Content uploaded by Árpád Rózsás
Author content
All content in this area was uploaded by Árpád Rózsás on Feb 28, 2016
Content may be subject to copyright.
Second Conference of Junior Researchers in Civil Engineering
Rózsás Á., Kovács N.: Carbonation of Concrete Infrastructure in Hungary in the Light of Climate Change
Carbonation of Concrete Infrastructure in Hungary
in the Light of Climate Change
Árpád Rózsás
BME Department of Structural Engineering, e-mail: rozsas.bme@gmail.com
Nauzika Kovács
BME Department of Structural Engineering, e-mail: nauzika@vbt.bme.com
Abstract
The costs related to corrosion are immense and they are expected to rise in the following decades due to human induced climate
change. This paper examines a slice of this problem: the carbonation process - which is the most common corrosion type of
concrete structures - in the light of climate change. Since current durability specifications are prevalently based on historical data,
experience and do not consider the effect of climate change, there is a pressing need to assess the reliability of them. For this
purpose full-probabilistic, time-dependent reliability analyses are performed utilizing Monte Carlo simulation technique.
According to the recommendations of the Intergovernmental Panel on Climate Change multiple future scenarios are used to assess
the effects of altering climate. The calculations indicate that the carbonation depth could increase by 21% compared to constant
CO2 level of year 2000 as a reference. The risk of depassivation for structures built per Eurocode and the superseded Hungarian
bridge standard (ÚT) may increase by 115% and 55% respectively up to the end of the century, due to rising CO2 level. The
findings reinforce that the effects of climate change should be reflected in the standards and the revision of current durability
specifications is required.
Introduction
The effects and costs of corrosion worldwide are enormous but they are typically greatly underestimated
since not accompanied by singular catastrophic events. The sum of direct and indirect costs of corrosion is
about 6% of world’s GDP [1]. The majority of our structures is made of reinforced concrete which is also
prone to corrosion and its repair and maintenance consumes great amount of money every year. Surprisingly,
contrary to its significance the current concrete durability specifications are prevalently based on historical
data and experience [2]. An additional effect which calls for the revision of present durability specifications
is climate change. It is predicted that in the following decades environmental conditions will considerably
alter which may amplify the corrosion processes. As a response to these concerns the primary aim of this
paper is to examine the effect of climate change on the durability of concrete structures and to assess the
sufficiency of the present and superseded durability provisions in Hungary. We restrict our attention to the
examination of the carbonation process. Although we employ the environmental conditions and predictions
valid for Hungary, by changing the appropriate parameters, the applied procedure can be used irrespectively
of the geographical location. Due to the significant uncertainty of variables full-probabilistic, time-dependent
reliability analyses are conducted. The calculations are completed for a representative structure with design
working life of 100 years considering multiple climate change scenarios, exposure classes and cement types.
Climate in the Future
Corrosion processes are governed by environmental factors therefore their realistic prediction is essential to
obtain reliable results. According to the recommendations of Intergovernmental Panel on Climate Change
(IPCC) multiple climate change scenarios are analyzed which are chosen to cover a wide range of possible
future progresses [3]. No distinction is made between these scenarios, each of them is treated with equal
probability of occurrence. The time course of atmospheric CO2 concentrations - which has a major influence
on the corrosion - per scenarios are shown in Fig. 1. The concentration is given in volumetric parts per
million [ppmv], the shaded areas show the ±1 standard deviation () range of the predictions.
