Corrosion failure of post-tensioning (PT) tendons with pre-packaged thixotropic grout has been documented in Florida bridges. Analysis of the deficient grout in Florida PT bridges, where severe corrosion developed, indicated elevated sulfate ion concentration, high pore water pH, low chloride ion content, and enhanced moisture content. Limited information is available on the corrosion behavior of PT strand in grout materials with elevated sulfate ion content. In this work, the corrosion performance of steel strands embedded in deficient grout was evaluated. The deficient grout was cast in large scale test assemblies that would represent geometry and material conditions found in Florida bridges, where PT strand corrosion occurred. Large scale mock up tendons (developed by the modified inclined tube [MIT] test) of about 4.57 m long were cast using expired grout materials, excess of 15% water, and sulfate ion concentrations to create deficiencies and corrosion conditions in the grout environment. Corrosion measurements such as corrosion potential (Ecorr) and corrosion current density (icorr), as well as potentiodynamic polarization tests were performed on the steel probes located along the MIT samples in contact with the deficient grout. Grout characterization parameters included the determination of the grout resistance (Rs), as well as the moisture and sulfate ion content. Also, visual inspection of the MIT samples (steel strand and steel corrosion probes) after a time was performed, as well as the corrosion products characterization. Testing confirmed that enhanced corrosion can occur in the deficient grout with high free sulfate ion accumulation.

Severe corrosion of post-tensioning (PT) tendons with pre-packaged low-bleed specified grout products have been documented in Florida bridges with less than 8 y of service.1-3  As reported elsewhere,1  grouts used in Florida bridges had susceptibility to segregate and form physically and chemically deficient grout of varying characteristics. Deficient grout with a wet plastic and white chalky consistency had enhanced moisture content sometimes greater than 50% to 70% by mass compared to moisture content less than 20% by mass in hardened grout.4  In 2014, Hamilton, et al.,5  used a modified version of the inclined tube (MIT) test, identifying material and casting conditions that may promote grout segregation. Results showed that parameters such as shelf life of bagged PT grout, storage conditions, and excessive mix water can promote formation of physically deficient grout.5  However, further validation that grout segregation can create adverse pore water chemistry promoting corrosion of steel is required.

The severe corrosion observed in the Florida bridges was consistently associated with deficient grout characterized by having high moisture content, low total chloride ion content, high free sulfate ion concentrations, and high pore water pH.4  The chemical characteristics of the deficient grout collected from one bridge where corrosion-induced failure of tendons occurred are presented in Table 1.4  The severe corrosion there was not well related to chloride ions and grout carbonation, and the locations of the severe corrosion were not consistently collocated in grout void spaces. This would indicate different corrosion mechanisms than those conventionally associated with corrosion of pre-stressed steel.

TABLE 1

Chemical Components of Grout Samples from the Recent Failed Bridge in Florida4,(A)

Chemical Components of Grout Samples from the Recent Failed Bridge in Florida4,(A)
Chemical Components of Grout Samples from the Recent Failed Bridge in Florida4,(A)

Similar corrosion failures of PT strand in contact with deficient grout have been reported in Europe.6-9  Carsana and Bertolini8  have recently documented a case of corrosion of PT strands in segregated grout that led to failure of external tendons of a bridge around 2 y after construction. The segregated grout had high content of soluble compounds such as alkalis, sulfate, and hydroxyl ions, which was suggested to provide active corrosion conditions where pH values can exceed 14.9  While some similarities in deficient grout characteristics are evident, namely grout segregation with elevated concentrations of alkali and sulfate ions, it is to be noted that the proprietary grout products used in the Florida bridges may create different pore water chemistries. Of note, deficient grout in the Florida cases often had accumulation of silica fume, of which hydration may lead to lower pH levels.10  Comparison of findings in Part 111  of this series indicates that pore solution pH is an important factor in corrosion development in alkaline sulfate solutions.

Information from the technical literature indicated that there could be conditions where elevated sulfate ion concentrations may lead to corrosion in alkaline solution;11  however, limited information is available on the role of sulfate ions in cementitious environment.12-15  In 1970, Gouda and Halaka16  investigated the role of sulfate ions (up to 8%) in the corrosion of steel in concrete. Reported test results showed no breakdown of steel passivity with the presence of sulfate ions and the authors there posed that alumina-containing cement phases react with the sulfate to form insoluble compounds. However, their testing with cementitious leachate solution with elevated sulfate ion concentrations did result in steel corrosion. Al-Tayyib and Khan17  reported only modest increase in corrosion rate of steel in concrete with up to 1.8 kg/m3 sulfate ion by weight of concrete.

