Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 172 (2017) 433 – 440

Modern Building Materials, Structures and Techniques, MBMST 2016

Experimental laboratory validation of reproducing road viaducts

concreting processes
Józef Jasiczak a, Agnieszka Ślosarczyka,*, Piotr Kulczewski a, Andrzej Karłowski a
a
Institute of Structural Engineering, Poznan University of Technology, 60-965 Poznan, Poland

Abstract

In the latest period, hundreds of concrete viaducts were built in Poland within a short time range. Their characteristic feature was
the use of high class concrete containing the CEM I 52.5 cement. The concreting works often took place during winter time.
Despite successful laboratory tests, the concrete prepared on the concreting site often did not achieve the assumed freeze-thaw
resistance.
The lack of freeze-thaw resistance in concrete had an unusual course. A higher than normative decrease of strength with
simultaneous high tightness was observed. There was no surface spalling, typical for the low freeze resistance concrete, no edge
curvature, etc. In addition, on the sample surface, a characteristic mesh of white leakage of unknown chemical compounds
around big aggregate grains was identified. Those compounds were thoroughly investigated with the use the SEM analysis.
It was found that the direct reason of lowering the concrete strength after 150 cycles of freezing and defrosting was the
occurrence of numerous micro-cracks formed during the hydration of high amount of clinker cement in concrete block.
Moreover, these phenomena were enhanced by the occurrence of white, needle-shaped, hard-soluble salts localised in the
boundary between aggregate and cement paste. Based on the EDS analysis, these salts were identified as nickel and calcium
phosphates, ettringite ana thaumasite formations and others, which delayed cement hydration [1,2,7,8,9].
The variety of occurring white salts forced the commencement of model laboratory tests on a specially built station for
reproducing the processes taking place during concreting and developing recommendations for the construction workmanship.

© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2016 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of MBMST 2016.
Peer-review under responsibility of the organizing committee of MBMST 2016
Keywords: concrete viaducts, frost resistance of concrete, SEM and EDS analyses, laboratory experiments.

*
Corresponding author. Tel.: +48 61-665-21-68
E-mail address: agnieszka.slosarczyk@put.poznan.pl

1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of MBMST 2016
doi:10.1016/j.proeng.2017.02.025
434 Józef Jasiczak et al. / Procedia Engineering 172 (2017) 433 – 440

1. Introduction

In the period before the 2012 UEFA European Football Championship, and later, several dozens of concrete
viaducts were built in Poland within a short time. Application of concrete class C40/50 with CEM I 52,5 cement and
building in periods of lowered temperatures was characteristic for those objects. Despite successful laboratory tests,
concrete made in the building site conditions often did not achieve the assumed F150 freeze resistance. According to
the Polish standard PN-B-06250, designation F150 indicates the concrete freeze-thaw resistance after 150 cycles of
freezing and defrosting within the range of temperature from -18 to +18°C. This designation means also that
predicting time of construction exploitation varied between 101 and 150 years. In technical literature, there was a
similar case from a few years ago, concerning the lack of freeze resistance with occurrence of salt leakage on the
border areas between aggregate [3]. Nonetheless, the reason for these phenomena has not been yet recognized.
The aim of the research was to define the reasons of the lack of frost resistance of C40/50 concrete, which was
used within 2011-2014 to build load-bearing structures of flyovers on routes: A1, A4 and S5. Flyovers’ construction
solutions were analysed, climate and technological conditions during concrete works were reproduced, parameters
of the used concrete mix were evaluated, the concrete used in the structure was analysed. Concrete compression
strength was defined on the basis of samples taken during the process of concreting, and then – when criteria for
frost resistance were not met – on the basis of core drills taken from various layers of the flyovers’ structural
elements. In order to explain the lack of frost resistance in both cubic and core samples, an analysis of concrete
microstructure was conducted with the use of method of scanning microscopy with EDS probe. The inconclusive
results of the EDS tests caused the commencement of model studies.

2. Results of completed road viaducts tests

The concreting works of load-bearing structures (girders and pavement slabs) were executed in the winter
seasons of 2011 to 2014. Concrete mixture and concrete with parameters as in Table 1 were used most commonly.

