University of Ljubljana Faculty of mathematics and physics

Department of physics

Primož Vavpetič

SEMINAR

Corrosion in concrete steel

ADVISOR: prof. dr. Žiga Šmit

Kamnik, april 2008

2

Abstract Reinforced concrete uses steel to provide the tensile properties that are needed in structural concrete. It prevents the failure of concrete structures which are subjected to tensile and flexural stresses due to traffic, winds, dead loads, and thermal cycling. However, when reinforcement corrodes, the formation of rust leads to a loss of bond between the steel and the concrete and subsequently delamination and spalling. If left unchecked, the integrity of the structure can be affected. Reduction in the cross sectional area of steel reduces its strength capacity. This is especially detrimental to the performance of tensioned strands in pre-stressed concrete.

3

Contents Introduction ...........................................................................................................4 Corrosion in concrete ............................................................................................5 The corrosion mechanism .......................................................................................5 Carbonation..........................................................................................................6 Chloride attack......................................................................................................6 Stress corrosion cracking........................................................................................7 Theoretical background for corrosion...................................................................8 Consequences of steel corrosion.........................................................................10 Corrosion: structural effects................................................................................11 Environmental influence in the corrosion process ..............................................11 Measurements of corrosion in concrete using electrochemical impedance spectroscopy (EIS)..............................................................................................12 Preventing corrosion ...........................................................................................14 Galvanisation .....................................................................................................14 Cathodic protection ............................................................................................14 Electrochemical chloride migration (Desalination) ...................................................15 Re-alkalisation ...................................................................................................15 Corrosion inhibitor repair techniques .....................................................................15 Conclusion ..........................................................................................................15 References ..........................................................................................................16

4

Introduction Standard terminology defines corrosion as “the chemical or electrochemical reaction between a material, usually a metal, and its environment that produces a deterioration of the material and its properties.” For steel embedded in concrete, corrosion results in the formation of rust which has two to four times the volume of the original steel and none of the good mechanical properties. Corrosion also produces pits or holes in the surface of reinforcing steel, reducing strength capacity as a result of the reduced cross-sectional area. Reinforced concrete uses steel to provide the tensile properties that are needed in structural concrete. It prevents the failure of concrete structures which are subjected to tensile and flexural stresses due to traffic, winds, dead loads, and thermal cycling. However, when reinforcement corrodes, the formation of rust leads to a loss of bond between the steel and the concrete and subsequently delamination and spalling. If left unchecked, the integrity of the structure can be affected. Reduction in the cross sectional area of steel reduces its strength capacity. This is especially detrimental to the performance of tensioned strands in pre-stressed concrete. Steel in concrete is usually in a non-corroding, passive condition. However, steel reinforced concrete is often used in severe environments where sea water or deicing salts are present. When chloride moves into the concrete, it disrupts the passive layer protecting the steel, causing it to rust and pit. Carbonation of concrete is another cause of steel corrosion. When concrete carbonates to the level of the steel rebar the normally alkaline environment, which protects steel from corrosion, is replaced by a more neutral environment. Under these conditions the steel is not passive and rapid corrosion begins. The rate of corrosion due to carbonated concrete cover is slower than chloride-induced corrosion.

Figure 1: Pictures of various types of corrosion damage namely localized corrosion (middle)

and general corrosion due to carbonation (left, right) when spalling occured.

5

Corrosion in concrete

The mechanism of corrosion in aqueous media is of electrochemical nature. This means that the oxidation of the metal is counterbalanced by the reduction of another substance in another region of the metallic surface. Therefore, zones (anodes and cathodes) with different electrochemical potential, develop. In the case of concrete the electrolyte is constituted by the pore solution, which is very alkaline. This pore solution is formed by mainly a mixture of KOH and NaOH presenting pH values ranging between 12,6-14. The solution is saturated in Ca(OH)2. Steel embedded in concrete is naturally protected by this high alkalinity and by the barrier effect of the cover itself. The two main causes of electrochemical corrosion are carbonation and the presence of chlorides (Figure 2). Carbonation usually induces a generalized corrosion while chloride will lead into pitting or localized attack. The corrosion can be easily recognized by the rust presence on the rebar and by the appearance of cracks running parallel to the rebars. In figure 1 is also identified another particular type of corrosion, the stress corrosion cracking (SCC), that develops in prestressed wires subjected to special aggressive conditions.

