Nuclear containment buildings are designed to withstand internal pressures and to act as a barrier to the release of radioactive products in the event of an accident. The walls are typically 1 - 1.5 m nominal thickness pre-stressed monolithic structures and may have a steel liner cast in some 300 mm from the inner surface. These structures are often cast using the slip-form technique, which is briefly described below.
Construction using the slip form technique
The process of slip-forming involves using a form which moves continually upwards at a rate of up to 300 mm per hour while fresh concrete is poured in 150 mm layers into the top. As shown in Fig. I, the forms are inclined inwards some 6 mm/m to allow the concrete to detach itself from the bottom of the form. Plastic concrete placed into the top of the form should be stable by the time the formwork leaves it below (a penetration resistance of between 50 and 200 psi measured in accordance with ASTM C 403 is normally required at the trailing edge of the formwork) and should therefore not sag.
Fig. 1. Slip form construction may cause problematic zones in concrete.
|
In Fig.1 it can be seen that the thickness of the concrete wall increases successively from the top of the form until the side support ends. This increase in thickness is partly hindered by the reinforcement near the surfaces. This means that the concrete outside the reinforcement can move more readily against the form than the inner concrete. The risk of cracks forming is thus dependent on the consistency of the concrete, Fig.1(a). This type of cracking is hazardous in open-air environments where aggressive elements such as Cl, CO2, SO2 and rainwater can penetrate to the reinforcement.
Another form of cracking is that caused by separation of the upper section of the plastic concrete in the form. A large horizontal separation crack develops and is masked by cement paste which is smeared by the form at the surface, Fig. 1 (b).
Fig. 2. A slip-formed concrete chimney after removal of the cover concrete by hydro-demolition. The large horizontal void had existed since the time of construction
|
In Fig.2 we see the result of concrete "hang-up" in the shell wall of a concrete chimney. This particular defect went unnoticed for many years after construction. The design of the concrete mix in this case was probably unsuitable and this combined with poor compaction led to the formation of a large internal void. Internal voids adjacent to the steel liner of an NPP deny the corrosion protection
normally provided by cement paste and this may have serious consequences in the warm and humid environment of a power station.
Structures dating from 1940, 1970 and 1985 have been inspected by the authors and have shown similar patterns of damage as described above.
Pre-stressed cables and their protection
Tensile stresses in the containment walls and roof of a NPP are intended to be restrained by pre-stressed
cables. These are high-tensile steel cables or wires which are placed in ducts of approximately 70 mm
diameter. After tensioning the cable ducts are injected with a cement grout to provide corrosion
protection. In some cases the ducts are filled with grease or oil. After injection with cement grout it is
not possible to remove the cables for inspection.
Generally speaking pre-stressed cables may be prone to one or more of the damage types listed below:
- corrosion of the steel cables due to external effects (intrusion of aggressive elements)
- initial damage caused by heat-treatment process
- handling damage during transport and construction
- weather conditions during construction
- injection process
- injection material and its chemical composition
- hydrogen embrittlement and stress corrosion
- leakage currents ( hydrogen formation at the cathode or anodic corrosion effects )
In the most extreme cases damage to pre-stressed cables may lead to failure and collapse, as occurring in the case of bridges and industrial buildings in Germany and the UK.
One of the primary areas with respect to corrosion attack of the stressed cables is a void in the cable duct since the steel is not protected by the alkaline cement environment and is exposed to intrusion of aggressive elements. The testing situation is thus one of first locating the cable ducts m the concrete members (preferably by non-destructive means) and then detecting voids in the grout filling of the duct. If the physical condition of the steel cables can be detected using NDT then this is of course a bonus.
Concrete moisture content and differential shrinkage
Fig.3. Moisture distribution in a concrete element with one surface exposed to a dry (room) environment whilst the opposite surface was occasionally wetted by condensed water.
|
Water is one of the major components in concrete, it reacts with cement and makes the fresh concrete mixture workable. Only part of the water will be consumed by the cement hydration, and the rest will be absorbed in the capillaries in concrete, ready to evaporate when the concrete is exposed to a dry ambient condition. Drying of the concrete causes moisture to evaporate from the capillaries, and the cement paste shrinks to compensate for the surface energy change. As the drying is taking place through the concrete surfaces, it will create an uneven moisture distribution from the surface Awards and consequently a differential shrinkage for the concrete member. This may lead to tensile stresses with resulting crack formation.
