Conference PaperPDF Available

Chloride ingress in concrete blocks at the Rødbyhavn marine exposure site - Status after 5 years

Abstract and Figures

This paper presents new results from a long-running study following the chloride ingress in concrete blocks exposed at the marine exposure site located at Rødbyhavn harbour in Denmark. The site was established in 2010 as part of the preparatory work for the planned Fehmarnbelt fixed link between Denmark and Germany. Chloride profiles have previously been measured on cores extracted from the blocks after 0.5 and 2 years of exposure to seawater-and now after 5 years. The studied blocks represent 15 different concrete mixes produced using a variety of binder types. The chloride ingress in the blocks have been investigated for both submerged and splash zone exposure conditions. Chloride transport parameters such as the achieved chloride diffusion coefficient (D_a), surface chloride concentration (C s) and the penetration parameter K_Cr were determined by fitting the error function solution to Fick's 2nd law to the measured profiles. Results from the Rødbyhavn exposure site are also compared to chloride data from the field exposure site in Träslövsläge, Sweden. Based on the findings from Rødbyhavn and chloride data from the literature, a simplified chloride penetration model for long-term marine exposure is proposed and discussed.
Content may be subject to copyright.
CHLORIDE INGRESS IN CONCRETE BLOCKS AT THE RØDBYHAVN MARINE
EXPOSURE SITE – STATUS AFTER 5 YEARS
Søren Lundsted Poulsen(1), Henrik Erndahl Sørensen (1) and Ulf Jönsson(2)
(1) Danish Technological Institute, Taastrup, Denmark
(2) Femern A/S, Copenhagen, Denmark
Abstract
This paper presents new results from a long-running study following the chloride ingress in
concrete blocks exposed at the marine exposure site located at Rødbyhavn harbour in
Denmark. The site was established in 2010 as part of the preparatory work for the planned
Fehmarnbelt fixed link between Denmark and Germany.
Chloride profiles have previously been measured on cores extracted from the blocks after
0.5 and 2 years of exposure to seawater – and now after 5 years. The studied blocks represent
15 different concrete mixes produced using a variety of binder types. The chloride ingress in
the blocks have been investigated for both submerged and splash zone exposure conditions.
Chloride transport parameters such as the achieved chloride diffusion coefficient (Da),
surface chloride concentration (Cs) and the penetration parameter KCr were determined by
fitting the error function solution to Fick’s 2nd law to the measured profiles. Results from the
Rødbyhavn exposure site are also compared to chloride data from the field exposure site in
Träslövsläge, Sweden.
Based on the findings from Rødbyhavn and chloride data from the literature, a simplified
chloride penetration model for long-term marine exposure is proposed and discussed.
Keywords: Concrete durability, chloride ingress, field exposure, marine environment,
modelling.
1. INTRODUCTION
Availability of reliable data from long-running investigations of field-exposed concrete is
very important for a number of reasons. For instance, the durability design of large concrete
structures typically involves the use of models to predict the service life of the given structure.
Such models must be validated against long-term data from field-exposed concrete in order to
demonstrate their applicability for the type of concrete and exposure environment of the given
structure. Also, data from long-running studies of field-exposed concrete is one of the primary
keys to improve our knowledge on concrete durability, e.g. the resistance against chloride
ingress for concrete in marine environments.
In 2010 Femern A/S established a marine field exposure site in the harbour of Rødbyhavn
as part of the preparatory work for the coming Fehmarnbelt Fixed Link between Denmark and
Germany [1]. A total of 15 large concrete blocks with different concrete mix designs have
been exposed to seawater at the exposure site since 2010, and measurement of chloride
penetration in the blocks has been performed after 0.5, 2 and 5 years of exposure.
This paper gives an update on the measured chloride penetration data from the exposure
site with an emphasis on the most recent data obtained after 5 years exposure. Based on the
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
192
presented data from Rødbyhavn, and other chloride data from the literature, a simplified
chloride penetration model is also proposed and discussed.
2. EXPERIMENTAL
2.1 Design and production of concrete blocks
The 15 large concrete block exposed at the marine exposure site in Rødbyhavn were
produced with dimensions of 2 x 1 x 0.2 m (H x W x D). The 15 different mix designs for the
blocks are presented in Table 1 in terms of the nominal composition of constituent materials.
The mix designs were composed by various combinations of three different cements and three
different mineral additions. The cements are: Low alkali sulphate resistant Portland cement
CEM I 42.5 N – SR5 EA, Rapid hardening ordinary Portland cement CEM I 52.5 N, and blast
furnace cement CEM III/B 42.5 N (slag cement). The mineral additions are: Fly ash, silica
fume slurry, and ground granulated blast furnace slag. The nominal eqv. w/c-ratio are 0.40 for
all concretes except for concrete H and I, which have eqv. w/c-ratios of 0.45 and 0.35,
respectively. Further details regarding the design and production of the concrete blocks can be
found in Ref. [1] and [2].