Figure 2: Types and morphology of the corrosion in concrete: generalized (carbonation), localized (chlorides) and stress corrosion cracking (in prestressed wires).

The corrosion mechanism Corrosion of steel reinforcement occurs by an electrochemical process which involves exchanges of electrons similar to that which occurs in a battery. The important part of the mechanism is the separation of negatively charged areas of metal or 'anodes' where corrosion occurs and positively charged areas or 'cathodes' where a harmless charge balancing reaction occurs (Figure 3). At the anode the iron dissolves and then reacts to form the solid corrosion product, rust. The rust is formed at the metal/oxide interface, forcing previously formed oxide away from the steel and compressing the concrete, causing it to spall.

6

Figure 3: Exposed steel will corrode in moist atmospheres due to differences in the electrical potential on the steel surface forming anodic and cathodic sites. The metal oxidises at the anode where corrosion occurs according to:

Fe (metal) --> Fe2+ (aq.) + 2e-

Simultaneously, reduction occurs at cathodic sites, typical cathodic processes being:

½O2 + H20 + 2e--(metal) → 2OH-(aq.)

2H+(aq.) + 2e- (metal) → H2(gas)

The electrons produced during this process are conducted through the metal whilst the ions formed are transported via the pore water which acts as the electrolyte. Carbonation Atmospheric carbon dioxide reacts with the calcium and alkaline hydroxides and cement phases, leading in a lowering of the pore solution pH value until values near neutrality. This process aims into the depassivation of the steel in contact with the carbonated zones. Carbonation is a diffusion process and therefore, its depth progresses by an exponential attenuation along the time. The modelling of carbonation is generally made by means of the

simplified expression: x = k(CO2) t , where x is the carbonation depth, t is the time and k(CO2) is the carbonation factor of the particular concrete. It does not develop if the concrete is water saturated or in very dry conditions. However, as cycling wet-dry periods are the usual environmental outdoor conditions, the carbonation front can advance relatively fast. As the corrosion is generalized, cracks will appear running parallel to the rebars. Usually they appear not before 20 years life for a cover of 20-25 mm, what means that the corrosion rates are in general low. Spalling will be produced at later stages. Chloride attack The chloride ions may be present in the concrete if they are added in the mix (admixtures, water or aggregates). However, this is fortunately not common. The most frequent is that chlorides penetrate from outside, either due to the structure is placed in marine environments or because deicing salts are used. Chlorides induce local disruption of steel passive layer dealing into pits or localized attack. In submerged zones or in fully saturated concrete, chlorides penetrate by diffusion. However, in aerial zones or when submitted to cycles (deicing salts), capillary absorption may be a faster mechanism of penetration. In both cases, the penetration is as well dependent of the square root of time. Therefore, its modelling may be made similarly to the carbonation, by

means of the simplified expression x = k(Cl) t . The chloride ingress is usually modelled by means of the so called error function equation (see Theoretical background for corrosion

section) which is a particular solution of Fick’s second law: Cx= Cs[1 – erf( 2x Dt )] where

Cx is the chloride concentration at the depth x, Cs is the surface chloride concentration and D the chloride diffusion coefficient.

7

Depending on how extended or localized is the corrosion, cracks may appear or not. In submerged zones sometimes the rebar corrodes without any external sign of cover cracking. Concerning the amount of chlorides needed to induce the onset of corrosion (threshold value), it depends on several factors not fully quantified. This multiple dependence makes difficult to fix a single value. The factors influencing the chloride threshold are:

- Type of cement: finess, amount of gypsum, blending materials etc. - Water/cement ratio – w/c (porosity). - Curing and compaction (porosity). - Moisture content and variation. - Type of steel and surface roughness and condition (pre-rusted or not). - Oxygen availability (corrosion potential when arriving the chlorides).