An example of a thick concrete element with one-sided water pressure is shown in Fig.3. The humidity profile in this case is shown for depths 150, 300 and 800 mm from the "dry" surface. The age of the structure at the time of test was approximately 25 years. Visible cracks at the surface were found to extend to approximately 300 mm depth.
Chloride induced reinforcement corrosion - Case histories
The presence of chloride (Cl) in reinforced concrete has been recognised, since the 1950s, as one of the major reasons for reinforcing steel corrosion in reinforced concrete, such as maritime structures, highway bridges and parking decks in areas where Cl-containing de-icing salts are used during the winter season. Corrosion has been attributed to the fact that cl destroys the protective passivation film which is formed on steel surfaces due to the high pH condition in concrete pore solution (pH=12-13).
In recent decades, research on concrete durability has greatly increased and has demonstrated that many of severe problems, such as cracking, delimitation and spalling of concrete, are related to cl induced steel corrosion. The corrosion induced damage is a serious problem for the infrastructure and commonly found in structural members such as pillars and beams. If sea-water is used as the coolant in power stations, it can of course affect the concrete cooling water system and could lead to stoppages in production.
Fig.4. Cl distribution in concrete.
Structure 1: Quay at a seaside, 23 years.
Structure 2: Water channel (for sea-water), 18 years. Cl- content was determined by chemical analysis of specimens sampled from the structures.
|
From the point of view of providing sufficient information to the owners of concrete structures to enable them to plan the best strategy for production, including repair and maintenance work, it is important to inspect the Cl distribution in concrete. Fig.4 demonstrates the variation of total Cl in concrete, i.e. the profile of Cl content in relation to depth from the surface which had been exposed to sea-water. The ingress of Cl depends on both the concentration of Cl in the seawater and the quality of
the concrete. This is evidenced by the samples taken from Structure 2, which despite the shorter service time and less saline seawater, the depth of the affected zone is found to be greater compared with Structure 1.
The surface layer of a few millimetres thickness contains less Cl . This is more obvious for the lower quality concrete (Structure 2) and for the concrete at semidry conditions, and can be mainly attributed to the deterioration of the concrete. The surface concrete in the seawater splash zone was carbonated by the CO2 m the surrounding air. It was washed by seawater and rain, which has caused decomposition of the cement paste due to leaching-out of lime, and was subjected to other physical and chemical deterioration processes such as sulphate and magnesium attacks.
The maximum Cl content is about 0.2 % by mass of concrete, occurring at about 5 mm under the surface for Structure 1. The maximum Cl content in Structure 2 is about 0.25% which seems to be quite high, considering the fact that the Cl content in the seawater in this area is only 1.8 g/l (or 0.05 mole/l). This can be explained by its lower designed strength (higher water/cement-ratio ) which corresponds to a higher porosity and water permeability. Experimental results also showed that about 70% of the total Cl in concrete could be extracted by water, which indicates that a large part of the Cl ions are only loosely adsorbed by the concrete and may participate in the steel corrosion process.
Indeed the in-situ measurement of electrochemical potential of Structure 2 showed active areas where the reinforcement may have started to corrode. According to Sagoe-Crentsil and Glasser [2], in a dilute Cl solution of 0.010-0.015 mole/l, the solubility of iron at pH appr. 13 abruptly increased from nearly zero to 0.170-180 mole/l, i.e. the corrosion is likely to occur even at the presence of such low concentration of cl. A higher threshold value proposed earlier by Hausmann [3] was mole ratio Cl/OH > 0.6 (Cl=0.06 mole/l in concrete with pH=13), which can be expected to prompt a rapid corrosion process.
Alkali-aggregate reactions
This is a form of chemical reaction which occurs between aggregates which contain alkali-reactive constituents and cement alkalis in the pore water of the concrete. The most common case is alkali-silica reaction (ASR), which is a reaction between alkali and some siliceous compounds in aggregates producing a type of gel. When in contact with water, the gels swell causing tensile stresses and ultimately cracking, which often results in a "map pattern" on the concrete surface. Of the Scandinavian countries ASR damage has so far only been a serious problem in Denmark.
