Long-term performance of different aluminum alloy designs as sacrificial anodes for rebars

This paper presents the performance of various cathodic-protection designs using Aluminum alloys to protect prestressed piles. The results obtained with different system designs (bracelete type-Al/Zn/In alloy, thermosprayed aluminum (3'year evaluation) and conventional Al/Zn/In anocies in an epoxy-painted steel bracelet (IZ-year evaluation)), indicated that all of these systems may be used as sacrificial anodes for pile protection. However, the thermosprayed aluminum type can not be used in prestressed concrete piles because the very negative potentials ( < 1 1 0 0 mV vs. CU/CUSO4) they supply to the reinforcement could lead to hydrogen embrittlement.


INTRODUCTION
The past few years have seen an increase in research on the use of sacrificial anodes for the cathodic protection of the reinforcement in reinforced concrete structures.This increase is due not only to simplicity of application and low system-maintenance requirements, but also to its application in prestressed concrete structures, in which the risk of hydrogen embrittlement would be eliminated, given the low and controlled protection potential.
Zinc is one of the materials that have undergone major evaluation, especially in the US^^' \ However, because of its low protective capacity in time, the location of the system is under constant modification.A new system, known as "Zinc-Hydrogel", was recently designed in an effort to control this situation.This system, which consists of a sheet of zinc foil backed by an ion-conducting adhesive, appears to be producing better results^^ ^l This system has been under study at Centro de Estudios de Corrosión (CEC), Universidad del Zulia, since 1984^^^'^^, following a promising line for the use of an Al/Zn/In alloy in specific situations, especially for prestressed piles in tide and splash areas^ \ Studies have therefore lately been centered on the search for a safe and easy way to install these anodes embedded in highstrength porous mortar^^^l designed to allow corrosion-product diffusion from this aluminum n Centro de Estudios de Corrosión, Facultad de Ingeniería -Universidad del Zulia Maracaibo -Venezuela.Tel/Fax: 58-61598784/525732.alloy.Other investigators, Funahashi^ ^ among them, have been evaluating an Al-lZ/Zn-O.Z/In alloy, with very good results.
This work presents the latest results obtained with three different anode locations: Bracelet-type Al/Zn/In, thermosprayed Al. (3-year evaluation) and conventional Al/Zn/In anodes in an epoxycoated steel bracelet (IZ-year evaluation).

Experimental reinforced concrete piles with cathodic protection system
Four cylindrical reinforced concrete piles were built (Table I): (0 = 15.24cm; height: 160 cm).Chloride at 0.1 % w/w (concrete-based), was added to the mixture.The reinforcements in these piles are a steel structure, which is the metallic element to be protected by applying cathodic protection.The steel structure comprises three rebars with 0 = 1.27 cm (Vz ii^) X 150 cm long, joined together with a helicoidal rebar 0.653 cm (V4 in) x 414 cm (total area: 0.261 m ).The rebars were welded at several points to ensure greater electrical continuity.Graphite was used for the reference electrodes (embedded in the concrete), which were prepared according to the CEC procedure^ K The electrodes were located 3 mm from the reinforcement, at two (2) different heights: at 50 cm and 1 m from the top of the reinforcement.A Cu/CuS04 external electrode was also used for measuring potentials.

Al/Zn/In anodes
The Al/Zn/In alloy was fused in accordance with the green and dry-sand molding method^^ \ The anodes consist of two 1.8 cm thick x 120 cm long bracelet shells (Fig. 1).
A hole was drilled into the upper part of the bracelets, into which a screw was later inserted, with a No. 12 TW plaited cable soldered to it.The cable was later covered with an epoxy resin that serves as an insulator and prevents the formation of a galvanic couple at that point.This connection would later be used for applying the cathodic protection system (anode -metal structure).A hole was also made in each bracelet in order to measure the reinforcement potentials vs. Cu/CuS04 once they had been placed on the piles.The two shells were screwed together with galvanized bolts.At each contact point, epoxy resin was spread on the outside of the bolt nuts to avoid the formation of a galvanic couple.

Aluminum anode
Located on one of the piles by thermospraying, using COR Resist 29220 powder supplied by Eutetic + Castolin.When the anode was in place, two rawlplugs (to be used for installing the cables for the cathodic protection system) were inserted into the pile.
In both cases, a mortar, specially evaluated in CEC^^^^ for this specific situation, was cast after the anodes were in place.Chloride (0.1 % w/w) was added to the mixture.Of the four piles, two were fitted with aluminum bracelet-type anodes; one with a thermosprayed aluminum anode and the fourth was left unprotected as reference control (for comparison).Once installed, mortar was cast to embed the anodes and create the electrolyte that would allow the cathodic protection system to act (Fig. 1).Reinforcement potentials were measured before connecting the cathodic protection system, using a multimeter and the graphite electrode embedded in the concrete, as well as the purpose-built Cu/CuS04 electrode inserted in the orifices.All of these piles, including one with conventional anodes in an epoxy-coated steel bracelet^ ^ (previously exposed for 12 years) and embedded in the same mortar, were placed on a purpose-built bench (Fig. 2a), which was then submerged in Lake Maracaibo (Fig. 2b).

Electrochemical measurements
The cathodic protection potential was measured periodically on all piles, using the graphite electrode embedded in the concrete and an external Cu/CuS04 electrode.The current drained by the anode was determined one year after the system was installed.A cathodic depolarization test using both electrodes was then run.The corrosion rate of the Al and Al/Zn/In anodes and the resistivity  of the concrete in which the anode was embedded were measured using

Anode/concrete interface analysis
Scanning Electron Microscopy (SEM) was used to evaluate the Anode/Mortar and Anode/Concrete interface to determine anode oxidation-product diffusion through the mortar/concrete.To that effect, 2.5 cm diameter cores were extracted from each cylinder at the level of one of the reinforcement steel bars.Core length depended on the rebar location.