Second Conference of Junior Researchers in Civil Engineering
Rózsás Á., Kovács N.: Carbonation of Concrete Infrastructure in Hungary in the Light of Climate Change
CO
2
concentration [ppmv]
CO
2
concentration [ppmv]
CO
2
concentration [ppmv]
CO
2
concentration [ppmv]
CO
2
concentration [ppmv]
CO
2
concentration [ppmv]
Time [year] Time [year] Time [year]
Time [year] Time [year] Time [year]
2000 2020 2040 2060 2080 2100
400
600
800
1000
A1B +σ
A1B µ
A1B -σ
2000 2020 2040 2060 2080 2100
400
600
800
1000
A1FI +σ
A1FI µ
A1FI -σ
2000 2020 2040 2060 2080 2100
400
600
800
1000
A1T +σ
A1T µ
A1T -σ
2000 2020 2040 2060 2080 2100
400
600
800
1000
A2 +σ
A2 µ
A2 -σ
2000 2020 2040 2060 2080 2100
400
600
800
1000
B1 +σ
B1 µ
B1 -σ
2000 2020 2040 2060 2080 2100
400
600
800
1000
B2 +σ
B2 µ
B2 -σ
Fig. 1. CO2 concentration in time per various climate change scenarios [3].
These describe economic (A1, A2) and environmental (B1, B2) focused future developments with local (A2,
B2) or global (A1, B1) orientations. The three groups within the A1 family are characterizing alternative
developments of energy technologies: A1FI (fossil fuel intensive), A1B (balanced), and A1T (predominantly
non-fossil fuel) [3]. Additionally, a reference concentration level (reference 2000) is chosen which represents
the CO2 level in year 2000 on which the current experiences and standards are roughly based. Other
environmental parameters, i.e. relative humidity, number of rainy days, are approximated by their measured
values in the previous years by the Hungarian Meteorological Service since no significant change is
expected in the future in their values and regional distributions.
Carbonation Model
The deterioration of reinforced concrete structures is initiated by the diffusion of CO2 into the concrete,
where it could reduce the pH of the surrounding of the rebar in time and create suitable condition for steel
corrosion. In this paper solely the initiation period characterized by carbonation is considered. The applied
carbonation model is based on the recommendation of CEB-fib Model Code (MC) [2]; this model has
relatively wide internationally acceptance. The carbonation in time is expressed as follows:
1
() 2 ()
cecNACS
x
tkkRCtWt
(1)
Where t stands for time, in the cases discussed in present paper it goes from 2000 to 2100. The unitless ke is
the environmental function which takes into account the effect of ambient relative humidity (RHreal), (Eq. 2).
1
1
e
e
e
g
f
real
ef
ref
RH
kRH
(2)
RHref is the value of ambient relative humidity maintained during the tests used to determine the carbonation
of concrete specimens; ge and fe are exponents calibrated to measured data. Along with the further
parameters these are treated as stochastic variables, their probability characteristics are summarized in
Table 1. Execution parameter kc expresses the effect of the length of curing time (tc), (Eq. 3).
7
c
b
c
c
t
k
(3)
In the above equation bc is a regression parameter. The diffusivity of concrete is characterized by the inverse
Second Conference of Junior Researchers in Civil Engineering
Rózsás Á., Kovács N.: Carbonation of Concrete Infrastructure in Hungary in the Light of Climate Change
carbonation resistance ( 1
NAC
R
) and can be calculated by using Eq. 4 from the inverse resistance ( 1
ACC
R
)
determined in the accelerated carbonation tests.
11
ttNAC ACC
kRR
(4)
The accelerated resistance is converted to normal conditions by means of a regression parameter (kt) and an
error term (
t). The total atmospheric CO2 concentration (CS) is expressed as the sum of predicted
atmospheric concentration (CS,atm) and the effect of local emission sources (CS,emi), such as the traffic in
cities, (Eq. 5). In urban regions the latter is taken as 15% of the atmospheric CO2 level of rural areas with
0.15 coefficients of variation (COV) following the measurements of Stewart et al. [4].
,,SSatmSemi
CC C (5)
Weather function W(t) is another unitless parameter which varies in time and shows the influence of the
wetness of concrete surface, (Eq. 6).
2
0
()
bw
SR
pToW
t
Wt t
(6)
In the above expression pSR means the probability that the examined surface is exposed to driving rain, ToW
(time of wetness) is the number of days in year with precipitation greater than 2.5mm, bw is a regression
exponent and finally t0 is the so called time of reference. The above model can only approximately consider
the time-varying CO2 concentration since it provides a point-in-time estimation of carbonation. To overcome
this issue the carbonation depth is determined by numerically solving the governing differential equation of
diffusion while using the described CEB-fib model [5].