The detrimental expansion takes several years to develop in field concrete structures, thus the potential risk is often evaluated in the laboratory under accelerated conditions. For example, some Norwegian rock types have been considered to be non-reactive only to be re-evaluated on the basis of experiment. In these studies it was found that for reactions to occur then at least 20 % of the total
aggregate content should be reactive, and that the alkali content of the concrete should be high i.e. more than 3 kg equivalent Na2O/m3 concrete [4]. A pre-requisite for AAR is of course high moisture levels. When a structure is suspected to have AAR, both the reactivity of the aggregates and humidity levels in concrete need to be examined, and the development of cracking closely monitored.
It should be noted that the development of cracking may accelerate after some years. The consequences, apart from reduced durability due to cracking, are primarily reduced tensile strength.
Non-destructive testing methods have been drawing more and more attention, in the sense of reliability and effectiveness, to the inspection of concrete structures. The importance of being able to test in-situ is being recognised and this means an increasing trend away from the traditional random sampling of concrete for material analysis to the use of sophisticated non-destructive techniques on the structure. There are generally speaking two families of testing techniques which are being used for large concrete structures:
- Quick and effective, non-destructive methods covering large areas and volumes
- Methods which may be used to characterize deviations or suspected defects
The methods may be direct or indirect, for example they may be used to give information about size, depth or physical condition or they may from one or more measured parameters give information which may be used to evaluate risk of damage or on-going deterioration processes. They may also give information relating to strength of material, potential durability or bearing capacity. There are of course advantages in having access to several techniques which complement each other.
In the following tables a summary is given of potential damage types and suitable non-destructive testing techniques for assessing these, including their present capability. The testing methods listed are exclusively those which the authors have first-hand experience of and represent state-of-the-art techniques for in-situ testing of concrete. The summary evaluation of the methods in terms of reliability and usefulness is based on practical experience in specific situations and it is not intended that these evaluations be universally applied to the given techniques.
|
Table 1. Non-destructive Testing of Concrete - Point of Interest / Damage Type
| | Item / Object of test | Description
|
| 1 | Cracking in surface concrete
|
| 2 | Cracking internally (concrete)
|
| 2 (a) | Deterioration of concrete with time (ex. development of cracking)
|
| 3 | Damaged concrete layers (reduced strength and elastic properties)
|
| 4 | Voids in concrete and homogeneity
|
| 5 | Elastic properties of concrete (E, G - moduli)
|
| 6 | Thickness of concrete member or layer
|
| 7 | Location of reinforcement
|
| 8 | Size of reinforcement
|
| 9 | Location of pre-stressed cable ducts
|
| 10 | Detection of voids in pre-stressed cable ducts
|
| 11 | Condition of pre-stressed cables (corrosion damage or fracture)
|
| 12 | Detection of actively corroding reinforcement (electrochemical)
|
| 13 | Estimation of reinforcement corrosion rate (electrochemical)
|
| 14 | Concrete resistivity (related to corrosion rate)
|
| 15 | Dielectric and conductive properties of concrete |
Table 2. Non-destructive Testing of Concrete - Testing Methods
* Rating for reliability and usefulness:
1 - Good, 2 - Fair. 3 - Poor.
| | Test Method | Principle | Application
(cf. Table 1) | Capability / Limitations | Reliability/
Usefulness *
|
| UPV | Ultrasonic wave
transmission -
measurement of
wave speed and
frequency-dependent
attenuation | 1, 2, 3, 4, 5
2 (a)
| Unknown factors may affect wave
speed. Interpretation demanding.
Quantification of information
requires complementary technique
| 2 / 2
Unknown
Ultrasonic
Pulse Echo | Transmission and
reflection | 1
2
4
6
9
10 | crack surfaces open
cracks parallel to surface/crack size
16 mm aggregate; depth ~ 700 mm
16 mm aggr.; depth ~ 1300 mm
16 mm aggr.
