30 by stariya


Wolfgang Brameshuber

Institute of Building Materials Research (ibac), RWTH Aachen University, Germany

   Frost damage in concrete is in principal caused by exceeding the critical degree of water
saturation [1]. To obtain information on the water saturation in members of concrete buildings
in connection with freezing of water, the humidity and temperature of concretes has been
monitored in different bridges, locks, tunnels etc. The so-called multiring electrode measuring
system has been used in connection with temperature sensors. The buildings cover the exposi-
tion classes XF1 to XF4 according to EN 206-1. The continuously monitoring of the electro-
lytic resistivity and the temperature data must be converted into water saturation values using
calibration curves determined in the laboratory. For a period of up to 6 years, the degree of
water saturation has been observed. From the data the long-term behaviour as well as short-
term occasions influencing the water saturation degree can be evaluated. For the prediction of
the service life of concrete in the case of frost resistance the data are compelling to describe
the action part in building constructions.

    Besides the descriptive specification in the form of limit values concerning the minimum
cement content and the water-to-cement ratio, the present European concrete standard EN
206-1 [2] also includes the possibility of the verification of equivalent concrete properties.
According to the CEB-FIB Model Code for Service Life Design [3], one possibility to realise
this concept consists in the application of fully or partly probabilistic durability models in
which the action is compared to the material resistance and the probability of failure is as-
sessed. As regards the durability of concrete, respective concepts concerning the penetration
of chlorides and the carbonation were developed in the past, e.g. [4-6]. For frost with and
without de-icing agents test methods as laboratory tests do exist, the frost attack, however, is
directly related to the pore structure of the concrete. Therefore, action and resistance influence
each other directly. The additional water absorption resulting from the so-called micro ice
lens pump [7] at the thawing of the water frozen in the pores is controlled by the water uptake
and the temperature (action) as well as by the pore structure (resistance). Hence, it must be
questioned if critical degrees of water saturation subject to the exposure class are reached at
   The method presented in this paper answers the question under which circumstances criti-
cal degrees of saturation in the concrete pore structure do actually occur in the building and
how buildings behave in the long term under different exposures. For this purpose, different
buildings - two tunnels, two bridges, two locks, one sewage plant and one quay wall - were
equipped with continuous measuring instruments. Up to now, the development of the satura-
tion degree in the different concretes has been observed over a period of up to 6 years. The
measurements were taken with the multiring electrode (MRE) [8] developed at the Institute of
Building Materials Research (ibac) of RWTH Aachen University. In connection with the pre-
cise recording of the temperature, the MRE can continuously and depth-dependently record
the electrolytic resistivity. Thus, the saturation developing in the building can be determined
with calibration curves which establish a correlation between the water content in the concrete
and the electrolytic resistivity. In this paper selected measuring results are presented and some
important aspects of the application of the MRE are described. The procedure for installation
the MRE and determination of calibration curves are given in [9-11].

                    degree of saturation
                                                                         calibration function:
              0.3           concrete Gäubahn tunnel, water
              0.2           concrete Gäubahn tunnel, NaCl
                            calibration function Gäubahn tunnel, water
                            calibration function Gäubahn tunnel, NaCl
                    1         10           100          1000       10000        100000 1000000
                                                       specific electrolytic resistivity in Wm
Figure 1: Correlation between the specific electrolytic resistivity and the degree of saturation
for a tunnel concrete

   At the calibration, a difference was made between exposure to water and to a solution of
3% NaCl because an influence of the electrolytic resistivity was to be anticipated as a result of
the chloride exposure in the street tunnel. In the present case, the conductivity is not increased
by the higher ion content. The constants for the calibration function can be gathered from Ta-
ble 1.
   For the subsequent installation, the calibration curves cannot be determined directly with
TEM as the resistance is influenced by the connecting grout. A method was developed regard-
ing the adsorption isotherm of concrete and connecting grout. As, due to failure, one measur-
ing point in the entrance area of the a. m. street tunnel had to be reinstalled after completion
of the tunnel, a calibration curve was also developed applying the method of subsequent in-
stallation for the same concrete. This method is illustrated in Figure 2.
Table 1: Constants for the calibration function according to Figures 4 and 6
          constant                                   direct installed                       subsequently installed
             -                               H2O                         NaCl                       H2O
             a                                8.5                        17.1                       85.9
             b                                220                         220                        518
             c                               3647                        3647                      -16477
             d                                4.5                         0.2                       24.0
             e                               -0.4                        -0.8                        -1.7

     water content in % by mass                                         water content of connecting grout in % by
12                                                                 12
                                                                        mass                   value pair of water
               connecting grout, water                                                         absorption under
10             concrete Gäubahn tunnel, water                      10                          atmospheric pressure