Each structure has its threshold which, in an already corroding structure can be verified by appropiate testing. To predict the threshold in a particular structure is difficult, but if the steel has depassivated in some areas, the testing of the concrete surrounding the rebar will enable a more precise knowledge. The threshold may be given as Cl-/OH- of the pore solution of % of Cl- by weight. In spite of this difficulty of fixing a reliable and general chloride threshold, all codes limit the chloride content in the mixing water. In absence of other value, this amount (in general 0.4% of cement weight) can be taken as reference. Stress Corrosion Cracking The SCC is a specialized type of corrosion which is produced when mechanical stresses act simultaneously to some specific aggressive agents. This type of corrosion may then develop in prestressed or postensioned wires. The mechanism of this type of corrosion is not yet well understood and several theories exist in the literature. The phenomenon may occur accompanied by an embritlement of the steel due to the penetration in the steel of hydrogen gas produced by a corrosion reaction. The three conditions necessary to develop the phenomenon are:

1) a type of steel susceptible to suffer this type of corrosion, 2) the steel has to be stressed beyond a minimum threshold below which the process

is very slow, and 3) a specific aggressive media (producing or not hydrogen gas)

When the three conditions are found simultaneously the process develops in three steps: 1) one or several microcracks are generated at the surface of the steel, 2) these cracks grow until they reach a certain depth and then they propagate very

quickly until 3) it aims into the brittle failure of the wire. This failure may be enhanced by hydrogen

embritlement. The phenomenon may then be slow during the generation of the crack and later it propagates very quickly leading to a sudden unexpected failure (Figure 4). The corrosion during the first step cannot be noticed by means of measuring the corrosion rate, as the loss of metal is negligible and the corrosion potential measured at the concrete surface may not indicate the developing of the process. Therefore, this type of corrosion cannot be electrochemically measured during its occurrence. Only the risk of its appearance may be approached by the traditional electrochemical techniques. The identification of the nature of the failure is not an easy task after being produced. It has to be detected by microscopical observation of the fractured surface when it is fresh, and not

8

corroded or contaminated. In the case of having occurred SCC or H2 embritlement, the fractography enables the identification by the brittle aspect of the steel fractured surface.

Figure 4: Start of stress corrosion cracking. This type of phenomenon has not to be mistaken with the failure induced by chlorides in prestressed wires. The ions inducing localized corrosion may also aim into a failure, but of simple reduction (localized attack) of the cross section produced by the electrochemical process. The SCC may not be of electrochemical nature and then its rates of propagation differ from those of normal localized corrosion. Typical contaminants promoting SCC are sulphides, sulphates or tiocyanides, while chlorides are less linked to this type of failure.

Theoretical background for corrosion The corrosion rate is controlled by many factors such as total chloride ion content of the pore solution, pH level, availability of oxygen, water content, temperature etc. In corrosion decay of steel in concrete several processes may be combined, making it difficult to identify a single mechanism. One of the mechanisms for surface penetration is intrusion of chloride-bearing water into capillary pores of unsaturated (dry) concrete by capillary action. Alternate wetting and drying can lead to buildup of chloride ions through absorption. If a structure is not dried to a high degree for prolonged period of time, chloride penetration into concrete by absorption and capillary suction is basically restricted to a small depth below the surface. If there is a differential head of chloride bearing water, permeability will also influence the ingress of chlorides for which higher permeability coefficient will permit higher rate of flow. The other dominant mechanism of the chloride ion transport is the diffusion which takes place under a concentration gradient. If outside concentration is higher than the inside of concrete, the migration of chloride ions through pore water in concrete will take place by diffusion. The relative importance of the two major mechanisms of chloride transport, namely diffusion and absorption, depend on the moisture content of concrete. Absorption may be dominant if a dry concrete with significant loss of pore water is wetted with chloride-bearing water, whereas for a reasonably moist concrete (sufficient level of pore water exists) diffusion process will prevail. However, researchers tend to agree that in most cases diffusion can be assumed to be the basic transport mechanism of chloride ions for reasonably moist structures. As part of the chloride in concrete is chemically bound due to reaction of chloride ions with constituents of cement, the free chloride concentration is of importance for corrosion initiation.

9

The chloride penetration can be modeled by Fick’s diffusion law without the reaction component when the pores are wide enough, so that the solid bodies around the pores (traps) may be neglected. The diffusion through the pores takes place only through the tortuous pores of concrete. As the pores are not straight, the diffusion effectively takes place over longer distance than it would be in homogenious medium. Also, the solids being impermeable, diffusion occurs over a smaller cross sectional area than that available in a homogenious material. The effects of the longer diffusion path and smaller areas can be lumped together in the definition of the effective diffusion coefficient.

De = τ

ε D (1)

where - ε is the total porosity of the material - τ is the tortuosity factor, which accounts for the sinuosity of the pores along the path

of diffusion - D is the diffusion coefficient of NaCl in water 1,26 ·105 cm2/s at the specific

temperature of 19 ○C. Ionic diffusion in a porous media is governed by Fick’s first law, which deals with the mass flux due to a concentration gradient,

x

CDJ e

∂

∂−= (2)

where - J is the mass flux - De is the effective diffusion coefficient - C is the concentration - x is the distance

-x

C

∂

∂is the concentration gradient in one dimension.