In the present work, the objective was to verify that grout segregation can create physical and chemical deficiencies (including elevated moisture and sulfate ion content) that can promote corrosion of steel.

Modified Incline Tube Test

In the present study, the MIT test, defined by Hamilton, et al.,5  was used for validating the role of sulfate ion on the corrosion of steel in deficient grout. The MIT test provided the benefit of assessing the extent of corrosion with varying degrees of grout deficiency resulting from pumping procedures and possible segregation associated with vertical elevation deviation.

To further exacerbate grouting conditions, all MIT samples here were cast with expired grout materials in non-ideal conditions in order to further promote grout deficiencies. Only one grout product was utilized in these experiments. Grout test conditions were made, as listed in Table 2. All grout materials used had at least an excess of 15% mix water over manufacturers’ recommended high-end limit. Enhancement of sulfate ion content (from pre-mixed solutions with sodium sulfate salt) was included to help elucidate the role of sulfate in corrosion development. A base case of deficient grout (Case 1) contained no external additions of sulfate ions. Cases 2 through 4 contained ∼0.09%, ∼0.9%, and ∼5.5% sulfate ion by cement mass, respectively. Because of the distress placed on the materials, results of testing were not meant to be indicative of the grout product performance when appropriately used, but rather to validate cause of strand corrosion in chemically deficient grout resulting from segregation.

TABLE 2

Grout Sample Conditions and Sulfate Content

Grout Sample Conditions and Sulfate Content
Grout Sample Conditions and Sulfate Content

Eight tendons that were 4.57 m long, made up of 7.62 cm diameter clear PVC duct, inclined 30° from the horizon, were cast with the varying grout conditions described above. Two 7-wire unstressed steel strands were embedded within the duct. Grout batching and mixing used a high-shear grout plant.

Corrosion probes were placed at five points (from higher to lower elevation) along the length of the tendons, as depicted in Figure 1. Each electrode set contained a three-electrode arrangement for conducting electrochemical measurements. The working electrode consisted of a 0.32 cm diameter steel rod with the bottom 2 cm length exposed in the grout. The remaining upper portion of the electrode was coated with a two-part epoxy. An activated titanium rod (periodically calibrated with a copper/copper-sulfate reference electrode [CSE]) and an embedded activated titanium mesh placed along the length of the tendon were used as the reference and counter electrodes, respectively. Some of the steel corrosion probes contained a rubber gasket in order to create a crevice between the gasket and the rod. Duplicate samples were made for each testing condition.

FIGURE 1.

Location of corrosion probes in the clear pipe.

FIGURE 1.

Location of corrosion probes in the clear pipe.

Close modal

Deficient Grout and Steel Characterization

Characterization of the grout materials from Case 1 included moisture content, chemical content, pH, grout resistance (Rs), and petrography in order to differentiate properties of the deficient grout without the influence of enhanced external sulfate additions. For Cases 2 through 4 with external sulfate additions, moisture content, sulfate ion content, pH, and Rs measurements were taken for comparison.

For analysis requiring grout extraction, material was sampled from the higher elevation (∼0.1 m from top) and the lower elevation (∼2 m from top) of the tendons, as detailed in Figure 1. The collected grouts were kept inside a closed environment until the beginning of their analysis. For the chemical analysis of grout for Case 1, additional tendon locations at intermediate distances were sampled. Moisture content was measured by gravimetric methods as described in ASTM Standard C566-13.18  An ex situ leaching method was used for obtaining grout leachate for chemical and pH analysis. The procedure included drying the powder samples at 55°C for 24 h and to combine 1 g of the dried powder with 1:10 leaching volume at 66°C for 15 h to 18 h. Grout leachate pH was measured with a pH glass electrode. Chemical analysis was done by ion chromatography, inductive coupled plasma, and other chemical analysis techniques. For Case 1, chemical analysis of Ca2+, Na+, K+, Cl, SO42−, S2−, NO3, and NO2 was performed. For Cases 2 through 4, only SO42− concentrations in the leachate were measured. In situ Rs measurements were made by three-point resistance. Thin-film petrography was conducted for two samples from Case 1 and two samples from deficient grout from a Florida bridge.