 8                                                                  8

 6                                                                  6

 4                                                                  4

 2                                                                  2                               measured values
 0                                                                  0
     30   40      50    60     70       80 90 100                       0     2      4      6      8     10    12
                                    rel. humidity in %                       water content concrete in % by mass

Figure 2: Adsorption isotherm of concrete and connecting grout at 20 oC (left) and derived
pairs of values at the same relative humidity, each, as well as water absorption under atmos-
pheric pressure (right)

   The determined correlation in connection with the calibration curve of the connecting grout
serves to gather the water content in the concrete from the electrolytic resistivity measured at
the building. The corresponding procedure is shown in Figure 3. If the calibration function of
the connecting grout is known (Figure 3, right), the water content of the concrete can finally
be gathered from the electrolytic resistivity of the connecting grout with the correlation be-
tween the water content of the connecting grout and that of the concrete which has already
been illustrated in Figure 2.
   Finally, the respective specific electrolytic resistivity of the connecting grout must be as-
signed to the adsorption isotherm pairs and thus to the water content of the old concrete at a
specific relative humidity. This results in the calibration curve for the correlation between the
specific resistance of the connecting grout and the saturation degree in the concrete shown in
Figure 4. The corresponding constants of the function are listed in Table 1.
     value pair of water
                                                                                                                water absorption under 150 bar
     absorption under
     atmospheric pressure                                                     12                                water absorption under atmospheric
                                                                                                                      calibration function:
                                                                                4            connecting grout,
                                       measured values                          2            calibration function
                                       trendline                                             of connecting grout
 0        2          4       6      8      10      12                               1                 100                 10000             1000000
                      water content of concrete in %                                        specific electrolytic resistivity in Wm
Figure 3: Calibration curve of the connecting grout (right) as well as correlation between wa-
ter content of connecting grout and concrete (left)

                                   degree of saturation in the concrete
                                                                                        calibration function:
                            0.90                                                        y=a/(ln(b*x+c)-d)+e
                            0.20                calibration function
                            0.10                combined measured values
                                   1           10          100         1000        10000     100000 1000000
                                                         specific resistivity of the connecting grout in Wm

Figure 4: Calibration curve to determine the water saturation degree in the concrete at subse-
quently installed MRE

   The resistivity measured at the building can be converted into water saturation degrees ap-
plying a temperature correction [11] and the respective calibration curves. For the direct as
well as for the subsequent installation, Figure 5 shows the saturation degrees in the area of the
tunnel side walls, i. e. about 1 m above the top of the road surface, which have been observed
over a period of 4.5 years. The saturation degrees directly at the entrance area (subsequent
installation) as well as 100 m into the tunnel (direct installation) are displayed.
   At this point, the chloride effect was not taken into account. In this case, the maximum oc-
curring error is smaller than 10%. Considering the expected accuracy of the chosen measuring
method, it is justified to disregard this influence. In general, the following findings can be
derived from the temporal development of the saturation degrees: At the beginning of the
measurements, relatively high saturation degrees occur at a larger depth. With utmost prob-
ability, on the one hand, they are to be ascribed to hydration effects – the concrete contains fly
ash and thus reacts more slowly, the calibration curve was determined at a concrete age of one
year - on the other hand, these saturation degrees are due to drying effects occurring in the
course of time. Both methods lead to comparable degrees of saturation at the same concrete.
          estimated degree of saturation     concrete temperature in °C                     estimated degree of saturation      concrete temperature in °C
   1.2                                                                    50          1.2                                                                    50
                      7 mm             12 mm               17 mm                                        7 mm             12 mm              17 mm
   1.1                                                                                1.1
                      22 mm            27 mm               42 mm                                        22 mm            27 mm              42 mm
                      concrete temperature                                40                            concrete temperature                                 40
   1.0                                                                                1.0
                        saturation under atmospheric pressure                                              saturation under atmospheric pressure
   0.9                                                                                0.9
                                                                          30                                                                                 30
   0.8                                                                                0.8