2.2 Exposure and sampling
The 15 concrete blocks were exposed to seawater at the field exposure site in Rødbyhavn
at a maturity of 43-49 days. The blocks are placed partly immersed in seawater with the upper
70 cm above normal water level. The chloride content of the seawater is approx. 0.7%
chloride [3] and the annual temperature variations of the seawater are typically between -1ºC
and 20ºC. After 5 years of exposure, two Ø100 mm cores were drilled from each concrete
block for determination of chloride (and calcium profiles), i.e. one core from the submerged
zone and one core from the splash zone.
2.3 Profile grinding and chloride analysis
Two chloride profiles were determined for each of the extracted cores: One profile from
the west and one profile from the east facing side of the cores. The profiles were measured by
grinding off material in layers parallel to the exposed surface of the drilled concrete cores.
The chloride content of each layer was subsequently determined according to the procedure
given in DS 423.28, which is similar to NT BUILD 208 [4]. The chloride contents were
measured using potentiometric titration rather than Volhard titration. The measured chloride
profiles were subsequently corrected for potential variations in paste content as a function of
profile depth. This was done by using measured calcium profiles (not shown) in conjunction
with the chloride profiles.
3. RESULTS
3.1 Chloride profiles
Fig. 1 displays the chloride profiles measured after 5 years exposure in the submerged zone
for a selection of five representative concrete types: A (100% SPRC), B (85% SRPC + 15%
FA), E (96% SRPC + 4% SF), F (84% SRPC + 12% FA + 4% SF), and K (100% slag
cement). For comparison, the chloride profiles measured after 0.5 and 2 years are also plotted
for these concretes. Furthermore, Fig. 1 also includes a diagram where the 5-year profiles for
the five representative concretes are plotted together. A complete collection of all the chloride
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
193
profiles measured for concretes A-O after 0.5, 2 and 5 years can be found on the website
www.expertcentre.dk [1], including profiles both from submerged zone and splash zone. For
all profiles, a significant decrease in chloride content is observed in the outermost 2-5 mm.
It is noted that the 2-year profile for Concrete A generally shows chloride concentrations,
which are lower than the measured chloride profile after 0.5 years. We have no explanation
for this unexpected result, and we question the reliability of the 2-years profile. Consequently,
the 2-year data for Concrete A has been excluded in the following. It is also noted that the 0.5
years chloride profile for Concrete B shows an isolated peak at an ingress depth of 6 mm,
which we interpret as a consequence of a significant deviation from the “normal” ratio
between the amount of aggregates and cement paste in the concrete. This interpretation is
supported by the appearance of a significant negative peak at a depth of 6 mm in the calcium
profile measured on the same concrete core from Concrete B that was used for measuring the
chloride profile.
3.2 Chloride penetration parameters
Values for the calculated chloride concentration at the exposed concrete surface (Cs) and the
achieved chloride diffusion coefficient (Da) have been calculated for all the investigated
concrete types (A-O) at the Rødbyhavn exposure site according to the procedure given in NT
Build 443 [5]. This was achieved by fitting the error-function solution to Fick 2nd law of
diffusion to the measured chloride contents by means of a non-linear regression analysis in
accordance with the method of least squares fit. Generally, a number of points of the chloride
profiles nearest to the concrete surface were omitted in the regression analyses, i.e. all points
between the concrete surface and the point representing the highest measured chloride content
were not included in the fitting procedure. As an example, the fitted curves are shown along
with the measured chloride profiles for Concrete E in Fig. 2.
The penetration parameter KCr (also named the “first year penetration”) was determined for
the measured profiles according to equation (2):
=
is
rs
1
a
erf2 CC CC
DK
Cr (1)
In equation (1) Cr is a reference chloride concentration, which was set to a standard value of
0.05 wt% (of concrete). Calculated values of Da, Cs and KCr are given in Table 2 for five
representative types of concrete (A, B, E, F, and K) from the Rødbyhavn exposure site. A
collection of calculated values of Da, Cs and KCr for all concrete types (A-O) can be found on
www.expertcentre.dk [1]. A pronounced decrease of Da is generally observed for all the five
concrete types during the first five years of exposure, whereas Cs generally increases
somewhat for all the five concrete types during the same exposure period.
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
194
Table 1: Mix design for concrete blocks. The activity factors used to calculate the equivalent
w/c-ratios are 2.0 for silica fume and 0.5 for fly ash.