16 mm aggr. Void or lack of bond ? | 2/2
1 or 2 / 2
1 / 1
1 / 2
2 / 2
2 / 2 or 3
| SASW | Spectral Analysis of
Surface Waves | 1
2
3
4
5
6 |
min. 50 mm depth
large voids
access to surface dimension | unknown
unknown
1 / 1
unknown
unknown
1 / 1
| Impact Echo | transmission and
reflection of
transient stress
waves | 2
4
6
10
| cracks parallel or near // to surface
void size = measuring depth
resolution & wave speed det. acc
cannot distinguish lack-of-bond
from void | 1 / 1
1 / 2
1 / 1
2 / 3
| G.P.R (Radar) | Transmission and
reflection of
electromagnetic waves | 3
4
6
7
8
9
15
| larger stratification
large voids
moisture restrictive
approx. 700 mm depth
special processing
approx. 700 mm depth (1 GHz)
(large amounts of steel have a
screening effect) | 2 / 2
2 / 2
2 / 2
1 / 1
unknown
1 / 1
2 / 2
| H.E.R | High energy
radiography
(X-ray) | 4
6
7
8
9
10
11 | thickness of member 1 m
thickness of member 1 m
thickness / access
thickness / access
thickness / access
approx. 800 mm thickness / access
800 - 1000 mm thickness / access | 1 / 2
2 / 2
1 / 1
1 / 2
1 / 1
1 / 1
1 / 1
| Percometer | Measurement of
conductivity and
dielectric constant | 14
15
| surface measurement
surface measurement
| unknown / 1
unknown / 1
Half-Cell
Potential | Measurement of
electrochemical
potential of steel in
concrete | 12
| reinforcement nearest measuring
surface; not in water saturated
concrete
| 1 or 2 / 1
| G.P.M | Polarisation
properties of
reinforcement | 12
13 | can be used in water saturated
concrete (slower)
applies to general corrosion | 1 / 1
unknown / 2
| | | | | | | | |
It may be pointed out that due to the special features of concrete construction, e.g. large dimensions, micro-heterogeneity of materials, etc., even the very conventional NDT testing methods need to be and have been modified and the mechanisms researched into in order to give efficient and confident analysis.
To assess reinforcement corrosion, half-cell potential has been used to determine the corrosion activity of steel in concrete [5]. It has been found that the potential criteria for corrosion likelihood are not applicable to all structures and in particular different environments, and so a potential contour map which emphasises potential gradients is commonly used to indicate active areas of corrosion on the steel. In recent years, a galvanostatic pulse measurement (GPM) technique has been developed and used in conjunction with the half-cell potential, which produces an enhanced potential map and enables the estimation of the corrosi
on rate [6].
Wiberg [7] has elaborated the dependence of pulse propagation through concrete on the material composition and properties, and shown and discussed the possible applications of ultrasonic techniques in detecting defects, even micro-cracks, in concrete.
The use of ultrasonic impulse echo techniques in testing concrete has been intensively studied since the 1980s. The theoretical background, results of experimental work and some field inspections are presented by many researchers, e.g. Krause et al [8]. Sansalone and Carino [9] demonstrated a generally successful use of impact echo method in detecting large flaws in concrete, and positive results in testing duct (without strands) embedded in concrete slabs [10]. On the other hand, Jansohn et al [11] reported uncertainties of using this method m detecting voids in metal ducts containing tendons. Moreover, reports on successful field inspection in detecting voids in tendon ducts have yet not been apparent.
X-ray radiography, a technique that visually reveals object under investigation, has been used for insitu concrete inspection since the 1950s. However, its field application continued only at a very limited extent, as was summarised by Mitchell [12], partly due to the cost and immobility of X-ray equipment which requires high voltages.
In recent years, more and more investigations have been carried out to explore the use of radiographic imaging techniques for concrete property control. For example, Clarke [13] presented the use of g-radiography m concrete, and the dependence of transmission of g-rays on concrete thickness, and the thickness of lead filter. BS 1881: Part 205 recommended 60 Cobalt, 192 Iridium, and Linac (8 MeV X-rays) as suitable sources of radiation. Kear and Leeming [14] demonstrated the possibility of using radiographic methods to inspect post-tensioned concrete structures, and especially the advantage of high energy X-ray source such as Betatron, m penetrating power, image sharpness and safety. The authors [15] employed a 6 MeV Betatron in a field inspection of a concrete structure, and showed that it effectively revealed the pre-stressing strands, concrete defects as well as the reinforcement in a 750 mm thick concrete slab.
