Professional Documents
Culture Documents
1 of 9
inhibition are claimed to be lower than inorganic inhibitors (6). There is however, little
published work on the mechanisms of the inhibition, the long-term performance of the
inhibitors or the effect of their concentration on the level of inhibition.
ctp52.doc
2 of 9
The present paper compares the inhibitive action of four systems, an inorganic oxidizing
inhibitor, sodium nitrite having nitrite as the main corrosion inhibiting component; a
simple organic amine, dimethylethanolamine; a quaternary amine nitrite,
dicyclohexylamine nitrite, which may combine the effects of both organic amines and an
inorganic oxidizing anion; and an aromatic amine based commercial migratory inhibitor
labeled MCI. In particular, the influence of these inhibitors on a prepassivated steel
surface, representing the state of the steel in a newly built structure (admixture situation),
and a corroding steel surface as would be found in the case of an existing corroding
structure (remedial situation), is investigated. The possibility of a synergistic effect of
the mechanism of barrier film formation by organic amines and interfacial chemical
reactions by inorganic inhibitors will be investigated. Attempts are made to correlate
inhibition capacity to the ratio of chloride: nitrite in the case of the nitrite based
inhibitors.
Since (in the remedial situation at least) the capacity of the inhibitors to perform their
function depends upon their rate of migration through concrete, this aspect of their
performance will also be examined. In this case the physico-chemical properties of the
inhibitors and the permeability of concrete play an important role. These inhibitors may
also be called upon to act in concert with cathodic protection. Since, in water, some of these
inhibitors are ionic in character, the possibility of a conjoint action of cathodic protection
with inhibitors, will be examined. Indeed, in the remedial situation, where an inhibitor is
required to reach the steel - concrete interface by migrating through concrete to control
further corrosion driven deterioration, this may be the only way in which an appropriate
quantity of the inhibitor can be applied at the steel surface in a short time frame.
2. EXPERIMENTAL METHODS
2.1 Linear Polarization Resistance (LPR) Measurements
Experiments were carried out in a simulated concrete pore water solution of saturated
calcium hydroxide. Fixed concentrations of inhibitors and of corroding chloride ion
(chosen to mimic a marine contaminated environment) were used. The chloride
concentration of the simulated pore water solution was 1.2% wt. and the concentration of
the inhibitors was 0.05 M in the case of both organic and inorganic inhibitors and 1% wt.
in the case of the commercial inhibitor MCI. Corrosion rates were estimated using the
linear polarization resistance technique, assuming a linear relationship between the
corrosion rate and the reciprocal of the polarization resistance icorr = B/Rp, where B is a
constant. The inhibitive capacity of each inhibitor is reported as the % inhibition. This
is the difference in corrosion rates between specimens immersed in inhibited and non
inhibited solutions expressed as a percentage of the corrosion rate in the uninhibited
solution (or a control solution). Details of the experimental procedure have already been
reported (7).
ctp52.doc
3 of 9
Organic amine inhibitors can diffuse through concrete in a gaseous phase besides
diffusing as dissolved species in a carrier solvent or water. Laboratory diffusion cells
have been developed to measure the diffusion/migration of these inhibitors (Figure I).
The diffusion cells have two compartments separated by a concrete membrane.
This design allows variability in the type of concrete membrane, i.e., water/cement ratio,
aggregate type, membrane thickness etc. Two different water/cement ratios, 0.45 and
0.6, and membrane thickness, 30 mm were used. Compartment A contains an aqueous
inhibitor solution and compartment B contains water. The amount of inhibitor diffusing
from compartment A to B through the concrete membrane is monitored using an amine
sensitive ORION 95-12 electrode(8). The specific amine sensitive electrode uses a
hydrophobic gas permeable membrane to separate the sample solution from the electrode
internal solution (0.1 M NH4Cl). Dissolved amine in the sample diffuses through the
membrane until the partial pressure of the amine is the same on both sides of the
membrane. In any given sample the partial pressure of amine is proportional to its
concentration and this is related to potential, E via the Nernst equation. The potentials are
measured for various known concentrations of a given inhibitor solution and a calibration
curve is thus obtained. This curve is used to measure inhibitor concentrations in
compartment B of the above cell.
2.3 Electromigration
The main consideration in this case is that the organic inhibitor molecule should have a
charged component such as a quaternary ammonium ion (R2NH2) in aqueous solution.