Concrete ID:
A B C D E F G H
Powder
composition
wt%
Low alkali SR cement CEM I 42.5 N 100 85 75 75 96 84 84 84
Rapid hardening cement CEM 1 52.5 N
Slag cement CEM III/B 42.5 N
Fly ash EN 450-1 N 15 25 25 12 12 12
Silica fume 50 %-wt slurry 4 4 4 4
GG blast furnace slag EN 15167-1
Concrete composition
Cement kg/m
3
365 322 300 336 340 300 310 276
Fly ash kg/m
3
57 100 112 43 44 39
Silica fume, solid matter kg/m
3
14 14 15 13
GGBFS kg/m
3
Water content l/m
3
146 140 140 157 147 140 145 145
Aggregate 0/2 kg/m
3
695 671 642 678 695 677 731 700
Aggregate 4/8 kg/m
3
377 374 367 349 377 377 386 380
Aggregate 8/16 kg/m
3
266 270 271 704 266 272 266 268
Aggregate 16/22 kg/m
3
529 538 541 529 543 530 534
Air entraining agent kg/m
3
1.7 1.7 2.3 4.0 0.7 1.6 0.0 1.5
Superplasticizer kg/m
3
2.8 2.3 2.2 2.9 2.7 2.9 3.82.
6
2.6
Eqv. w/c ratio - 0.40 0.40 0.40 0.40 0.40 0.40 0.40 0.45
Concrete ID:
I J K L M N O
Powder
composition
wt%
Low alkali SR cement CEM I 42.5 N 84 84 96
Cement CEM 1 52.5 N 30
Slag cement CEM III/B 42.5 N
100 100 100
Fly ash EN 450-1 N 12 12
Silica fume 50 %-wt slurry 4 4 4
GG blast furnace slag EN 15167-1 70
Concrete composition
Cement kg/m
3
330 350 360 375 410 108 340
Fly ash kg/m
3
47 50
Silica fume, solid matter kg/m
3
16 17 14
GGBFS kg/m
3
252
Water content l/m
3
135 163 144 150 164 144 147
Aggregate 0/2 kg/m
3
671 687 689 702 686 689 695
Aggregate 4/8 kg/m
3
374 354 373 381 353 374 377
Aggregate 8/16 kg/m
3
270 713 263 269 712 263 266
Aggregate 16/22 kg/m
3
538 525 535 525 529
Air entraining agent kg/m
3
2.3 2.2 0.8 0.0 1-6 1.0 0.0
Superplasticizer kg/m
3
3.6 3.4 2.3 2.6 2.9 2.9 3.7
Eqv. w/c ratio - 0.35 0.40 0.40 0.40 0.40 0.40 0.40
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
195
Figure 1: Chloride profiles measured on cores drilled from the permanently submerged part
of concrete blocks at the Rødbyhavn marine exposure site after 0.5, 2 and 5 years of exposure.
Two profiles were determined for each of the extracted cores after 5 years: One profile from
the west (W) and one profile from the east (E) facing side of the cores.
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
196
In Fig. 3, the 5-year values of Da and Cs for five representative concrete types are
compared to similar data determined from chloride profiles measured after 0.5 and 2 years of
exposure at the Rødbyhavn exposure site.
Figure 2: Measured chloride profiles for concrete E in the submerged zone
plotted along with calculated profiles obtained by fitting the error-function solution to
Fick’s 2nd law of diffusion to the measured profiles.
Figure 3: Achieved chloride diffusion coefficients (Da) (left) and surface chloride
concentrations (Cs) (right) after 5 years marine exposure for five representative concrete types
(A, B, E, F, and K) from the field exposure site in Rødbyhavn, Denmark. Values for Da and
Cs were obtained from chloride profiles measured on cores drilled from the submerged zone
of the exposed concrete blocks. Similar data determined after 0.5 and 2 years of exposure [2]
are shown as well for comparison.
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
197
4. DISCUSSION
4.1 Influence of binder type on chloride ingress
The data presented in Fig. 1, Fig. 3 and Table 2 demonstrates that the resistance against
chloride ingress for the investigated types of concrete is strongly influenced by the binder
composition used in the concrete mix. For example, the chloride profiles in Fig. 1 show that
the chloride ingress is significantly lower in Concrete K with slag cement compared to the
ingress observed for all the other concrete types. The same tendency is also reflected in the
very low values of Da observed for Concrete K (Fig. 3). The presented data generally suggest
that the investigated binder types may be ranked in the following order in terms of their
ability to ensure a high resistance against chloride ingress: (1) slag cement, (2) blends of
Portland cement + fly ash + silica fume (3) Portland cement + fly ash blends, (4) Portland
cement + silica fume blends, and (5) pure Portland cement. The improved resistance against
chloride ingress as a result of using mineral addition such as slag, fly ash or silica fume has
previously been confirmed in numerous studies (e.g. [2] and [6]).