Reliability Analysis
The above presented corrosion model and climate predictions are used in the time-variant reliability analysis.
Full-probabilistic (Level III) approach is adopted where the probability of failure is determined by crude
Monte Carlo simulation. The stochastic properties of the relevant variables are summarized in Table 1.
Table 1. Stochastic properties of model parameters.
Parameters Distribution Mean COV Ref.
Relative humidity of carbonated layer (RHreal), [-] Beta 0.70 0.125 [2]
Reference relative humidity (RHref), [-] Constant 0.65 0 [2]
Exponent (ge), [-] Constant 2.50 0 [2]
Exponent (fe), [-] Constant 5.00 0 [2]
Regression exponent (bc), [-] Normal -0.567 0.042 [2]
Period of curing (tc), [day] Constant 2.00 0
Inverse carbonation resistance (RNAC-1), [mm2/yr/(kg/m3)] Normal varying1 varying1 [2]
Regression parameter (kt), [-] Normal 1.25 0.28 [2]
Error term (
t), [mm2/yr/(kg/m3)] Normal 315.50 0.152 [2]
Equivalent water cement ratio, [-] Constant varying2 0 [6]
CO2 concentration of atmosphere (CS,atm), [ppmv] Normal varying3 varying3 [3]
CO2 concentration due to emission (CS,emi), [ppmv] Normal 0.15·CS,atm 0.15 [7]
Probability of driving rain (pSR), [-] Constant 0.10 0
Exponent of regression (bw), [-] Normal 0.446 0.365 [2]
Time of reference (t0), [yr] Constant 0.0767 0 [2]
Time of wetness (ToW), [-] Constant 0.20 0
Nominal concrete cover (cmin,dur), [mm] Weibull varying 8/a [2, 8]
1depends on the cement type and w/c. 2depends on the exposure class.3see Fig. 1.
Second Conference of Junior Researchers in Civil Engineering
Rózsás Á., Kovács N.: Carbonation of Concrete Infrastructure in Hungary in the Light of Climate Change
The limit state function (g) is formulated the following way:
min, 5()
cdur
g
cmmxt (7)
Where cmin,dur is the minimal required concrete cover due to durability requirements. The 5mm reduction
reflects the experimental fact that the depassivation occurs when the carbonation front reaches the 5mm
vicinity of rebar [9]. The reliability of three Eurocode exposure classes related to carbonation and the one ÚT
specification are examined. As an example a representative structure with design working life of 100 years
(tSL) within S6 structural class is taken for evaluating each durability measure. The parameters corresponding
to the above classifications are summed up in Table 2.
Table 2. Investigated concrete covers.
Standard Exposure class w/c1 c
min,dur [mm] tSL [yr]
Eurocode [8] XC2 0.60 35 100
XC3 0.55 35 100
XC4 0.50 40 100
ÚT [10] - 0.60 30 100
1follows the recommendations of [6].
Analysis Results and Discussion
The reliability calculations are performed with two cement types: CEM I 42.5 R which is ordinary Portland
cement with high early strength and CEM I 42.5 R+FA which is fly-ash containing (22% of cement) Portland
cement with high early strength. CEM I type is chosen since it is the most widely used cement type by the
construction industry, the two types are intended to illustrate the effect of additional constituent. Fig. 2.
shows the mean values of carbonation depth in time for various scenarios, cement types and exposure
classes.