Such positively charged amine species can be driven to the negatively charged cathode
by the application of an electric field, thus enhancing their rate of diffusion. The
objective of these studies is to optimize electrical conditions required for successful
inhibitor injection. The effect of electric field strength and application time is being
investigated and preliminary results will be discussed in the presentation.
ctp52.doc
4 of 9
5 of 9
the building up of the passive film followed by its subsequent rupture as the available
nitrite is used up. In the case of dimethylethanolamine there is a continuous fall in the
corrosion potential as the inhibitor efficiency increases, in this is consistent with a
predominantly cathodic inhibition process. A possible explanation of the decrease with
time of inhibition by the dimethylethanolamine is that the inhibitor is absorbed at bare
metal sites, produced by the action of chloride ions on the passive film and the decrease in
the inhibitor efficiency corresponds to the complete use of the available inhibitor leading
to insufficient inhibitor molecules to block new bare metal sites. It should be noted that
this mechanism is in some conflict with the trend in the Ecorr values. This is yet to be
resolved.
Figure IIA- % Inhibition (Admixture) Dicyclohexylamine Nitrite, Dimethylethanolamine,
MCI and Sodium Nitrite
ctp52.doc
6 of 9
Evidence that the chloride attack is responsible for the steady decline of the inhibition
with time was obtained by removing coupons that had reached maximum protection
during a remedial experiment, and placing them into a chloride free environment with
and without added inhibitor. In these cases the % inhibition was maintained for at least
700 hours.
The behavior of dicyclohexylamine nitrite under remedial conditions is complex.
Ionization and subsequent hydrolysis (10) will yield NO2 - anions and (R2NH2) + cations
and R2NH2OH where R = C6H11. The inhibitor is thus assumed to be both anodic and
cathodic. The nitrite anion itself is an anodic inhibitor and hence reacts chemically to
assist in repairing the oxide film at bare anodic sites. This can be evidenced by the
increase in the Ecorr values on addition of the inhibitor to the corroding steel.
Adsorption of the amine component of the inhibitor molecule over the bare metal or oxyhydroxy sites may result in the formation of barrier film. This adsorption of the inhibitor
appears fairly rapid as seen from the rise in the % inhibition upon addition of inhibitor.
In addition, the cooperative adsorption (7,9,11) of the protonated (R 2NH2)+ groups over
the chloride covered sites on the steel surface may result in the stabilization of the
inhibitor barrier film and hence a many fold increase in the inhibition level, compared
with the other inhibitors. The corrosion potential of coupons exposed to the commercial
MCI inhibitor showed a more or less continuous decrease, suggesting that it is acting
primarily as a cathodic inhibitor. It differs from the other cathodic inhibitors in that the
level of inhibition, once attained is maintained for the duration of the experiment. In the
absence of the full knowledge of components of MCI a rationalization of this behavior
may be proposed on the basis of the better adsorptive capacity of this inhibitor resulting
in a barrier film difficult to break or penetrate by the corrosive chloride ions.
Figure IV is concerned with the admixture situation and compares plots of inhibition
levels with time for varying concentrations of sodium nitrite and dicyclohexylamine
nitrite. The negative inhibition observed in the case of low concentration (0.01M) of
sodium nitrite is characteristic of an anodic inhibitor. It appears, for both of the
inhibitors, that an increase in the concentration results in a corresponding increase in the
inhibition. Sodium nitrite, although showing negative inhibition values at 0.01 M
concentration (chloride: nitrite ratio 26.4), improved to the inhibition levels exhibited by
dicyclohexylamine nitrite, as the inhibitor concentration level increased to 0.1 M
(chloride: nitrite ratio of 2.6). This is in accordance with Berke and Rosenberg (12), who
ctp52.doc
7 of 9
recommend a ratio of chloride to nitrite to be less than 2 in the case of calcium nitrite, for
any meaningful inhibition to be obtained.