For unsteady state in one dimensional diffusion, Fick’s second law is

x

J

t

C

∂

∂−=

∂

∂. (3)

Substituting for J from (2)

2

2

x

CD

t

Ce

∂

∂=

∂

∂ (4)

where De has been assumed to be constant.

10

The applicable boundary conditions for the solution of this differential equations for semi-infinite domain are: Cx = Ci, at t = 0 when 0 < x < ∞ Cx = Ci, at x = 0 when 0 < t < ∞ Cx = Ci, at x = ∞ when 0 < t < ∞ By combination of variables, the solution for Eq.(4) is

tD

xerf

CC

CC

eis

ix

21−=

−

−

where - Ci is the initial chloride concentration - Cx is the chloride concentration at a depth x - Cs is the surface chloride concentration - erf is error function - De is the effective diffusion coefficient - t is the time elapsed Using the above solution, when the initial chloride concentration Ci, the surface chloride concentration Cs and the effective diffusion coefficient De at a time t are known the chloride concentration Cx at a particular depth x can be determined.

Consequences of steel corrosion In the case of reinforcement corrosion, the most simple and descriptive model for service life is shown in Figure 5. This well known model considers:

- An initiation period which consists of the time from the erection of the structure until the aggressive agent (either chlorides or the carbonation front) reaches the rebar and depassivates the steel.

- A propagation period from the steel depassivation until a certain unacceptable level of deterioration is developed in the structure.

Figure 5: Service life model for reinforcement corrosion (T is temperature, RH is relative humidity).

11

Corrosion: structural effects In the case of non prestressed reinforcements, the first direct effect of the reinforcement corrosion is its section decrease due to the corroding process. Iron oxides (rust) resulting from the corrosion process have a larger volume than the original steel, in the case of reinforced concrete structures, and this effect induces internal stresses in the concrete which may lead to cracking or even spalling of concrete cover. Corrosion also may reduce the steel elongation at maximum load, affecting subsequently the structure ductility. On the other hand, the composite action of concrete and steel in a reinforced concrete structure is based on the bond between them, and this is also affected by corrosion through several mechanisms: a) increasing of hoop stresses due to pressure of rust, producing concrete cracking, b) change of properties of the interface concrete-steel, and c) the corrosion of stirrups. Accordingly, reduction of structural capacity of reinforced concrete elements affected by rebar corrosion is mainly due to the following three main phenomena, which are direct consequence of corrosion:

- reduction of rebar section due to corrosion - reduction of bond strength - loss of concrete integrity due to cover cracking and/or spalling

The rate of developing of these phenomena is function of different parameters as corrosion current (Ionic current in Figure 3), type of aggressive, time since propagation period was initiated, and reinforcement or structural detailing. In the case of SCC, the main difference lies in the fact that rust may not be generated and therefore the cracking of the cover is not produced. The main concrete characteristics affected are the ductility of the bars and their reduction in cross sectional area. When partial failure is produced, the reduction in structural ductility and the collapse of the main other parts, can be immediately achieved.

Environmental influence in the corrosion process The reinforcement corrosion is the consequence of the ageing of the concrete in a particular environment. Then, the environmental actions are the aggressive which may shorten the concrete durability. The main influencing parameter on the corrosion among the environmental conditions is the moisture content of the concrete, which is dependent on the external temperature and humidity. The concrete moisture controls the penetration of gases such as CO2 and O2. Thus, when the material is water saturated the carbonation is delayed and the access of O2 is limited. On the opposite, chlorides penetrate quicker if the material is saturated. Regarding the corrosion process, the concrete moisture content will influence the electrical resistivity (porosity). All these effects are not uniform in the concrete mass, but a gradient of moisture is produced from the concrete surface towards the interior. It will depend on the rebar position (cover depth) how the climatic cycling will influence the corrosion process. Temperature will influence as well the corrosion process. It produces also two opposite effects: acceleration or retardation of the reaction. When temperature rises, evaporation of pore water is induced and oxygen is removed from the pore solution. Therefore, although the corrosion process is stimulated by the rise in temperature, this may be counterbalanced by the increase in resistivity (evaporation) and the removal of oxygen (smaller solubility at higher temperatures). An opposite effect is induced by a lowering of temperature in semi-dry concretes as condensation is induced.