The corrosion probes as well as the 7-wire strand embedded in the deficient grout were visually inspected at the end of the test. Then, steel corrosion products (removed from the strand) were characterized by x-ray diffraction (XRD) by using a diffractometer D5000 with a data acquisition Diffrac Plus software. The testing procedure included diffraction scans with 2θ from 10° to 60° with a 2°/min scan rate. XRD data from corrosion products from a Florida bridge4  and from steel electrodes tested in sulfate alkaline solutions from Part 111  were also compared.

Corrosion Measurements

The corrosion potential (Ecorr) was measured by using a CSE electrode via access points along the length of the tendon. Also, corrosion current densities (icorr) of embedded steel probes were measured by the linear polarization resistance (LPR) method following the Equation (1) presented below,

formula

where the Stern-Geary Coefficient (B) was assumed to be 26 mV for active corrosion conditions. Rp is the resolved polarization resistance in Ω compensating for the Rs, and A is the area of the probe (∼2.2 cm2) exposed to the grout.

Anodic Polarization Tests

After ∼500 d of open-circuit potential and LPR measurements, anodic polarization tests were conducted by using the three-electrode arrangement explained earlier. For this test, a saturated calomel electrode (SCE) was used as a reference electrode. The working electrode was cathodically conditioned to −1 VSCE for 30 min prior to being subjected to polarization from −1 VSCE to 500 mVSCE at a scan rate of 0.05 mV/s. The anodic polarization tests were intended to evaluate the anodic behavior of steel in deficient grout with enhanced sulfate ion levels and determine if more adverse corrosion conditions exist in deficient grout. The data were compared with test results conducted in Part 111  in order to identify discrepant corrosion behavior between experiments in solution and in grout environments. Measurements were performed in duplicate samples.

Grout Characteristics

Subjectively, the visual appearance and texture of the grout seen through the clear PVC was not greatly different at the top of tendon than compared to the bottom low-elevation sections immediately after casting. The tendon samples appeared to be fully grouted immediately after casting, but void spaces subsequently developed by the next day, typically at the top valve, indicating some level of grout segregation and volume change.

After completion of corrosion testing, autopsy of the tendons showed that grout segregation occurred at the high point of the tendons for all test conditions, although it never fully replicated the most severe physical grout deficiencies observed in the field. Examples of grout segregation are represented in Figure 2. The grout segregation was thought to be related to the displacement of excess water content after mixing. Indeed, the moisture content of the grout in the higher tendon elevation was as much as 60% to 70% in Cases 1 and 3, as can be seen in Figure 3. The moisture content in the lower tendon elevations was lower (typically less than 20%). Of note, when external additions of sodium sulfate were incorporated, the measured moisture content in the higher tendon elevation was sometimes only marginally higher than that observed in the hardened grout in the lower elevations. This was thought to be related to the increase in the solid to water content ratio as a result of the additions of sodium sulfate and where the amount of available moisture for cement hydration was lower.

FIGURE 2.

Photos of deficient grout at higher ([a] through [c]) and lower ([d] through [f]) elevations of tendons.

FIGURE 2.

Photos of deficient grout at higher ([a] through [c]) and lower ([d] through [f]) elevations of tendons.

Close modal
FIGURE 3.

Moisture content of the deficient grout at higher and lower elevation of the tendons.

FIGURE 3.

Moisture content of the deficient grout at higher and lower elevation of the tendons.

Close modal

Results from thin-film petrography showed that the deficient grout from the higher elevations consisted of silica fume admixture, some carbonate flour and scant amount of hydrated Portland cement (Figure 4). Portions of the segregated grout showed darker silica fume-rich paste, soft porous portlandite and fine carbonate-rich paste. The water-to-cement (w/c) ratios were estimated to be in the range of 0.40 to 0.70. The hardened grout section consisted of Portland cement and less abundance of silica fume. The hardened grout material compared well to fully hydrated Portland cement clinker (Figure 4). The grout was more uniform and consistent in composition and had w/c ratios estimated at 0.40 to 0.45. Similar conditions were observed for the field samples (Figure 4). In those samples, polished section of segregated grout showed abundant shrinkage microcracking. The segregated material was very soft (Mohs < 2). The bottom section of the field sample showed a gradation of darker- to lighter-colored paste with presence of entrained-sized air voids that were partially filled with ettringite and portlandite.

FIGURE 4.

Micrographs of deficient grout. ([a] and [b]) Segregated grout from the base case. ([c] and [d]) Hardened grout from the base case. ([e] and [f]) Segregated grout (left) and hardened grout (right) from field samples.

FIGURE 4.