   0.7                                                                    20          0.7                                                                    20
   0.6                                                                                0.6

   0.5                                                                    10          0.5                                                                    10

   0.4                                                                                0.4
                                                                          0                                                                                  0
   0.3                                                                                0.3

   0.2                                                                                0.2
                                                                          -10                                                                                -10
   0.1                 Gäubahn tunnel, MP1, entrance area,                            0.1
                       subsiquently installed                                                       Gäubahn tunnel, MP3, entrance area + 100 m
   0.0                                                                    -20         0.0                                                                    -20
    01.03.02      01.03.03    29.02.04     28.02.05    28.02.06    28.02.07           01.03.02     01.03.03     29.02.04     28.02.05   28.02.06    28.02.07

   Figure 5: Temporal development of saturation degrees in a street tunnel at the entrance
area (left) and 100 m into the tunnel (right)

   It must be taken into account that there are, however, some inaccuracies at the determina-
tion of the adsorption isotherms with the method of the subsequent installation and that, up to
now, no experience could be gathered elsewhere. In both cases, events at different times can
be indicated which hint at a fast increase in the saturation degree. Therefore, for the measur-
ing point located 100 m into the tunnel, shortened periods of the winters of 2004 and 2006
were evaluated in connection with the precipitation data gathered at a meteorological station
nearby. The result is demonstrated in Figure 6.
            degree of               concrete temperature in °C                            degree of               concrete temperature in °C
            saturation                   precipitation in mm/d                            saturation                   precipitation in mm/d
    1.2                                                                       100   1.2                                                                100
                         7 mm                         12 mm                                            7 mm                      12 mm
    1.1                  17 mm                        22 mm                   90    1.1                17 mm                     22 mm                 90
                         7 mm NaCl                    12 mm NaCl                                       7 mm NaCl                 12 mm NaCl
    1.0                  precipitation                concrete temp.          80    1.0                precipitation             concrete temp.        80
    0.9               saturation under atmospheric pressure                   70    0.9          saturation under atmospheric pressure                 70
    0.8                                                                       60    0.8                                                                60
    0.7                                                                       50    0.7                                                                50
    0.6                                                                       40    0.6                                                                40
    0.5                                                                       30    0.5                                                                30
    0.4                                                                       20    0.4                                                                20
    0.3                                                                       10    0.3                                                                10
    0.2                                                                       0     0.2                                                                0
    0.1        Gäubahn tunnel, MP3, entrance area + 100 m
                                                                              -10   0.1      Gäubahn tunnel, MP3, entrance area + 100 m
    0.0                                                                       -20   0.0                                                                -20
    01.12.03         01.01.04       01.02.04 03.03.04              03.04.04         19.12.05       19.01.06       19.02.06 22.03.06            22.04.06

  Figure 6: Observed development of the saturation degrees 100 m into the tunnel during the
winters of 2004 (left) and 2006 (right)

   The calculated degree of saturation is shown for the edge area in a depth of 7 and of
12 mm with and without consideration of the influence of the chlorides on the electrolytic
resistivity. The short-term, significant increases of the saturation degree illustrated in this pe-
riod are only observed during a thawing process at simultaneous precipitation events. Here, it
must be pointed out that the b-value for correction of temperature effects [11] was chosen to
be constant. In the following chapter, it is demonstrated that this assumption is not correct and
that it should be evaluated especially at high saturation degrees with low b-values, see also
[12]. Hence, the distinct exceeding of the saturation under atmospheric pressure does not cor-
respond to reality. If a saturation-dependent b-value was considered, lower saturation degrees
would be calculated, the results, however, hint at a frost suction corresponding to the model of
the micro ice lens pump [7]. Subsequent to the period of the fast water absorption the concrete
dries again so that a further saturation cannot take place.