12

0,0E+00

1,0E+03

2,0E+03

3,0E+03

4,0E+03

5,0E+03

6,0E+03

7,0E+03

0,0E+00 2,0E+03 4,0E+03 6,0E+03 8,0E+03 1,0E+04 1,2E+04 1,4E+04

Real (Ohm)

-Im

ag

(O

hm

)

All this means that the effect of daily and seasonal variations of RH (relative humidity) and T (temperature) on the corrosion of rebars, cannot be directly deduced and quantified. The experimental evidence shows that the humidity content of the concrete varies depending upon the external RH and the direct exposition to rain. Raining events dramatically affect the saturation degree of the concrete. The saturation degree is the most relevant parameter linked to the resistivity and the corrosion rate values. The exposition of concrete to outdoor conditions will induce a saturation degree related to the particular climate. External moisture and temperature cycles will be followed by parallel changes in the concrete moisture content, which in consequence induce an evolution in the corrosion rate following dayly and seasonal cycles.

Measurements of corrosion in concrete using electrochemical

impedance spectroscopy (EIS) Electrochemical impedance spectroscopy (EIS) is relatively new method for testing corrosion and its evolution. With this method we can determine the main corrosion mechanism from the measured impedance spectrum and thus evaluate basic values of corrosion parameters. Its main feature is electrical stimulation of corrosion system with known sinus signal with small voltage amplitude in particular frequency range. While the system is being stimulated the current response is measured. For further details see References section. The acquired spectrum is then modeled with relatively simple theoretical models constituating basic electrical components such as capacitors, resistors etc. From the modeled spectrum we can characterize start of corrosion process and degree of corrosion and its evolution.

Figure 6: Nyquist diagram of measured impedance spectrum of steel rebars in fresh concrete

(dots) and its electrical model (line) with impedance equation. The spectrum has real and imaginary impedance on x and -y axis respectively. This spectrum represents basic state of steel rebars in concrete where steel is still in its passive state and corrosion has not started yet. Electrical model (Rs stands for solution resistance, Cdl stands for double layer capacity (passive film) and Rp is for polarization resistance (passive film)). Sign ω is angular frequency.

13

5,0E+01

1,0E+02

1,5E+02

2,0E+02

2,5E+02

3,0E+02

3,5E+02

4,0E+02

4,5E+02

5,0E+02

2,0E+02 4,0E+02 6,0E+02 8,0E+02 1,0E+03 1,2E+03 1,4E+03 1,6E+03 1,8E+03 2,0E+03 2,2E+03

Real (Ohm)

-Im

ag

(O

hm

)

ω → ∞

ω → 0

Figure 7: Nyquist diagram of measured impedance spectrum of steel in carbonated concrete,

wetted in 3,5 % NaCl solution and its electrical model with impedance equations. This spectrum represents that corrosion of steel is present and already evolved to a great extent in concrete sample. Electrical model ( Rs stands for solution resistance, Cdl is for double layer capacity (passive film still present in some areas), Rct stands for charge transfer resistance (rust layer) and w is Warburg impedance (diffusion impedance in rust layer). F is Faraday constant, A is rebar cross section, D is diffusion coefficient, n is number of particles (ions), T is temperature, CR/O

∞ is concentration of oxidants and reducents at infinity, ω is angular frequency and R is gas constant).

Figure 8: Concrete sample before exposing in various solutions took place and steel rods

when breaking the sample after last measurement in 3,5 % NaCl solution. Corrosion products on steel rods are well visible.

14

Preventing corrosion There are a variety of methods for preventing corrosion or at least to slow down the corrosion process. The most common are listed below. Galvanisation Galvanized reinforcing steel is effectively and economically used in concrete where unprotected reinforcement will not have adequate durability. The susceptibility of concrete structures to the intrusion of chlorides is the primary incentive for using galvanized steel reinforcement. Galvanized reinforcing steel is especially useful when the reinforcement will be exposed to the weather before construction begins. Galvanizing provides visible assurance that the steel has not rusted and requires no on-site repair, unlike most other coatings. Galvanized reinforcing steel can withstand exposure to chloride ion concentrations several times higher (at least 4 to 5 times) than the chloride level that causes corrosion in black steel reinforcement. While black steel in concrete typically depassivates below a pH of 11.5, galvanized reinforcement can remain passivated at a lower pH, thereby offering substantial protection against the effects of concrete carbonation. Figure 9: Galvanized rebar can be treated exactly like black rebar when it is being installed