Micrographs of deficient grout. ([a] and [b]) Segregated grout from the base case. ([c] and [d]) Hardened grout from the base case. ([e] and [f]) Segregated grout (left) and hardened grout (right) from field samples.

Close modal

The Rs measurements of the segregated grout are depicted in Figure 5. In general agreement with the measured moisture content, the Rs is lower for the grout in the higher elevations than for the grout from lower elevations, indicating a more permeable grout at higher elevations associated with grout deficiency. It is highlighted that Rs did not increase with time. It was thought that the grout at the lower elevations would be of lower permeability and Rs would increase as the cement component of the grout continues to hydrate. For the grout in the higher elevations, the Rs decreased with time, which may be indicative of further moisture and ion accumulation at higher elevations with time. A lower Rs in the deficient grout would promote adverse galvanic interaction and enhance macrocell corrosion development. Rs was not well correlated to the external admixed sulfate ion content levels in the higher or lower elevations of the tendons. It was evident that moisture availability is important in the transport of sulfate ions. Indeed, in Case 4, precipitation of sodium sulfate crystals was apparent as the moisture level throughout the tendon was relatively low (even in the upper elevation locations that contained some level of deficient grout).

FIGURE 5.

Grout resistance with time for samples collected at higher (full symbols) and lower (empty symbols) elevation of the tendons.

FIGURE 5.

Grout resistance with time for samples collected at higher (full symbols) and lower (empty symbols) elevation of the tendons.

Close modal

The sulfate ion content for all test cases is shown in Figure 6. Higher sulfate ion concentrations were found at the upper elevation than at the lower elevations of the tendons (Figure 6). Case 2 showed slight sulfate ion increment (<0.1%); Case 3 showed significant sulfate ion increment (>1%). As expected, Case 4 (where high levels of sodium sulfate were pre-mixed) indeed showed high levels of sulfates (<1%) throughout the tendon. Importantly, the sulfate level in the segregated grout in the non-doped base condition (Case 1) showed high sulfate levels (>1%), verifying that vestigial sulfate ion content in the grout can have significant accumulation as a result of grout segregation. The accumulated sulfate ion levels in the lab tests were comparable to the elevated levels observed in the Florida bridges.

FIGURE 6.

Sulfate ion content determined in the deficient grout at higher and lower elevation of the tendons. Arrows indicate corrosion development.

FIGURE 6.

Sulfate ion content determined in the deficient grout at higher and lower elevation of the tendons. Arrows indicate corrosion development.

Close modal

Sulfate ion presence has been emphasized in much of the work; however, other ionic species were also tested for in grout collected from the base condition (Case 1): calcium, potassium, sodium, nitrate, nitrite, phosphorus, sulfate, sulfide, and chloride (Figure 7). For comparison purposes, sulfate concentrations for Cases 2 through 4 were also included in the figure. Grout samples were collected along the length of the test tendon from near the middle of the pipe, where the grout was hardened, to near to high-elevation pipe cap where grout segregation was prominent. Similar to findings in field samples, elevation of potassium, sodium, sulfate, and chloride was observed in the higher portion of the test sample, where grout segregation was prominent. The chloride ion content, although elevated, remained low. The role of low-level vestigial chloride ion content on corrosion in alkaline sulfate solution is described in later publications. As described by others, the enhanced alkali content may be important in the development of high pore water pH.10 

FIGURE 7.

Chemical characteristic profiles along the tendon for the base case 1. Sulfate ion levels for test cases 2 through 4 are presented for comparison.

FIGURE 7.

Chemical characteristic profiles along the tendon for the base case 1. Sulfate ion levels for test cases 2 through 4 are presented for comparison.

Close modal

The results of pH measurements from grout leachate are depicted in Figure 8. Also, pH measurements on extracted grout fragments using a pH color indicator were performed. The pH indicator showed a purple color indicative of pH 12 for all test cases. The pH obtained from the leachate typically had values between 12 and 13 for all test cases except for the segregated grout from Case 1 where the pH was low (pH < 11). Both pH measurements from grout leachate and by color indicator were taken at different times. pH measurements using a color indicator were done immediately after opening the tendons, exposing the grout to the environment. For pH measures from the leachate, grout samples collected from the tendon were taken to the lab in a sealed bag until the beginning of the analysis. Hence, as the pH indicator was applied immediately after opening the tendon and exposing the grout to the environment, the somewhat lower pH of the leachate was thought to be a testing anomaly.

FIGURE 8.

pH measurements for the leachate obtained from grout samples collected at higher and lower elevation of the tendons.