4.1 Dependence of constant b on moisture content
   The gathered resistance data were converted into electrolytic resistivity assuming a b-value
independent of the saturation. As this b-value also depends on concrete technological parame-
ters, it has to be redetermined for each concrete of a building. For this purpose, the measuring
data of larger measuring depths are used where a short-term change in the water saturation is
not to be expected, nevertheless temperature changes are possible. Figure 7 exemplarily
shows the determination of the b-value. Because of the building data, it is ensured that
changes in the resistance are not caused by changes in the moisture content of the concrete.
While the time interval on May 1, 2004 enables an evaluation, the time interval on May 5,
2004 is unsuited which is also illustrated by the coefficient of determination. Generally, b-
values were calculated at the evaluation of the building data when the coefficients of correla-
tion were larger than 0.90. All other evaluations with smaller coefficients of correlation are
subject to additional influences as e. g. the change in the water saturation degree.

        temperature in °C       resistance in W                   logarithm of resistance
   35              temperature 01.05.2004
                                                  135000   11.5
                   temperature 05.05.2004                              series of measurment 01.05.2004
   30              resistance 01.05.2004          120000               series of measurment 05.05.2004
                   resistance 05.05.2004                                    y = 1778.1x + 4.8725
   25                                             105000                         R = 0.2058
   20                                             90000

   15                                             75000
   10                                             60000                   y = 5197x - 6.9637
                                                                             R = 0.9234

    5                                             45000
    0                                             30000      0.00334     0.00338 0.00342 0.00346 0.00350
     0:00   4:00   8:00 12:00 16:00 20:00 0:00                                 reciprocal temperature in 1/K
   Figure 7: Evaluation of temperature compensation and determination of b-value

   The tests with TEM which were conducted to determine the calibration function [9], [10],
[11] can also be used to determine the b-values. As a saturation degree was adjusted in a tar-
get-oriented way, the b-value can directly be derived from measurements at different concrete
temperatures. For the concrete of the street tunnel, Figure 8 displays the dependence of the b-
value on the saturation degree.
                                                                                                   resistance in W         temperature in °C
                                                                                         1000000         7 mm                 12 mm                  45
                                                                                                         17 mm                27 mm
                                                                                                         87 mm                concrete temperature   40
                                                                                                         air temperature      water temperature
        b-value in K                             coefficient of correlation                                                                          35
7000                                                                          1.00                  lock Hohenwarthe, MP3, h=56.0 m
                                                                                          100000                                                     30
6000                                                                                                                                                 25
5000                                                                          0.94                                                                   20
                                                                              0.92         10000                                                     15
                                                                              0.90                                                                   10
3000     concrete                                                             0.88                                                                   5
         Gäubahn tunnel
2000                                                                          0.86          1000                                                     0
             b-values, r>0.98                                                 0.84
1000                                                                                                                                                 -5
             coefficient of correlation                                       0.82
   0                                                                          0.80
       0,0   0,1    0,2     0,3    0,4    0,5   0,6   0,7 0,8 0,9 1,0                        100                                                   -15
                                                      degree of saturation                   18.01.06        21.01.06       24.01.06         27.01.06

Figure 8: Saturation-dependent b-values de-                                          Figure 9: Increase in resistance in the build-
termined with TEM                                                                    ing concrete caused by freezing of pore water

   Therefore, the saturation degrees shown in Chapter 3 should be corrected by an iterative
calculation method. At high saturation degrees, lower values would result after conducting a
corresponding correction calculation so that the indicated values concerning the development
of the moisture balance in the building are on the unfavourable, i. e. on the safe side. Particu-
larly for the area near the surface, there are maximum deviations of about 20% of the satura-
tion degrees actually existing in the building concrete.