4.2 Simplified model for chloride ingress in concrete structures
In a previous investigation dealing with the chlorid ingress in a number of Danish concrete
bridges in marine environment it was observed that the values for achieved chloride diffusion
coefficient (Da) and surface chloride concentration (Cs) appear to become more or less
constant for exposure times beyond 5 to 10 years [7]. These findings led to the suggestion of a
relatively simple model for chloride ingress into concrete, which is based on a linear
correlation between the ingress depth (xcr) of a given reference chloride concentration (cr) and
the square root of exposure time:
crcrcr btax += (2)
where acr is a factor of proportionality and bcr is the intercept with the y-axis in a plot of xcr
against the square root of exposure time (t). The value of acr is an expression of the rate of
chloride ingress of the reference chloride concentration (cr), while the value of bcr is
interpreted as a result of a “fast” initial ingress of chloride during the first few months (or
years) after the first exposure to the marine environment. Hydration reaction will typically
still occur during this early period, which means that the permeability of concrete will be
somewhat higher compared to the same concrete in a more mature state. The initial
penetration depth (bcr) could also partly be a consequence of initial capillary suction of
seawater at the time of the first exposure to a submerged marine environment. Therefore, bcr
will most likely be different for the same concrete depending of the maturity, as well as the
moisture content in the concrete, at the time of the first exposure to seawater.
Equation (2) can be rearranged for t:
2
=cr
crcrabx
t(3)
In principle, this equation can be utilized to estimate the time until initiation of reinforcement
corrosion in a concrete structure, i.e. the duration of the initiation phase, which is sometimes
used as a definition of the service life of a concrete structure. Such an estimation can be
obtained by equating the value of xcr with the thickness of the concrete cover above the
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
198
reinforcing steel and by setting the reference chloride concentration (cr) at a level equal to the
threshold value for initiation of chloride-induced reinforcement corrosion for the given
concrete.
Table 2: Initial chloride concentration in the concrete (Ci), chloride concentration at the
concrete surface (Cs), achieved chloride diffusion coefficient (Da) and penetration parameter
K0.05 determined for the 15 concrete types (A-O) at the marine exposure site in Rødbyhavn,
Denmark after 5 years of exposure. Parameters were determined for both submerged zone
(SUB) and splash zone (SPL), and for each combination of concrete type and exposure
environment, two sets of parameters were determined: One set calculated from each of the
chloride profiles measured on the west and east facing side of the extracted cores. Similar
parameters determined after 0.5 and 2 years exposure can be found in Ref. [2].
Con-
crete
ID
Exposure
environ-
ment
Ci
[wt% of
concr.]
CS
[wt% of concr.] Da
[*10-12 m2/s] K0.05
[mm/years0.5]
West East West East West East
A SUB 0.012 0.57 0.63 2.02 2.01 21 21
SPL 0.012 0.65 0.63 1.94 2.07 21 21
B SUB 0.012 0.84 0.81 1.08 0.93 16 15
SPL 0.012 0.86 1.00 1.02 1.14 16 18
C SUB 0.011 0.89 0.81 0.75 0.84 14 14
SPL 0.011 0.83 0.83 0.98 0.99 16 16
D SUB 0.013 0.95 0.83 0.60 0.28 13 8
SPL 0.013 0.94 0.95 0.84 0.71 15 14
E SUB 0.012 0.58 0.65 1.33 1.32 17 17
SPL 0.012 0.66 0.70 1.53 1.69 19 20
F SUB 0.014 0.77 0.55 0.70 0.77 13 13
SPL 0.014 0.72 0.72 0.80 0.78 14 14
G SUB 0.010 0.67 0.63 0.66 0.56 12 11
SPL 0.010 0.65 0.69 0.83 0.76 13 13
H SUB 0.012 0.65 0.65 1.07 0.80 15 13
SPL 0.012 0.69 0.67 1.17 1.21 17 17
I SUB 0.011 0.68 0.72 0.63 0.63 12 12
SPL 0.011 0.73 0.73 0.71 0.65 13 12
J SUB 0.010 0.80 0.76 0.74 0.80 13 14
SPL 0.010 0.99 0.99 0.73 0.68 14 13
K SUB 0.021 0.70 0.73 0.22 0.21 7.6 7.4
SPL 0.021 0.66 0.60 0.21 0.20 7.3 7.0
L SUB 0.020 0.67 0.65 0.19 0.15 6.8 6.1
SPL 0.020 0.69 0.71 0.22 0.21 7.4 7.3
M SUB 0.023 0.82 0.66 0.20 0.20 7.5 7.2
SPL 0.023 0.82 0.86 0.25 0.22 8.4 7.9
N SUB 0.012 0.53 0.36 0.19 0.20 6.4 5.7
SPL 0.012 0.80 0.87 0.19 0.26 6.8 8.2
O SUB 0.011 0.48 0.58 1.01 1.19 14 16
SPL 0.011 0.59 0.63 1.41 2.02 17 21
The suggested model for chloride ingress was initially derived from an analysis of (1) data
from the abovementioned investigation of chloride ingress in a number of Danish coast
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
199
bridges [7], and (2) data from 20 years marine exposure of concrete slabs in a Swedish field
exposure site [6]. In Fig. 4, measured chloride data from Rødbyhavn is used to further
examine the validity of the proposed ingress model. This is done by plotting the penetration