Electrochemical Evaluation
Figures 3 and 4 show the average reinforcement potential variation after the sacrificial anodes were connected and the test piles located in Lake Maracaibo.Protection potentials are obtained in all cases, with more negative potentials obtained from the one thermosprayed with Al.However, the potentials reached with the Al/Zn/In anodes, after approximately 3 years exposure, are in the order of-900 mV vs. CU/CUSO4.Figures 5 and 6 show the depolarization tests results.The test piles protected with the bracelettype anodes reached depolarization values > 100 mV after 4 h.However, this was not observed with the piles thermosprayed with Al.Apparently, one of the rawlplugs used for the electrical connection to the anode through the cable made contact with the reinforcement; so when the cables were disconnected externally, the reinforcement potential remains very low.
Average currents, measured at the end of the exposure period, were 5.7 mA/m and 4.9 mA/m for the system with Al/Zn/In anodes (bracelet type) and Al (thermosprayed), respectively.These currents coincide with the ones used to protect reinforced concrete in several countries^^' ^' ^' ^' ^^ ^^' ^' ^ ^^\ Figure 7 shows the results of the test piles exposed approximately 12 years ago^ % with Al/Zn/In anodes distributed on an epoxy-coated steel bracelet.Note that potentials are still at protection levels (< -850 mV vs. CU/CUSO4) after all this time.In this case, the current drained is 4.3 mA/ml Additionally, after depolarization, the test piles were allowed to become completely depolarized and the corrosion rate of Aluminum and the      Al/Zn/In alloy was assessed, as well as the electrical resistance and resistivity of the mortar embedding the anodes.Table II shows the results.It can be seen that the corrosion rate is not very nign (0.012 -0.031 mm/year for the Al/Zn/In and Al anode, respectively), which would give a prolonged useful life to these anodes.However, the Ah thermosprayed anode has the least corrosion potential (< -1000 mV vs. CU/CUSO4) and greater corrosion rates (3.543 |LlA/cm^ (0.031 mm/year), approximately), which is not needed for guaranteeing reinforcement protection.On the contrary, this would mean a shorter useful life for the anodes and a more negative potential for the reinforcement something that could be harmful for presttessed steel.It was also observed that the resistivity of the mortar covering the anodes remains low and almost constant, regardless of anode type (5 kn • cm), which is good for conducting the galvanic current in the electrolyte.The unprotected cylinder was also evaluated (Fig. 8).Note the low potentials, indicative of reinforcement corrosion.The corrosion rate measured was also very high (>0.5|LlA/cm), as expected.

Anode/Concrete, Anode/ Mortar Interface SEM analysis
Al diffusion through the mortar/concrete was observed in all cases.Figure 9 shows a typical case^ the morphology of the area evaluated for the Ah thermosprayed anode/embedding mortar interface and mapping per element (Al, Ca, CI).Note the Al corrosion-product diffusion toward the outside as it goes further away from the anode/rtlOftat interface, as well as CI enrichment at the interface.
It is important to point out that no cracks were observed in the mortar embedding the anodes when the cores were extracted.However, an inside crack was seen at the anode/concrete interface in some parts of the cylinders in which the bracelettype anode was inserted.This might have occurred     at the time this anode was installed in the concrete cylinder.However, this did not affect reinforcement polarization, as can be seen in figures 3, 4 and 8. Some localized fissures were also seen in the spots where the anode was connected to the reinforcement.

CONCLUSIONS
The following conclusions may be drawn from the electrochemical evaluations: -Al/Zn/In anodes (bracelet type) may be used as sacrificial anodes for protecting prestressed piles.-The Al anode (thermosprayed type), may also be used as a sacrificial anode, but no in prestressed steel piles because they supply very negative potentials to the reinforcement -which could lead to hydrogen brittleness.
-The resistivity of the mortar covering the anodes remains low and constant, regardless of anode type (5 KO • cm), which is good for conducting the galvanic current in the electrolyte.

>
a: #8 cables welded to rebars (Vz") > b: #12 cables welded to the graphite electrode, set at 3 mm from the steel (0 = 1/2") at 0.50 m y 1 m from the top > c: Concrete Pile > d: #10 cable joined to the sacrificial anode by means of resin-coated screws (welded to the cable to enhance electrical conductivity) > e: Rebar (steel) 0 = V2" > f: Helicoidal stirrup 0 = 14" > g: Headless screws with two nuts, one at each end, coated with epoxy resin to avoid a galvanic couple > h: Sacrificial anode (Al/Zn/ln) I: Mortar covering the anode.

Figure 2 .
Figure 2. a) Workbench, b) Workbench with the different piles submerged in Lake Maraca i bo.

Figure 7 .
Figure 7. Piles with Al/Zn/ln anodes distributed on an epoxy-coated steel bracelet.

Figure 8 .
Figure 8. Test-Pile reinforcement potential variation with time, before and after immersion.

Figure 9 .
Figure 9. a) Morphology of the area evaluated for the interface of the Al-thermosprayed anode/embedding mortar and b) mapping per element (Al, Ca, Cl).

Table I .
Mixture used for experimental concrete piles

Table II .
Electrochemical parameters of the different sacrificial anodes evaluated