Table 1. Formula and basic technical data of bridge concrete C40/50, F150.
Laboratory formula Mixture and concrete physical and mechanical parameters
Cement 52.5 N-HSR/Na - 400 kg f cm 7 = 47.5 MPa
Water - 157 l fcm28 = 73.4 MPa
Sand - 670 kg/m3 air content 5.20 %,
Grit 2/8 - 438 kg/m3 texture measured using the concrete slump test:
12 cm, after 1 hour - 16 cm
3
Grit 8/16 - 679 kg/m volume 2349 kg/m3
Plastifier - 0.50 % per month absorbability 4.1 %
Super-plastifier - 0.80% per month water tightness degree W8, water penetration 1.8 cm
Air entraining additive - 0.08% per month frost-resistance degree F150 at strength loss of ΔR = 5.2 % and mass
of ΔG = 0.8 %

The execution of test batches and results of fresh concrete mixture as well as testing of cured concrete samples
confirmed the proper selection and proportions of the components and the positive values of all concrete parameters
specified in the design specification.
During the execution of the road viaducts, several deviations from the laboratory conditions, however, were
noted, which resulted in the lowering of the concrete mixture and concrete parameters, which in consequence
resulted in the loss of frost resistance of the structural concretes [4].
The basic deviations include:

x change in texture: instead of a slump of 12/16 cm, a slump of 18 cm or even ≥ 25 cm was noted,
x the degree of aeration of the concrete mixture at the building site fluctuated from 3.5 % to 3.9 % and was
 Józef Jasiczak et al. / Procedia Engineering 172 (2017) 433 – 440 435

lower from the test aeration by 5.2%,

x lower 28-day compressive strength of concrete by 3.7 % to 9.1% caused by concreting during the period of
lower temperatures,
x lower water absorption by weight of building site samples by 9.7 %, higher concrete tightness,
x substantially lower strength of concrete subjected to recurring freezing and defrosting by 30% to 54%,
x increase in the weight of samples subjected to freezing and defrosting by 0.95% to 1.63%,
x lack of concrete scaling on the sample surface, visible sample cracks in the entire volume, filled with white
salt residue (Fig. 1a).

The process of recurring freezing lowers the strength, but does not cause destruction of the samples and weight
loss, as is the case for non-aerated concretes, which do not comply with the concrete frost resistance criterion.
Due to the non-standard appearance of the samples, the frost resistance tests were followed up by additional non-
destructive, ultrasound tests of core samples collected from viaduct slabs, in order to compare the concrete structure
prior, during and after 150 freezing cycles. The results of designations are presented in Fig. 2a. We also noted the
increase in weight after the frost resistance tests, which is presented in Fig. 2b.

a) b)

Fig. 1. Samples from core drills: a) grid of micro-bruises in samples, filled with white residue, cut out from the viaduct’s structure, b) sample
cracks after 150 freezing and defrosting cycles.

Fig. 2. Results of recurring freezing and defrosting of samples collected from the executed road viaducts, a) ultrasound wave propagation rate,
b) change in sample weight.

When analysing the dependencies presented in Fig. 2, it may be stated that the flow rate of impulses through the
436 Józef Jasiczak et al. / Procedia Engineering 172 (2017) 433 – 440

non-frozen samples is within the range of 4500-5000 m/s (concrete strength of 50-75 MPa, respectively), which
confirms the concrete’s high technical parameters, its cohesion and low absorbability. On the other hand, the frozen
samples are characterised by a very high spread of both the impulse flow rate (from 2700 to 3300 m/s with the
strength of 35 to 65, respectively), which confirms the increasing destruction of the concrete’s internal structure and
indicates the cause of the concrete’s lower frost resistance. The phenomenon of gradual destruction of the
microstructure resulted in its evaluation prior to the recurring freezing process. The result is presented in Fig. 3.

Fig. 3. SEM and EDS image demonstrating the concrete’s micro-cracks lowering the microstructure after subsequent freezing cycles.