depth (x0.05) of 0.05 wt% chloride concentration against the square root of exposure time for
five of the investigated concrete types (A, B, E, F, and K). For each exposure time (0.5, 2 and
5 years), the values of x0.05 were determined by linear interpolation between the two data
points of each measured chloride profile being closest to 0.05 wt% chloride. It is generally
observed that the data for each concrete type plots more or less along a straight line, thus
supporting the linear correlation between x0.05 and t½ implied in Equation (2). A linear
regression analysis has been performed on the data set of each of the five concrete types
presented in Fig. 4, and the resulting regression lines are plotted here as well. The scale of the
x-axis lines is set to 10 years0.5 (= 100 years), which means that the penetration depths (x0.05)
after e.g. 100 years can be estimate directly in the plot by reading the value on the y-axis for
each regression line at x = 10 years0.5. The regression lines indicates that x0.05 after 100 years
exposure will be approx. 163 mm, 87 mm, 148 mm, 74 mm, and 58 mm for Concrete A, B, E,
F, and K, respectively.
Figure 4: Penetration depth of 0.05 wt% chloride concentration (x0.05) in submerged exposure
conditions as a function of the square root of exposure time. The plots includes data for five
selected concrete types (A, B, E, F, and K) from the Rødbyhavn marine exposure site. For
each concrete type a linear regression analysis has been performed on the plotted data and the
results are displayed as the dotted lines. The correlation coefficient (R2) from each regression
analysis is also shown along with the optimised parameters (a0.05 and b0.05) inserted in the
equation x0.05 = a0.05(t)0.5 + b0.05.
The validity of using the calculated regression lines in Fig. 4 as models for estimating the
chloride ingress in comparable concrete types from other locations has been examined by a
comparison with chloride ingress data from concrete slabs exposed at the marine exposure site
in Träslövsläge, Sweden [6,8]. In Fig. 5, the regression lines for concrete A, B, E, and F is
plotted together with extracted data sets of x0.05 for five selected concrete types (2-40, H8, H4,
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
200
10-40, and 12-35) from Träslövsläge. The selected concrete types have compositions that are
comparable to either Concrete A, B, E, or F, and the values of x0.05 were determined from
measured chloride profiles by linear interpolation between the two data points of each
chloride profile being closest to 0.05 wt% chloride.
Generally, each data set of x0.05 for the Träslövsläge concretes plots very close to the
relevant regression line from the analysis of the Rødbyhavn chloride data. Specifically, the
x0.05 values for Concrete 2-40 (100% PC) plots along the regression line for Concrete A
(100% PC), the x0.05 values for Concrete H8 (80% PC + 20% fly ash) plots close to the
regression line for Concrete B (85% PC + 15% fly ash), and the x0.05 values for Concrete 10-
40 (78.5% PC + 17% fly ash + 4.5% silica fume) and Concrete 12-35 (85% PC + 10% fly ash
+ 5% silica fume) plots close to the regression line for Concrete F (84% PC + 12% fly ash +
4% silica fume). An exception is the x0.05 values for Concrete H4 (95% PC + 5% silica fume),
which for some unknown reason show a less convincing compliance with the regression line
for Concrete E (96% PC + 4% silica fume).
Figure 5: Penetration depth of 0.05 wt% chloride concentration (x0.05) in submerged exposure
conditions as a function of the square root of exposure time. The plots includes data for five
selected concrete types (2-40, H4, H8, 12-35, and 10-40) from the marine exposure site in
Träslövsläge, Sweden [6,8]. The calculated regression lines for Concrete A, B, E, and F from
the marine exposure site in Rødbyhavn are included for comparison.
A linear regression analysis has also been performed on the data set of x0.05 for each of the
selected concrete types from Träslövsläge, and the optimised parameters (a0.05 and b0.05) from
the analyses are given in Table 3 along with the optimised parameters from the linear
regression analyses of the chloride data for Concrete A, B, E, F, and K. It is generally
observed that the concretes from Rødbyhavn and Träslövsläge with comparable binder
compositions have rather similar values for a0.05 and b0.05. Again, an exception is the two
concretes with a binder of Portland cement + silica fume (E (Rødbyhavn) and H4
(Träslövsläge)), which show somewhat diverse values for a0.05 (15 mm/years0.5 vs. 8
mm/years0.5) and b0.05 (1 mm vs. 9 mm). The reason for this discrepancy is unknown at this
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
201
point. Furthermore, validation against reliable long-term field data is generally still needed to
calibrate and further develop the suggested chloride ingress model.