variable CO
2
reference 2000
(ÚT) EC - XC2 - CEM I 42,5
R
Carbonation depth [mm]
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
0
5
10
15
20
25
Carbonation depth [mm]
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
0
5
10
15
20
25
Carbonation depth [mm]
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
0
5
10
15
20
25
Carbonation depth [mm]
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
0
5
10
15
20
25
Carbonation depth [mm]
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
0
5
10
15
20
25
Carbonation depth [mm]
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
0
5
10
15
20
25
A1B
A1FI A1T
A2
B1 B2
variable CO
2
reference 2000
EC - XC3 - CEM I 42,5
R
A1B
A1FI A1T
A2
B1 B2
variable CO
2
reference 2000
EC - XC4 - CEM I 42,5
R
A1B
A1FI A1T
A2
B1 B2
variable CO
2
reference 2000
(ÚT) EC - XC2 - CEM I 42,5 R+FA
A1B
A1FI A1T
A2
B1 B2
variable CO
2
reference 2000
EC - XC3 - CEM I 42,5 R+FA
A1B
A1FI A1T
A2
B1 B2
variable CO
2
reference 2000
EC - XC4 - CEM I 42,5 R+FA
A1B
A1FI A1T
A2
B1 B2
Fig. 2. Expected values of carbonation depth in time.
From the above figures it is clear that the predictions per climate change scenarios are quite coherent despite
that they represent significantly different future developments. Compared to the reference scenario the
expected carbonation depth increase is 10-21%. As Fig. 3. shows this moderate mean carbonation depth
Second Conference of Junior Researchers in Civil Engineering
Rózsás Á., Kovács N.: Carbonation of Concrete Infrastructure in Hungary in the Light of Climate Change
increment yields to significant (21-115%) increase in probability of depassivation.
tSL = 100 years
tSL = 100 years
A1FI
A1T
A2 B2
B1
EC - XC2 - CEM I 42.5
R
Probability of depassivation [%]
0
0.05
0.1
0.15
0.2
0.25
0.3
EC - XC2 - CEM I 42.5 R+FA
A1B
A1FI
A1B
A2
A1T
B1
B2
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Probability of depassivation [%]
0
0.05
0.1
0.15
0.2
0.25
0.3
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Probability of depassivation [%]
0
0.05
0.1
0.15
0.2
0.25
0.3
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Probability of depassivation [%]
0
0.05
0.1
0.15
0.2
0.25
0.3
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Probability of depassivation [%]
0
0.05
0.1
0.15
0.2
0.25
0.3
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Probability of depassivation [%]
0
0.05
0.1
0.15
0.2
0.25
0.3
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
tSL = 100 years
A1FI
A1FI
A1T
A1T
A2
A2
B2
B2
B1 B1
EC - XC3 - CEM I 42.5 R
A1B
A1B
tSL = 100 years
EC - XC4 - CEM I 42.5 R
tSL = 100 years
A1FI
A1FI
A1T
A1T
A2 A2
B2 B2
B1
B1
EC - XC3 - CEM I 42.5 R+FA
A1B A1B
tSL = 100 years
EC - XC4 - CEM I 42.5 R+FA
tSL = 100 years
ÚT - CEM I 42.5 R
Probability of depassivation [%]
A1FI
A1B
A2
A1T
B1
B2
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 2100
tSL = 100 years
A1FIA1T
A2
B2
B1
ÚT - CEM I 42.5 R+FA
A1B
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
Probability of depassivation [%]
Time [yr]
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
variable CO2
reference 2000
variable CO2
reference 2000
variable CO2
reference 2000
variable CO2
reference 2000
variable CO2
reference 2000
variable CO2
reference 2000
variable CO2
reference 2000
variable CO2
reference 2000
Fig. 3. Probability of depassivation for various exposure classes and climate change scenarios.
To evaluate the adequacy of concrete covers the typically applied 10% probability of failure can be used as a
limit value against depassivation [2]. It can be seen from Fig. 3. that only XC4 exposure class complies with
this criterion. It is important that even a 5mm cover difference can have a significant effect on the probability
of depassivation as can be seen by comparing EC-XC2 and ÚT cases (Table 2). As the figures show the
cement type has considerable effect on the carbonation depth and on the probability of depassivation as well,
CEM I 42.5 R+FA has higher diffusion resistance. This is due to that fly ash reduces the permeability of
concrete through pozzolanic reaction and improving the cement particles’ reactivity [11]. An extraction of
the results - comprising the maximum carbonation depth (xc) and depassivation probability (Pf) increases
- is presented in Table 3.
Table 3. Increases in carbonation depth (xc) and depassivation probability (Pf) compared to reference 2000.