3.2 Inhibitor Migration
In the simulated concrete solution, the concentration of the organic inhibitor at the steelsolution interface is the same as that in the test cell, whereas in reinforced concrete, in the
remedial situation, inhibitor concentration at the metal surface is lower than that at the
point of inhibitor application. The build up of inhibitor at the rebar will depend on the
diffusion rate of the inhibitor from the surrounding areas. Hence a further aim of this
work is the determination of the rate of diffusion of the organic amine inhibitors through
concrete. This has been achieved by designing a series of two compartment cells
(5,13) as shown in Figure 1. Diffusion of dicyclohexylamine nitrite,
dimethylethanolamine and commercial inhibitor MCI (based on an aromatic amine
compound) through 30 mm thick concrete membranes having 0.45 and 0.60
water/cement ratio, is being monitored. Figure V shows the inhibitor, which has
migrated through the concrete membrane with water/cement ratio of 0.6, as a percentage
of the original concentration in compartment A. In this case, the high permeability of the
concrete clearly allows the slow but steady diffusion of the commercial inhibitor MCI;
dicyclohexylamine nitrite, however diffuses only to a limited extent. In the case of
dimethylethanolamine, though there is an increase in the diffusion up to 1% initially, at
longer times, there is a decrease in the concentration. This is unexpected and may be due
to either volatilization of the inhibitor from compartment B or physical absorption onto
surfaces, which would decrease the free amine content. Preliminary experiments using
concrete with water/cement ration of 0.45 suggest neither MCI nor dicyclohexylamine
nitrite diffuses significantly through this less permeable concrete. On the other hand,
dimethylethanolamine shows better diffusion after only 10 days. The insignificant
diffusion in the case of dicyclohexylamine nitrite could be attributed to its low solubility
in water and comparatively low vapor pressure (14).
Figure IV - Comparison of Different Concentrations of Dicyclohexylamine Nitrite and
Sodium Nitrite
ctp52.doc
8 of 9
These early results suggest that the inhibitor diffusion through concrete depend both on
the concrete composition/permeability and the physical and chemical properties of the
inhibitors. The degree of permeability of concrete is dependent on water/cement ratio,
degree of microcracking and thickness of the cover (15). Higher water/cement ratio
apparently leads to either a greater number of pores or to larger pores, both of which can
lead to increased permeability. High water/cement ratio results in a more rapid migration
of inhibitor. The vapor pressure and the solubility of the inhibitor in water are the other
two factors determining the transport of these surface applied migratory inhibitors
through concrete. A high vapor pressure may, however, result in both faster migration of
the inhibitor through concrete towards steel reinforcement and also heavy losses to the
atmosphere. Similarly, high solubility of the inhibitor in water might lead to leaching out
of the inhibitor. Further, the higher the vapor pressure or solubility is likely to reduce the
residence time of the inhibitor molecules in the barrier film formed over the steel surface,
thereby adversely affecting the long term inhibition properties. A detailed investigation
of all these factors is necessary in order to predict the success rate of migratory inhibitors.
Given that most of the amine inhibitors under investigation will be protonated, and
therefore positively charged, in aqueous solution, one method to enhance migration of the
inhibitors towards steel may be to use electromigration (16). This technique has certainly
been used successfully in dechlorination of concrete (17). Work in our laboratories is
underway to investigate the synergy between cathodic protection and chemical inhibition.
Preliminary experiments have begun based on the diffusion cell described above (Figure
I). The effect of electric field strength and application time is being investigated.
4. CONCLUSIONS
The results indicate that organic aromatic amine based inhibitors, dicyclohexylamine
nitrite and commercial MCI, offer excellent inhibition for longer times in contrast to the
inorganic anodic inhibitor sodium nitrite. It is also shown that the effectiveness of these
inhibitors is dependent on the relative concentrations of chloride and inhibitor species and
the corrosion state of the steel at the time of the treatment. In the case of a quaternary
amine inhibitor dicyclohexylamine nitrite, a very high efficiency is observed for extended
periods due to the synergistic effect of the mechanism of barrier film formation by
organic amines and interfacial chemical reactions by inorganic inhibitors. Preliminary
results of the diffusion measurements in the case of concrete with high water/cement ratio
indicate that there is a rapid initial diffusion of dimethylethanolamine, and a slow but
ctp52.doc
9 of 9
2.
S. Berke, D. W. Pfeifer and T. G. Weil, Concrete International, 10, No 12, pp. 45-55,
December (1998).
3.
4.
5.
6.
7.
8.
9.
10 of 9
16. Hettiarachchi and A. T. Gaynor, Materials Performance, 31, No 3, pp. 62-66, March (1992).
ctp52.doc
11 of 9