Table 3: Penetration parameters a0.05 and b0.05 for concretes exposed in submerged marine
environment (see text for further explanation). The parameters were obtained by linear
regression analyses performed on the data plotted in Fig. 4 and selected chloride data from the
marine exposure site in Träslövsläge, Sweden [6,8]. SRPC = sulphate-resistant Portland
cement, OPC = ordinary Portland cement, FA = fly ash, SF = silica fume. *Calculated
assuming an efficiency factor of 2.0 and 0.5 for silica fume and fly ash, respectively.
**Calculated assuming an efficiency factor of 1.0 and 0.3 for silica fume and fly ash,
respectively [6,8].
Concrete ID Binder Eqv. w/c-
ratio a0.05
[mm/year0.5] b0.05 [mm]
A (Rødbyhavn) 100% SRPC 0.40 14.60 7.79
2-40 (Träslövsläge) 100% OPC 0.40 17.60 1.81
B (Rødbyhavn) 85% SRPC + 15% FA 0.40* 7.73 11.35
H8 (Träslövsläge) 80% SRPC + 20% FA 0.30** 4.65 11.05
E (Rødbyhavn) 96% SRPC + 4% SF 0.40* 14.67 0.96
H4 (Träslövsläge) 95% SRPC + 5% SF 0.40** 7.61 8.99
F (Rødbyhavn) 84% SRPC + 12% FA + 4% SF 0.40* 6.36 10.09
10-40 (Träslövsläge) 78.5% SRPC + 17% FA + 4.5% SF 0.40** 6.18 9.32
12-35 (Träslövsläge) 85% SRPC + 10% FA + 5% SF 0.35** 7.10 7.60
K (Rødbyhavn) 100% slag cement 0.40 5.59 1.63
5. CONCLUSIONS
The following conclusion can be drawn from the investigation of the most recent chloride
ingress data from the marine exposure site in Rødbyhavn, Danmark and from a comparison
with relevant chloride data from the field exposure site in Träslövsläge, Sweden:
-Among the studied concrete types from Rødbyhavn, the highest resistance against
chloride ingress is observed for Concrete K with a binder of slag cement.
-Intermediate resistance against chloride ingress is observed for concrete types with
binders consisting of Portland cement in combination with fly ash and/or silica
fume.
-Lowest resistance against chloride ingress is observed for Concrete A with a binder
of 100% Portland cement.
-Analysis of chloride data from submerged concretes at the exposure sites in
Rødbyhavn and Träslövsläge support the validity of a proposed model for
estimation of chloride ingress in concrete structures that is based on a simple square
root of time dependency.
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
202
REFERENCES
[1] http://www.expertcentre.dk/concrete/fehmarn-belt-exposure-site.aspx.
[2] Sørensen, H.E., Jönsson, U. and Munch-Petersen, C., ‘Chloride penetration into concrete –
Comparison of results from field exposure tests and laboratory tests’ in Proceedings of 2nd ICDC
(International Congress on Durability of Concrete), New Delhi, India, 4-6 December 2014.
[3] Andersen, I., ’Salt- og temperaturforhold i de indre danske farvande (Salt and temperature
conditions in the inner Danish waters)’, Danish Meteorological Institute, Copenhagen, Technical
Report No. 94-4, 1994.
[4] NT Build 208, ‘Concrete, Hardened: Chloride content by Volhard titration’, edition 3, Nordtest,
1996. Available at http://www.nordtest.info.
[5] NT Build 443, ‘Concrete, Hardened: Accelerated Chloride Penetration’, Nordtest, 1995. Available
at http://www.nordtest.info.
[6] Boubitsas, D., Luping, T. and Utgenannt, P. ‘Chloride Ingress in Concrete Exposed to Marine
Environment - Field data up to 20 years’ exposure’, SBUF report 12684, 2014, 1-137.
[7] Poulsen, S.L. and Sørensen, H.E., ‘Chloride ingress in old Danish bridges’ in Proceedings of 2nd
ICDC (International Congress on Durability of Concrete), New Delhi, India, 4-6 December 2014.
[8] Luping, T., ‘Chloride ingress in concrete exposed to marine environment Field data up to 10
years exposure’. SP Swedish National Testing and Research Institute, SP report 2003:16., 2003,
1-62.