4.2 Freezing processes
   In the presented example of the street tunnel, the saturation degrees even in the near sur-
face area are normally that low that freezing processes, i. e. the abrupt increase in resistance,
are not observed. In buildings as e. g locks, even the larger pores are filled with water, satura-
tion degrees near or slightly above the saturation at atmospheric pressure are found. Figure 9
shows such a phenomenon of abrupt change in resistance at the example of a lock chamber.
The measuring point is located directly below the headwater. On January 23, 2006 as well as
on January 24, 2006 the resistance increases abruptly which could at first be interpreted as
desiccation. Laboratory tests regarding this phenomenon show that the water freezing in the
pores is responsible for the change in resistance because of the low temperatures at a simulta-
neously occurring, relatively high saturation of the concrete [13]. When converting the resis-
tance into saturation degrees disproportionately low values would be determined temporarily.
Regarding a subsequent thawing process at simultaneous activation of the micro ice lens
pump [7], an increase in saturation is conceivable if this process is repeated several times.
Only then, respective frost damage can actually occur.

4.3 Carbonation
   A possible carbonation of the concrete significantly influences the electrolytic resistivity.
Therefore, it must be known to which extent the surface area of the concrete is actually car-
bonated. In any case, corresponding calibration curves of the non-carbonated as well as the
carbonated concrete must be developed. Figure 10 exemplarily shows the calibration curves
of a carbonated and a non-carbonated concrete of a quay wall (fender). The concrete was ex-
posed to water as well as to a solution of 3% NaCl.

                             degree of saturation
                                                                        fender concrete, water, n. c.
                       0,9                                              calibration function, water, n. c.
                       0,8                                              fender concrete, water, c.
                                                                        calibration function, water, c.
                                                                        fender concrete, NaCl, n. c.
                       0,6                                              calibration function, NaCl, n. c.
                       0,5                                              fender concrete, NaCl, c.
                                                                        calibration function, NaCl, c.
                                 calibration function:
                       0,1       y=a/(ln(b*x+c)-d)+e
                             1                 100       10000        1000000        100000000
                                                           specific electrolytic resistivity in Wm

Figure 10: Calibration curves for carbonated (c.) and non-carbonated (n. c.) concrete when
exposed to water or to NaCl solution

   The results show that the correlation between the saturation degree and the specific electro-
lytic resistivity is influenced by a previous carbonation. This has to be ascribed to the change
of the structure resulting from carbonation. This influence is higher than that of a sodium
chloride solution on a non-carbonated concrete. The ion content of this concrete is obviously
that high that the additional ions from the NaCl exert hardly any influence on the calibration
curve. If carbonated concrete is exposed to a sodium chloride solution instead of water, the
difference between both calibration curves is considerably higher due to the ion content in-
creased by NaCl. At lock buildings the concrete surface area is carbonated because a longer
period of time expires until the lock is operated for the first time. This entails the additional
effect that, because of the subsequent exposure to water over many years, even if a change in
structure took place as a result of carbonation, the surface area, however, was realkalised si-
multaneously. This has to be taken into account at the determination of the carbonation depth
with simple tests. Otherwise, there would be no correct values of the measured saturation de-
grees in the building.

   The present paper reports on the determination of the water content/saturation degree of
concrete in different buildings. The determinations are based on the measuring of the electro-
lytic resistivity and the calibration using laboratory tests. The method and the obtained meas-
uring values are explained at the example of a street tunnel. In the present case, no degrees of
water saturation which could be regarded as problematic for a frost attack were determined at
the entrance area as well as in the tunnel. The vertical elements of tunnel buildings are nor-
mally classified in exposure class XF2. This entails a maximum water-to-cement ratio of 0.5
if no air entraining agent is applied. The measurements illustrate that the classification into
class XF2 is justified despite the relatively extreme weather conditions in the entrance area.
The method of measuring the electrolytic resistivity in the building has meanwhile been well
proven over many years thanks to its robustness. The evaluation of the data must however
take into account some boundary conditions as the dependence of the constant b on the satura-
tion degree, the carbonation of the concrete surface area at a possible simultaneous exposure
to salt and the freezing of the pore solution. This leads to a distinctly more accurate assess-
ment of the development of the water content in the concrete, especially if it is less the long-
term development but the single events that shall be considered. Measurements at buildings
with the most different exposures as tunnels, bridges, locks, quay walls and sewage plants
demonstrated that a critical water saturation of the concretes applied today does obviously not
occur under exposure according to XF1/XF2. Only when exposed to XF3/XF4 conditions,
water saturation degrees are determined which are within the range of atmospheric water satu-
ration and which even sometimes exceed them at single events. As a rule, the concretes ap-
plied today are sufficiently dimensioned for these exposure classes. Minor deviations, e. g. in
the water-to-cement ratio, could then however lead to an inadequate frost resistance.