3. Model testing for the reproduction of concreting conditions

In 2014-2015, the laboratory of the Poznań University of Technology conducted a test aimed at the determination
of the temperature gradient in the concrete block and the testing of the block’s frost resistance. The execution of the
concrete block with the dimensions of 50 x 50 x 50 cm was carried out using a steel mould (Fig. 4a) and PVC pipe
grating (Fig. 4b) as the support structure for distributing the temperature sensors inside the block. The concrete
mixture formula was adopted as in Table 1.

a) b)

Fig. 4. Model testing station: a) steel mould for block concreting, b) PVC grating fitted with temperature sensors, inserted into the mould.

In order to reproduce the conditions occurring during the viaduct’s concreting in the winter period, the bottom
part of the concreted mould was placed in a freezer to subject the block (from the bottom) to cooling down to the
temperature of -5C. A warm air vent was connected from the top to simulate the curing of concrete (protected from
 Józef Jasiczak et al. / Procedia Engineering 172 (2017) 433 – 440 437

the wind with black film) during the day. The temperature measurements were taken automatically using the
Metronic Instruments MPI-G device, which recorded the read concrete temperatures every 30 minutes.

Fig. 5. Concreted block placed in the freezer. Fig. 6. Metronic Instruments MPI-G device for measuring
the block’s temperature.

The results of the monitoring of hydration heat temperature inside the concrete block in time are presented in Fig.
7.

Fig. 7. Results of the concrete block’s temperature monitoring.

Table 2. presents the maximum and minimum temperatures recorded for particular sensors. 3D concrete block
visualisations with measurement point locations were developed for better imaging (Fig. 8).

Table 2. Overview of the concrete block’s temperature measurements.

No. Sensor Maximum Minimum Difference

positioning temperature temperature
438 Józef Jasiczak et al. / Procedia Engineering 172 (2017) 433 – 440

1 Point 3.0 32.8 2.9 29.9

2 Point 2.0 37.3 9.3 28.0
3 Point 1.0 32.8 18.7 14.1
4 Point 3.1 28.6 1.6 27.0
5 Point 2.1 31.9 8.8 23.1
6 Point 1.1 30.2 16.6 13.6
7 Point 3.2 29.9 2.1 27.8
8 Point 2.2 32.3 8.7 23.6
9 Point 1.2 30.4 15.1 15.3
10 Point 3.3 29.9 2.3 27.6
11 Point 2.3 32.8 9.3 23.5
12 Point 1.3 30.7 14.8 15.9
13 Point 3.4 30.0 2.1 27.9
14 Point 2.4 33.0 8.8 24.2
15 Point 1.4 30.6 16.0 14.6
16 Under block -4.8 -7.9 3.1
17 Above block 27.2 24.2 3.0

Fig. 8. 3D visualisation of the concrete block’s temperatures.

The measurement data were used to determine the maximum temperature gradient for the centre and 4 edges in
the vertical direction.

Table 3. Temperature gradients depending on the sensor’s positioning.

 Józef Jasiczak et al. / Procedia Engineering 172 (2017) 433 – 440 439

Temperature for maximum gradient Δ Distance between Temperature

point gradient
(°C) (°C)
(m) (°C/m)
Maximum Minimum
Centre 31.2 19.6 11.6 0.25 46.4
Side 1 25.3 13.9 11.4 0.25 45.6
Side 2 25.1 14.4 10.7 0.25 42.8
Side 3 26.3 14.5 11.8 0.25 47.2
Side 4 25.4 14.5 10.9 0.25 43.6

The next stage of the experiment was to collect Ø100 mm core drills for the purpose of determining the
concrete’s mechanical features and their susceptibility to recurring freezing and defrosting. The compressive
strengths of three samples are as follows: 62.81 MPa; 64.40 MPa; 66.40 MPa. The change in weight and the rate of
flow of ultrasound waves through rolled samples prior to, during and after 150 freezing cycles were also subject to
monitoring. The results are presented in Fig. 9. The concretes obtained the frost resistance of F150.

Fig. 9. Results of recurring freezing and defrosting of samples collected from the concrete block executed in the laboratory of the Poznań
University of Technology, a) ultrasound wave propagation rate, b) change in sample weight.

