re‐sale the article in any format without written approval of 
CI‐Premier PTE LTD 

Visit Our Website for more information 

www.cipremier.com  
 29 th Conference on OUR WORLD IN CONCRETE & STRUCTURES: 25 - 26 August 2004 , Singapore

CONCRETE CARBONATION IN ARID CLIMATE

H. AI-Khaiat*, Kuwait University, Kuwait

M. N. Haque, Kuwait University, Kuwait
N. Fattuhi, England

Abstract

Fifty-four different concrete mixes were exposed to hot arid weather in Kuwait to
monitor their extent of carbonation. Various parameters affecting carbonation such
as water/cement ratio, cement content, type of cement, type and dosage of
admixture, water-curing period and type of coating were investigated. For many of
these variables, the carbonation depth and rate constant (K) are reported for an
exposure period of 8.6 years.
The results show that the water/cement ratio is one of the main factors affecting
carbonation. The values of 'K' ranged from 2.1 to 7.8 as the water/cement ratio
increased from 0.45 to 0.8. Concrete made with white Portland cement carbonated
less than ordinary or sulphate resisting cements. Two curing compounds were
effective in reducing carbonation while two coating materials prevented carbonation
after eight years of exposure.

Keywords : admixtures, cement, carbonation, coatings, curing compounds, field,

water-curing, concrete

1. Introduction
Reinforcement corrosion is recognized as the most important durability problem of reinforced concrete
structures . The corrosion of reinforcing steel is accelerated when the concrete carbonates . This is
due to a reduction in alkalinity of concrete cover to reinforcement, which provides the protective
environment.
Results of a long-term investigation of concrete exposed to natural weather conditions in Kuwait
are presented in this paper. Concrete prisms were tested for up to 8.6 years of exposure. Various
parameters affecting carbonation are considered in the study such as water/cement ratio , cement
content, type of cement, type and dosage of admixture, water-curing period and type of coating .
It has been generally accepted that the carbonation depth , d, is related to exposure time, t, by the
equation

(1 )
Fookes [1J, however, reports that in hot dry environments carbonation penetrates at about 1 mm
per year depending on the concrete: it may be a little more in dry situations and significantly less in
wetter situations . Also, Roberts [2] has reported an average Portland cement concrete may show a
depth of carbonation of 5-8 mm after 10 years and 10-15 mm after 50 years . Whilst Sims [3] and
Roberts [2] have reported the K values which may be of more relevance in temperate environment,
Haque and AI-Khaiat [4J presented K values which are more realistic for the hot dry coastal exposure
conditions of Kuwait (or the Gulf region). Of course, the carbonation is a diffusion phenomenon and
the rate of penetration of C02 depends main ly on the concrete quality and the exposure conditions.

161
 In this study 'd' is presented in mm, 't' in years and thus 'K' in unit of mm.yea(O.5. The
performance of different mixes and curing methods etc. were assessed by comparing the 'K' values
calculated from the best fit result for the d-t relationship.

2. Materials and testing

2.1 Materials and Preparation of Specimens
Ordinary, white and sulphate-resisting Portland cements were used . However the bulk of the
specimens were made with 0 rdinary Portland cement delivered in 0 ne batch. The fine and coarse
aggregates consisted of natural desert sand and crushed desert gravel. Drinking tap water was used
for mixing and curing the concrete. Some properties of the various materials used have been
described elsewhere (5] .
Ten proprietary admixtures, four curing compounds / aids and two coatings specifically designed
to reduce carbon dioxide diffusion into concrete were used. Details of the properties of the
constituents of the various concrete mixes are shown in Table 1.
The various mixes were prepared in the laboratory and 100 x 100 x 500 mm prisms were cast.
The prisms were demolded after 24 hours. Most prisms were cured in air and / or water for various
periods. However, some prisms were brushed with curing compounds . Also, some prisms were
brushed with anti-carbonation coatings. Details of the application of the various curing compounds /
aids and coatings were described by Fattuhi and AI Khaiat [5].