CEM I 42.5 R CEM I 42.5 R+FA
Standard Exposure class
xc [%] Pf [%] xc [%] Pf [%]
Eurocode XC2 11 - 20 33 - 61 13 - 21 55 - 90
XC3 12 - 21 44 - 73 12 - 20 50 - 82
XC4 12 - 20 70 - 115 10 - 19 65 - 100
ÚT - 11 - 20 21 - 36 13 - 21 33 - 55
Second Conference of Junior Researchers in Civil Engineering
Rózsás Á., Kovács N.: Carbonation of Concrete Infrastructure in Hungary in the Light of Climate Change
Conclusions
Based on full-probabilistic, time-dependent reliability calculations the following main conclusions are
deduced: (i) climate change has significant effect on the carbonation of concrete infrastructure in Hungary
and its effect should be incorporated into standards; (ii) the calculations indicate that many present and
previous durability specifications are not sufficient to provide adequate resistance against depassivation even
with year 2000 CO2 level; (iii) compared to year 2000 level the carbonation depth is expected to increase by
12-21% depending on the particular climate change scenario. The results suggest that the expected relative
carbonation depth increase is very slightly sensitive to the material properties of concrete (Table 3.). The
probability of depassivation may increase by 115% to the end of the 21st century; (iv) the cement type has
considerable effect on the diffusivity of concrete and consequently on the probability of failure. In our case
CEM I 42.5 R+FA cement’s probability of depassvation is about half that of the CEM I 42.5 R.
The predictions can be improved by implementing the propagation period and by using more sophisticated
carbonation model. Moreover, in the light of new climate predictions the boundary conditions of corrosion
models should be updated. Based on the findings we recommend further research on the topic and the
evaluation of potential remedial actions in order to prevent the great economic burden of amplified
corrosion.
Acknowledgement
The work reported in the paper has been developed in the framework of the project „Talent care and
cultivation in the scientific workshops of BME" project. This project is supported by the grant TÁMOP-
4.2.2.B-10/1--2010-0009.
References
[1] Koch G.H., Brongers M.P.H., Thompson N.G., Virmani Y.P., Payer J.H.: Corrosion Costs and Preventive Strategies in the
United States, National Technical Information Service, Springfield, VA FHWA-RD-01-156, (2001).
[2] CEB-fib: Model Code for Service Life Design, Bulletin 34, Lausanne, Switzerland: International Federation for Structural
Concrete. 2883940746, (2006).
[3] IPCC: Climate Change 2007: Synthesis Report, Intergovernmental Panel on Climate Change, Geneva, Switzerland, (2007).
[4] Stewart M.G., Teply B., Králová H.: The effect of temporal and spatial variability of ambient carbon dioxide concentrations on
carbonation of RC structures, "9th International Conference on Durability of Building Materials and Components",
Brisbane, Australia, (2002).
[5] Rózsás Á., Kovács N.: Probabilistic Assessment of the Effect of Climate Change on the Carbonation of Concrete Structures in
the Carphatian Basin, "Creative Construction Conference 2013", Budapest, in press, (2013).
[6] CEN: Concrete - Part 1: Specification, performance, production and conformity, in EN 206-1, Brussells, Belgium: European
Committee for Standardization, (2000).
[7] Stewart M.G., Wang X., Nguyen M.N.: Climate change impact and risks of concrete infrastructure deterioration, "Engineering
Structures", Vol. 33, pp. 1326-1337, (2011).
[8] Eurocode 2: Design of Concrete Structures, (2004).
[9] Yoon I.-S., Çopuroğlu O., Park K.-B.: Effect of global climatic change on carbonation progress of concrete, "Atmospheric
Environment", Vol. 41, pp. 7274-7285, (2007).
[10] Magyar Útügyi Társaság: Design of Concrete, Reinforced Concrete and Prestressed Concrete Road Bridges. in Code of
Roadbridges 4, in Hungarian, (2002).
[11] Newman J., Choo B.S.: Advanced Concrete Technology 1: Constituent Materials, Butterworth-Heinemann, 978-0750651035
(2003).