and is always protected from corrosion by the hot-dip galvanized coating. Cathodic protection (CP) In this process the anodes, power supply and control systems are permanent, and a range of anodes can be used. The aggressive anodic reaction is isolated to a corrosion resistant anode while the harmless cathodic reaction occurs at the surface of the steel reinforcement. This process creates additional hydroxyl ions, rebuilds the passive alkaline layer and repels chloride ions. CP has been used on hundreds of reinforced concrete structures around the world and has potential for the conservation of historic brick and stone masonry, terracotta and statuary where steel and iron has been used to provide reinforcement or a structural frame

15

Electrochemical chloride migration (Desalination) This process uses a temporary anode, power supply and monitoring system to apply 50 volts direct current to the steel. The positive charge repels the negatively charged chloride ions and rebuilds the passive layer over a period of four to six weeks.The technique has been used to successfully treat more than 50 structures in the UK, continental Europe and North America. Re-alkalisation This system is the equivalent of desalination for carbonated structures. It relies on the principle that the hydroxyl ions produced at the cathode re-alkalise the concrete from the reinforcement outwards. This is linked with a wet anode at the surface that contains calcium carbonate, which moves under electro-osmotic pressure and re-alkalises the concrete from the surface inwards Corrosion inhibitor repair techniques A recent development is the impregnation with chemical corrosion inhibitors which are widely used in the power generation, chemical and manufacturing industries. Recently, attempts have been made to introduce these chemicals into hardened concrete. If successful, then these could be good, relatively simple methods of increasing the life span, reducing maintenance and providing a 'minimum intervention' method of slowing or stopping corrosion. One of the most effective corrosion inhibiting systems is also one of the simplest. An inorganic admixture made with calcium nitrate, which is added to the concrete before casting, performes equally well or better than more complicated systems that include sealers applied to the concrete or coatings on the steel bars.

Conclusion Corrosion of steel in concrete can be seen to be a significant problem for many reinforced concrete structures if moisture is present. If there is no salt to cause corrosion in the short term, carbonation will affect most structures over the centuries. If the structure cannot be kept dry then there is a range of techniques that can be used depending on the structure, its condition and the cause and extent of the problem. Repairing damage caused by corrosion is a multi-billion problem. Observations of numerous structures show that corrosion of reinforcing steel is either a prime factor, or at least an important factor, contributing to staining, cracking and/or spalling of concrete structures. The effects of corrosion often require costly repairs and continued maintenance during the life of the structure.

16

References 1. R. Greef, R. Peat, L. M. Peter, D. Pletcher, J. Robinson: Instrumental methods in electrochemistry, John Wiley & Sons, New York (1985) 2. Robert Cottis, Stephen Turgoose: Electrochemical impedance and noise, NACE International, Houston (1999) 3. Leopold Vehovar: Korozija kovin in korozijsko preskušanje, Ljubljana (1991) 4. Dr. Milenko V. Šušić: Osnovi elektrohemije i elektrohemijske analize, Beograd (1980) 5. J. R. Macdonald, Ed.: Impedance Spectroscopy Emphasizing Solid Materials and Systems, John Wiley & Sons, New York (1987) 6. V. D. Jovic, Determination of the correct value of Cdl from the impedance results fitted

by the commercially available software (2003), http://www.gamry.com 7. Equivalent Circuit Modeling Using the Gamry CMS300 Electrochemical Impedance Spectroscopy Software, Instruction manual, GAMRY Instruments© 8. Milena Turičnik Deutsch: Vpliv aminskega inhibitorja na korozijo armiranega betona, Magistrsko delo, Fakulteta za kemijo in kemijsko tehnologijo, Ljubljana (2001) 9. Primož Vavpetič: Študij korozijskih procesov v betonskem jeklu z elektrokemijsko

impedančno spektroskopijo, Diplomsko delo, Ljubljana (2005) 10. http://www.nrc-cnrc.gc.ca/highlights/2008/0803bridges_e.html 11. http://www.cadman.com/section.asp?catid=1235&subid=1238&pageid=3162 12. http://www.galvanizeit.org/showContent,278,322.cfm#top 13. http://www.concretenetwork.com/concrete/concrete_admixtures/corrosion_protection.htm