2.2 Testing
To measure the depth of carbonation, a slice with a minimum thickness of 20 mm was broken off each
100 x 100 x 500 mm prism. Then a phenolphthalein indicator solution was sprayed uniformly on the
freshly broken surface. The depth of the carbonated surface layers of concrete was measured with a
steel ruler of 0.5 mm accuracy. The average of the four measurements, from each edge, was taken
to be the depth of carbonation for each prism . Each carbonation depth presented is t he average
obtained from the two slices. Carbonation testing was performed periodically for the concrete.

3. Weather conditions
Kuwait weather is characterized as hot and dry environment. Average annual mean temperature is

25 .5°C . For months of May to September, the average mean temperature exceeds 30°C . The

maximum mean temperature from June to August exceeds 40°C. The average rainfall totals 85 mm

. per year. The mean relative humidity ranges from 20 to 56% . For seven months a year, the mean

maximum relative humidity is less than 50% [6]. The highest mean temperature recorded was 48°C

while the lowest mean was 3°C during this period of 8.6 years . For 60% of the period the maximum

mean temperature was higher than 30°C. The relative humidity reached highest values (80 - 90%)

during winter months, and its lowest values (5 - 40%) during the summer months .

4. Results and discussion

4.1 Cement and aggregate content
Two sets of prisms with a water/cement ratio of 0.5 and water cured for six days were considered . No
admixture was used during the preparation of the prisms, but the cement content of each set of
prisms was either 435 or 460 kg/m 3. The results indicate that the carbonation was marginally lower
for prisms with a cement content of 460 kg/m3 and a fine to total aggregate ratio of 0.32. Therefore,
by increasing the cement content by 5.7% (i.e. from 435 up to 460 kg/m 3), the carbonation decreased
by 5.4% (i.e. 'K' from 3.11 down to 2.95). However, when the fine to total aggregate ratio in the
prisms was increased from 0.32 to 0.39 (i.e. over sanded prisms), the carbonation increased by 4.1 %
(i.e. 'K' from 2.95 to 3.07). Of course, no firm conclusion can be drawn from these results and should
be considered indicative only. There is 0 nly marginal difference in the cement content 0 f the two
mixes tested .

4.2 Cement type

There was virtually no difference between the carbonation depths measured for concrete mixes made
with ordinary (K =3.23) and sulphate resisting (K =3.11) Portland cements. However, concrete made
with white Portland cement (K = 2.72) exhibited the least carbonation (i.e. about 19% lower than
concrete made with ordinary Portland cement). The maximum carbonation depths recorded were
15.5, 15.0 and 13.5 mm for sulphate resisting, ordinary and white Portland cement concrete prisms,
respectively.

]62

 4.3 Water/cement ratio

Concrete prisms were made with water/cement ratios ranging from 0.45 to 0.80, and the cement
content varied between 456 and 256 kg/m 3 . Fig . 1 shows that as the water/cement ratio increased ,
there was a significant increase in the depth of carbonation . The increase in carbonation depth
reached more than three folds when the water/cement ratio was increased from 0.45 to 0 .80 (K from
2.05 to 7.79). This shows the importance of selecting and maintaining a low water/cement ratio in
concrete mixes in order to achieve an adequate durability. The maximum carbonation depth recorded
were 13.0 and 24.0 mm for prisms with water/cement ratios of 0.45 and 0.80, respectively. Again, it
has to be noted that the difference in the carbonation depth is not due to the difference in
water/cement ratio per se. It has been reported, that it is, in reality, the lowering of the compressive
strength and the open pore structure of the concrete which enhances the depth of carbonation with an
increase in the water cement ratio [7] . Carbonation depth has been reported to be nearly a linear
function of the concrete strength [4] . Accordingly, the average carbonation depth of the concrete
mixes with water/cement ratio of 0.45, 0.5, 0.6, 0.7 and 0.8 has been plotted against the average
compressive strength of these mixes in Fig . 2. There is a near linear relationship between the two
which agrees well with the earlier findings [4]. Other factors remaining unchanged strength is a
function of water/cement ratio .