4. Conclusions

4.1. After analysing over 20 cases of road viaduct concreting, it seems very probable that the cement matrix in
the bridge structure blocks, concreted in the winter period using a large amount of high calorific cements in the
concrete’s unit of volume, will feature micro-bruises. A high temperature gradient features micro-bruises of the
cement matrix, which during latter hydration and hydrolysis process are filled mainly with limestone compounds,
which creates an apparent continuous structure of the concrete. During the freezing and defrosting cycles occurs the
activation and watering of concretions, which occur in the form of white leaks. The concrete’s porosity thus
increases, and its compressive strength decreases, which is confirmed by the ultrasound wave flow rate, measured in
both facilities. If the concrete is sufficiently aerated, the freezing water does not cause additional destruction, which
is why the samples do not lose their weight. Furthermore, in the climatic factor conditions of the IV and I quarter of
the 1990’s and with the use of the CEM I 52.5 cement in the amount of 400 kg/m3 of mixture and the water-cement
ratio of = 0.39, it was necessary to apply active water curing of the concrete surface, or even its cooling, which is
recommended by W. Kiernożycki in his monograph [5].
4.2. The tested measurement set allows a wide range of numerous experiments with purposefully changed
parameters. It is possible to prepare variants of the quality and quantity of cements, water-cement ratio, chemical
additives for concrete, mixture aeration degree and climatic conditions or curing methods. The parameters can be
440 Józef Jasiczak et al. / Procedia Engineering 172 (2017) 433 – 440

changes numerous times until determining the proper formula of the concrete mixture and specifying the climatic
conditions ensuring the concrete properties necessary for the structure’s safety. The laboratory experiment is a cheap
process, however if prepared properly, it can protect the investor against severe financial consequences. The global
literature describes such cases, e.g. the widely known article [6] of Martin R-P., Toutlemonde F.: Theoretical and
experimental validation of a simple method to reproduce representative DEF-prone conditions in laboratory,
Materials and Structures (2013) 46, 1245 – 1255.

Acknowledgements

The financial help of the Polish Ministry of National Education (Project No. 11- 861/2016 DS) is gratefully
acknowledged.

References

[1] J. Małolepszy, E. Tkaczewska, Effect of fly ashes from co-burning bituminous coal and biomass on hydration and properties of cement,
Conference Proceedings Dni betonu, Polski Cement, Wisła, 2006.
[2] E. Tkaczewska, E. Kłosek-Wawrzyn, Effect of phosphate ions PO42- on cement hydration, Cement Wapno Beton 2 (2008) 401-408.
[3] W. Wołowicki, A case of unreliability of frost-resistant concrete test, Conference Proceedings Dni betonu, Polski Cement, Wisła, 2008.
[4] J. Jasiczak, A. Ślosarczyk, Cases of concretes freeze resistance lack in new viaducts, CCC’13: Concrete Structures in Urban Area, Wroclaw
2013, pp. 469-473.
[5] W. Kiernożycki, Betonowe konstrukcje masywne, Polski Cement, Kraków, 2003, p. 320.
[6] R-P. Martin, F. Toutlemonde, Theoretical and experimental validation of a simple method to reproduce representative DEF-prone conditions
in laboratory, Materials and Structures 46 (2013) 1245 – 1255.
[7] L. Divet, A. Pavoine, Delayed Ettringite Formation in massive concrete structures: an account of some studies of degraded bridges,
International RILEM TC 186-ISA workshop on Internal Sulfate Attack and Delayed Ettringite Formation, Villars, Switzerland, 2002, pp. 98-
126.
[8] Y. Shao, C.J. Lynsdale, C.D. Lawrence, J.H. Sharp, Deterioration of heat-cured mortars due to the combined effect of delayed ettringite
formation and freeze/thaw cycles, Cement and Concrete Research 27(11) (1997) 1761-1771.
[9] H.F.W. Taylor, C. Famy, K.L. Scrivener, Delayed ettringite formation, Cement and Concrete Research 31 (2001) 683-693.