4th International Conference on Service Life Design for Infrastructures (SLD4) –
27-30 August 2018 – Delft, Netherlands
203
... The purpose was to support the design and operation of the reinforced concrete structures for the fixed link. Chloride profiles taken from concrete subject to tidal and submerged exposure for five years at the Fehmarn Belt Exposure Site [1] and petrographic analysis of concrete subject to five years' tidal exposure [2] have already been reported. Recently, chloride ingress data from several exposure sites, including the Fehmarn Belt Exposure Site, have been compiled and analyzed using a square root of exposure time vs. chloride ingress depth approach [3]. ...
... For completeness calcium calibrated chloride profiles are shown in Appendix C. The equation used to calibrate for paste fraction is given in Eq. (1). ...
... With the exception of the outlier PC-S-W, all the chloride profiles measured after ten years show higher maximum chloride content (C max ) and deeper chloride penetration than for previous exposure times. Furthermore, when using the penetration depth of 0.1% chloride by wt. of concrete as a measure for the chloride ingress resistance, the ranking in chloride ingress resistance suggested for the blocks after five years by Poulsen et al. [1] is maintained after ten years of exposure (PC < 4SF < 15FA < 12FA4SF < 25 FA < SG). The increased resistance towards chloride ingress as a result of mineral additions and in particular GGBS is confirmed in numerous previous studies [64][65][66]. ...
Article
Full-text available
After ten years of marine exposure, chloride and calcium profiles and petrographic data were obtained from the tidal and submerged zones of six concrete panels differing in binder composition. Moisture and portlandite profiles were also determined on the submerged concrete. The data enables us to improve our understanding of the impact of sea water exposure and can also be used for service life modeling. The depth of the maximum chloride content and the depth of the microstructurally changed zone were comparable. Both depths progressed over time and reached a depth of as much as 10 mm after ten years of exposure. When using these and other field data for testing of chloride ingress prediction models, we recommend excluding datapoints from the microstructurally changed zone, i.e., the outermost datapoints including the maximum chloride content, unless reactive transport models are used.
... Poulsen and Sørensen [15] observed a linear relationship between a reference concentration of 0.05% chloride by mass of concrete and the square root of time and termed this the ''square root observation''. Poulsen et al. [16] showed an excellent correlation with experimental data of up to 20 years for submerged exposure in Träslövsläge Field Exposure Site in Kattegat and up to five years for submerged exposure at Fehmarn Belt Exposure Site. ...
... According to Poulsen et al. [16] the ingress depth of a reference concentration of 0.05% chloride by mass of concrete (x 0.05 ) can for their data be described by Eq. (1). ...
... where a 0.05 is the slope of a straight line, when x 0.05 is plotted against the square root of exposure time, and b 0.05 is the intercept. Poulsen et al. [16] interpreted a 0.05 as an indicator for the long-term chloride penetration rate of the reference concentration (C r-= 0.05% chloride by mass of concrete), while b 0.05 is interpreted as an indicator for the ingress relatively faster than a 0.05 occurring at early age. Hereafter the slope of the straight line for a C r is mentioned as a Cr (ingress parameter), the intercept as b Cr (early ingress depth), and the ingress depth as x Cr . ...
Article
Full-text available
A recent observation showed a square root time dependency of the ingress depth of a fixed (reference) chloride concentration of 0.05% chloride by mass of concrete for submerged exposure in Kattegat and the Baltic Sea. The purpose of this paper is to assess the applicability and limitations of the observation, widen the scope of validity and propose it as a method. Field data from submerged, tidal, splash, atmospheric and inland deicing salt exposure at various geographical locations was analyzed at a range of reference concentrations. In total 237 combinations of concrete, exposure, and reference concentration were analyzed. Our results showed that chloride ingress of a reference concentration followed a linear relationship with an average R^2 of 0.96, when the penetration depth of the reference concentration was plotted against the square root of the exposure time. The square root observation appeared valid for the studied Portland cement based concretes with fly ash, silica fume and ground granulated blast furnace slag exposed in submerged and tidal exposure zones, when applying reference concentrations of 0.1-1.8% chloride by mass of binder, and reference concentrations of 0.1-0.5% chloride by mass of binder in atmospheric exposure zone. It was found that the parameters describing the straight line depended on the chosen reference concentration and concrete composition, and that the slope of the straight line (ingress parameter) in addition depended on the exposure. It was concluded that the square root method appears to be a promising method for predicting further chloride ingress into concrete.
... Composition of concretes exposed at the Fehmarn Belt Exposure Site [kg/m 3 ][13]. SG = Ground granulated blast furnace slag cement (slag content: 67 wt.%) ...
Article
Full-text available
Due to stochastic and systematic variations in the paste fraction, data for total chloride content are occasionally calibrated using parallelly measured calcium content as a measure of the actual paste fraction − assuming non-calcareous aggregates and no calcium leaching. Data from concrete exposed at the marine Fehmarn Belt Exposure Site questions the latter assumption. In the outer zone experiencing calcium leaching (ten mm after ten years), errors will be introduced by calcium calibration. To account for the wall effect, calcium profiles from cores taken before exposure might be used to correct for the systematically higher paste fraction at cast surfaces.