25~----------------------------------------------------------~

Water/Cement Ratio (Summer)

-e- 0.45 (K = 2 .05)

o
20 o
.. 0 0.5 (K = 3.23)
- -6 - 0.6 (K = 4 .23)

- -A- - 0.7 (K = 5 .58)

E 15 - 0-0.8 (K=7.79)

..

§.
.r:

Q.
CI)
~ --­
IC

.6
"" - ­

_ _ -4- ­
C 10 /I - - - ­
o
o

O~------.------.r-----~------~-------r-------r------'-----~
o 400 800 1200 1600 2000 2400 2800 3200
Time (Days)

Figure 1. Carbonation of Concrete with various Water/Cement Ratios

4.4 Type and dosage of admixture

Four main types of admixtures were used in the mixes. These included high-range water-reducing,
high-range water-reducing and retarding , water-reducing and retarding , and plasticizing admixtures.
Fig. 3 shows the variation of the carbonation depth of the various prisms containing admixtures with
time. K values were lowest for prisms containing A 7 admixture, a plasticiser based on calcium
lignosulfonate. However, K values were highest for prisms containing admixtures AIO and A9.
Admixture A9 is a superplasticizer based on condensate of sodium naphthalene sulfonate with
formaldehyde . Admixture AlO is a superplasticizer and retarder based on condensate of sodium
naphthalene sulfonate with formaldehyde and these prisms had the highest average carbonation
depth of 9.2 mm at about 3100 days (K = 3.19). The difference in the carbonation depth for prisms
containing the various admixtures was 13%. No definitive conclusions can be made on the effect
differing admixtures used on the depth of carbonation. In practice, when superplasticisers are
introduced other concomitant changes are made in water/cement ratio etc. In retrospect it was
advisable to keep either the workability or the strength of these concretes comparable to compare the
effect on carbonation .

163
 20
•
18

16

.s 14
'tI
,
~ 12
C-
al
'tI 10 •
I:
0

;: 8

I'll y = -O.3936x + 28 .387

I:
0
.c
... 6
I'll
U
4

20 25 30 35
40 45 50 55 60

compressive strength (MPa)

Figure 2. Relationship between Strength and Depth of Carbonation

4.5 Water-curing period

A further ten sets of prisms with a water/cement ratio of 0.5, were water cured for 0, 2, 6, 13 or 27
days and only half of them contained admixture A2. Fig. 4 shows the results of the five sets that
contained admixture A2, while Fig . 6 shows the results of the other five sets of prisms without the
admixture A2. Examination of Figs. 4 and 5 indicate that prisms with a cement content of 460 kg/m3
. and containing no admixture exhibited slightly lower carbonation than those with a cement content of
410 kg/m 3 with an admixture. Also, the results show considerable reduction in the carbonation when
the water curing period was increased from 0 to 6 days. Haque [8, 9]. has reported that in hot
exposure conditions 6 days of water curing is almost mandatory for the proper strength development
and this study further establishes that at least 6 days of curing is highly desirable to minimize the
carbonation depth and hence to improve the overall performance and durability of concrete.

4.6 Curing aids

Four curing compounds / aids Cl, C2, C3 and C4 were used to brush prisms. Each set of prisms
was demolded, then immersed in water for a period of 3 - 4 hours. After surface drying they were
brushed several times with a curing compound according to the specifications of manufacturers. All
specimens were then air cured in the laboratory for 27 days. The prisms had a water/cement ratio of
0.5 and contained 1% dosage of admixture A2. Curing compound Cl was based on sodium silicate,
C2 on resin, C3 on wax emulsion, while C4 was based on white resin.
The carbonation results for the prisms are shown in Fig . 6. It should be noted that prisms
brushed with curing compounds / aids Cl - C3 were prepared during the summer, however prisms
brushed with curing compound C4 were prepared during the winter. Additional prisms brushed with
curing compound C2 and prepared during the winter are denoted by C2W .
The carbonation depth measurement results for the prisms prepared during the summer showed
- -
that prisms brushed with curing compounds C2 and C3 performed better (K values of 3.6 and 3.7,
respectively) than prisms brushed with curing compound C1 (K = 4.21). It should be noted that similar
prisms containing admixture A2 and air cured exhibited an average carbonation depth of 11.6 mm
= - -
after 3100 days of exposure (K 4.1). These results indicate that the use of compounds C2 and C3
have reduced the carbonation depth rate by about 13% in 8-year exposure period .