[1]    Fagerlund, G.: On the Service Life of Concrete Exposed to Frost Action. London: E & FN Spon, 1997. -
       In: Freeze-Thaw Durability of Concrete. Proceedings of the International Workshop, Sainte-Foy, Que-
       bec, 1997, (Marchand, J.; Pigeon, M. ; Setzer, M.(Ed.)), S. 23-41
[2]    DIN EN-206-1:2001-07; Concrete – Part 1: Specification, performance, production and conformity;
       German version EN 206-1:2000
[3]    Schießl, P.; fib ; Model Code: Model Code for Service Life Design. Lausanne: Federation
       Internationale du Beton, fib. - In: Bulletin (2006), Nr. 34
[4]    Büteführ, M. ; Fischer, C. ; Gehlen, C. ; Menzel, K. ; Nürnberger, U.: On-Site Investigations on
       Concrete Resistivity - A Parameter of Durability Calculation of Reinforced Concrete Structures. In: Ma-
       terials and Corrosion 57 (2006), Nr. 12, S. 932-939
[5]    Schießl, P.; Gehlen, C. ; Pabsch, G.: Prediction of Service Life of Existing Structures by Applying
       Probalistic Service Life Design (SLD) Procedures. Bagneux : RILEM, 2003. - In: Advances in Concrete
       and Structures, Proceedings of the International Conference ICACS 2003, Xuzhou, Jiangse, China, 17-
       19 September 2003, (Yuan, Y. ; et al (Ed.)), Vol. 1, S. 593-601
[6]    Schießl, P.; Gehlen, C.: New Approach of Service Life Design for Concrete Structures. In: Restoration
       of Buildings and Monuments 12 (2006), Nr. 4, S. 285-298
[7]    Setzer, M.J.: Modelling and Testing the Freeze-Thaw Attack by Micro-Ice-Lens Model and CDF/CIF-
       Test. Proceedings of the International Workshop on Microstructure and Durability to Predict Service
       Life of Concrete Structures Sapporo, Japan (2004) p. 17-28. Hokkaido University (2004)
[8]    Raupach, M.: Zur chloridinduzierten Makroelementkorrosion von Stahl in Beton. Berlin : Beuth. - In:
       Schriftenreihe des deutschen Ausschusses für Stahlbeton (1992), Nr. 433 = Dissertation Stand: Y 925
[9]    Brameshuber, W.; Spörel, F.: Measuring the humidity content in concrete buildings in dependence of
       depth, (in German). Berlin : Deutscher Ausschuss für Stahlbeton, 2005. - In: Beiträge zum Kolloquium
       29. und 30. September 2005 im Forschungsinstitut der Zementindustrie in Düsseldorf, S. 83-90
[10]   Spörel, F. ; Brameshuber, W.: Monitoring of Moisture Distribution in Concrete Structures. Zürich: Inter-
       national Vereinigung für Brückenbau und Hochbau, ETH-Hönggerberg, 2006. - In: Conference on Op-
       eration, Maintenance and Rehabilitation of Large Infrastructure Projects, Bridges and Tunnels, Proceed-
       ings of IABSE, Copenhagen, May 15-17, 2006, (Vincentsen, L.J.; Larsen, S.E. (Eds.)), 8 S.
[11]   Spörel, F. ; Brameshuber, W.: Monitoring the Degree of Water Saturation in Concrete Structures :
       Freiburg : AEDIFICATIO Publisher, 2007Proceedings of 5th International Essen Workshop, Duisburg-
       Essen, June 2007, S. 105-115
[12]    Elkey, W. ; Sellevold, E.J. ; Norwegian Road Research Laboratory: Electrical resistivity of concrete.
        Oslo: Norwegian Public Roads Administration, Norwegian Road Research Laboratory, 1995. - Publica-
        tion No. 80
[13]    Spörel, F.:Concerning the water content of frost-stressed concrete buildings; PhD-Thesis, Institute of
        Building Materials Research, RWTH Aachen University, 2009

To top