Preprint
Full-text available
Clinker substitution is limited by the availability of suitable supplementary cementitious materials. Therefore, flexible methods must be adopted by practitioners to assess lab performance that can be extrapolated to long-term field conditions. In this study, we assess the square root law for chloride ingress in cementitious materials using a single-species reactive transport model and new analysis of published experiments. Our results illustrate the suitability of the square root law as a first approximation extrapolation tool. As the square root law is defined by a diffusion-controlled process, this analysis opens the door to even quicker methods which are related to the effective diffusion coefficient.
Article
Results of more than 5 years of in situ reinforcement corrosion monitoring are presented including continuous measurements of concrete resistivity, temperature, and open circuit corrosion potential measurements. Depth dependent resistivity and temperature measurements were obtained by means of multi-ring electrodes, while so-called instrumented rebars were used for location- and time-dependent open circuit corrosion potential measurements. The presented results highlight the complex interaction between mass transport, fracture, exposure, and electrochemistry and underline the importance of in situ measurements, which are imperative for model testing to enable accurate and reliable long-term performance predictions of reinforced concrete structures.
Conference Paper
Full-text available
Femern A/S, the Owner of the coming Fehmarnbelt Fixed Link, has built a marine field exposure site in the harbour of Roedbyhavn as a part of the preparations for the coming 17.6 km long immersed tunnel between Denmark and Germany, which will be the longest combined rail and road tunnel to date. A total of 15 large concrete blocks (2 x 1 x 0.2 m) with different concrete mix designs and 15 small concrete blocks (1 x 1 x 0.2 m) with concrete mix designs matching the large blocks were produced in 2010 by the Danish Technological Institute (DTI is external concrete laboratory for Femern A/S). The large blocks were placed partly submerged in seawater in Roedbyhavn field exposure site and the small blocks were placed outside the laboratory. The 15 mix designs include concrete with three different cement types (two Portland cements and one blast furnace cement) and three different pozzolans (silica fume, fly ash and blast furnace slag). The field exposed concrete blocks will be monitored at least until the end of the construction period (2021). Chloride penetration profiles have already been measured after ½ years and 2 years field exposure in immersed zone and in splash zone. The results are presented and compared to results from accelerated laboratory testing of samples from the small blocks by the chloride migration test according to NT BUILD 492 at the maturity age of 1 month and 6 months.
Conference Paper
Full-text available
The ingress of chloride into three Danish bridges located in marine environments has been studied by means of chloride profiles measured on concrete cores from submerged conditions. The sample materials represent three different concrete types and exposure times ranging from 30 to 34 years. Chloride transport parameters such as the achieved chloride diffusion coefficient (Da), surface chloride concentration (Cs) and the penetration parameter KCr were determined by fitting the error function solution to Fick's 2 nd law to the measured profiles. Comparison with data from earlier investigations shows that Da and Cs are almost constant for exposure times beyond ten years. The findings from the old Danish bridges are also compared to data from 20 years marine exposure of concrete slabs in a Swedish field exposure site, and the possibility to apply a simplified chloride penetration model for long term marine exposure is discussed.
Salt-og temperaturforhold i de indre danske farvande (Salt and temperature conditions in the inner Danish waters)', Danish Meteorological Institute
  • I Andersen
Andersen, I., 'Salt-og temperaturforhold i de indre danske farvande (Salt and temperature conditions in the inner Danish waters)', Danish Meteorological Institute, Copenhagen, Technical Report No. 94-4, 1994.
Concrete, Hardened: Chloride content by Volhard titration
NT Build 208, 'Concrete, Hardened: Chloride content by Volhard titration', edition 3, Nordtest, 1996. Available at http://www.nordtest.info.
Concrete, Hardened: Accelerated Chloride Penetration
NT Build 443, 'Concrete, Hardened: Accelerated Chloride Penetration', Nordtest, 1995. Available at http://www.nordtest.info.
Chloride Ingress in Concrete Exposed to Marine Environment -Field data up to 20 years' exposure
  • D Boubitsas
  • T Luping
  • P Utgenannt
Boubitsas, D., Luping, T. and Utgenannt, P. 'Chloride Ingress in Concrete Exposed to Marine Environment -Field data up to 20 years' exposure', SBUF report 12684, 2014, 1-137.
Chloride ingress in concrete exposed to marine environment -Field data up to 10 years exposure'. SP Swedish National Testing and Research Institute
  • T Luping
Luping, T., 'Chloride ingress in concrete exposed to marine environment -Field data up to 10 years exposure'. SP Swedish National Testing and Research Institute, SP report 2003:16., 2003, 1-62.