164
 14

12 Admixture Type
--&- A6 (K = 3.08)

10
-.- A7 (K 2.68) =
··~ · A8 (K = 3.16)

-E - .If- . A9 (K =
3.19)

-E
:Q.
5
8 - ­ .. ­ A10 (K = 3.19)

CI) 6
C

4 t

0
0 400 800 1200 1600 2000 2400 2800 3200
Time (Days)
Figure 3. Carbonation of Concrete Containing Different Admixtures

14
Water Curing Period (days)
(without Admixture)
12
--&-27 (K = 2.78)
-.- 13 (K = 3.04)
10
= 0

E
··~· · 6 (K 3.07)
-·1f-·2 (K 3.03) =
Go<

0
E 8 -0- 0 (K = 4.05)

J:
c.
4)
C
6 0 0

0
,
0 400 800 1200 1600 2000 2400 2800 3200
Time (Days)
Figure 4. Carbonation of Concrete Cured for Various Periods: Without Admixtures

The results also showed that prisms C2W which were prepared during the winter and brushed
with curing compound C2, exhibited a higher carbonation depth rate (K = 4.3). On the other hand,
prisms brushed with curing compound C4 exhibited a lower carbonation depth rate (K = 3.34).
In summary, the curing compounds used have been found to effective in mitigating the
carbonation depth of concrete exposed to hot ambient conditions.

165
Conclusions
1. The water/cement ratio was the main parameter affecting the rate of concrete carbonation. The K
=
value varied from about 2.1 to 4.2 for w/c 0.45 to 0.6.
2. Six days of water curing was found necessary and adequate to lower the carbonation depth to a
level comparable to 13 or even 27 days of water curing.
3. All the four curing compounds brushed onto concrete prisms were found effective in reducing the
carbonation of concrete.
4. White Portland cement concrete appeared to exhibit the lowest rate of carbonation. However,
ordinary and sulphate resisting Portland cement concretes appeared to have had a similar rate of
carbonation.

References:
[1] Fookes, P.G., "Concrete in Hot Dry Salty Environments", Concrete, 29 (1995) 34-39.
[2] Roberts, M.H., "Carbonation of Concrete Made with Dense Natural Aggregates", BRE,
Information Sheet, (1981 ).
[3] Sims, I., "The Assessment of Concrete for Carbonation", Concrete, 28 (1994) 33-38.
[4] Haque, M.N. and AI-Khaiat, H.. "Carbonation of Concrete Structures in Hot Dry Coastal
Regimes". Cement and Concrete Composites, 19 (1997) 123-129.
[5] Fattuhi, N.I. and AI-Khaiat, H., "Shrinkage of Concrete Exposed to Hot & Arid Climate", Journal
of Materials in Civil Engineering (ASCE) 11 (1) (1999) 66-75.
[6] Meteorological Department, "Climatological Annual Summary", Climatological Division Report,
Kuwait International Airport (1996).
[7] AI-Khaiat, H. and Haque, M.N., "Carbonation of Some Coastal Concrete Structures in Kuwaif'.
ACI Material Journal, 94 (1997) 602-607 .
[8] Haque, M.N., "Some concretes need seven days initial curing" . American Concrete Institute,
Concrete International : Design and Construction, 12(2) (1990) 42-46 .
[9] Haque, M.N., "Give it a Week: Seven days initial curing". ACI, Concrete International, 20(9)
(1998) 45-48 .
[10] Davies, "Protection of concrete in the Arabian Gulf', Paint and Ink Int. 8(3) (1995) 489-501.

167