FIGURE 8.

pH measurements for the leachate obtained from grout samples collected at higher and lower elevation of the tendons.

Close modal

Electrochemical Testing

Figure 9 shows the Ecorr and icorr trends for the corrosion probes with time after casting. More active potentials and greater icorr for probes located in the upper ∼0.3 m region of the tendons than compared to probes at lower elevations were measured regardless of the level of added external sulfate ions. The Ecorr for the corrosion probes embedded in the upper tendon elevation were generally more negative than −300 mVCSE. Nominal icorr values were 0.02 µA/cm2 to 0.4 µA/cm2. One of the corroded steel probes embedded in the top portion of a tendon, initially cast with 0.9% sulfate ions by cement mass (Case 3), is shown in Figure 10. Similar Ecorr and icorr data were obtained for steel sensors with crevice conditions (Figure 9).

FIGURE 9.

Corrosion potential and corrosion current density of steel probes embedded in the deficient grout at higher (full symbol) and lower (empty symbol) elevation of the tendons. ([a] and [b]) Without crevice condition. ([c] and [d]) With crevice condition.

FIGURE 9.

Corrosion potential and corrosion current density of steel probes embedded in the deficient grout at higher (full symbol) and lower (empty symbol) elevation of the tendons. ([a] and [b]) Without crevice condition. ([c] and [d]) With crevice condition.

Close modal
FIGURE 10.

Images of steel probes after corrosion testing at higher (1) and lower (5) elevation of the tendons. Left side (from the arrow) of probes were embedded in the deficient grout.

FIGURE 10.

Images of steel probes after corrosion testing at higher (1) and lower (5) elevation of the tendons. Left side (from the arrow) of probes were embedded in the deficient grout.

Close modal

High moisture and sulfate content in alkaline conditions (as in severe deficient grout) were deemed to provide adverse corrosion conditions where active corrosion potentials and current densities can develop. Steel in hardened grout is expected to maintain passive-like conditions. If electrodes were to be coupled in such dissimilar materials as is expected for tendons with grout segregation, adverse macrocell coupling can occur. Large macrocell current that can lead to early corrosion failure would occur if the steel in the deficient grout at upper elevations show Tafel-like anodic behavior and efficient coupling to extended cathodes with readily available oxygen. Generally, testing in solution and MIT showed that enhanced steel corrosion activity can form in conditions representative of deficient grout. In solution, anodic polarization curves for active corrosion had Tafel slopes of the order of 100 mV/decade. As for the MIT testing, even though the distinctly enhanced corrosion current densities appeared to be marginally high (generally of the order of 0.1 µA/cm2), adverse galvanic coupling could enhance those rates. Preliminary testing incorporating an approximate 1:1 coupling of tendon sections from a failed field tendon with deficient and hardened grout showed that the coupled mixed potential can provide as high as 50 mV to 100 mV anodic polarization resulting in an ∼20% to 50% increase in corrosion current.4  Similar levels of polarization in aqueous environments may result in a 5 to 10 times increase in corrosion rate.11  The field observations in the Florida bridges showed highly localized development of severe deficient grout that formed local anodes, thus likely creating significantly higher cathode-to-anode ratios and thus greater macrocell current.19  The low grout resistivity and oxygen availability through unsealed vents would support cathodic reactions and galvanic coupling. As conjecture, if the macrocell enhanced corrosion current density in the grout were to be as high as of the order of 1 µA/cm2, by Faradaic conversion, it would take ∼50 y to cause total corrosion loss of 15% of 22 7-wire strands in a 1 foot length. If deficient grout conditions are better represented by the electrochemical behavior in alkaline sulfate solution (where non-macrocell enhanced current density were measured of the order of 1 µA/cm2), then a macrocell enhanced current density of 10 µA/cm2 would result in ∼5 y to cause that level of metal loss. This amount of metal loss was considered to be a ballpark figure of the severe corrosion in the Florida bridge where tendon corrosion failure occurred within 8 y.

As shown in Figure 9, the difference in Ecorr for the steel probes located in the upper and lower elevation of the MIT samples was as much as 100 mV. It is evident that some level of polarization would occur if electrodes were coupled. If there is strong passive to active transition and pitting events (as described in Part 111 ), enhanced corrosion degradation would occur. The anodic polarization curves of the steel probes in the MIT tendons were generally indicative of passive-like conditions (up to 100 mVSCE) for all test cases, as depicted in Figure 11. It was not discerned why the anodic polarization behavior of steel in alkaline sulfate solution was apparently much more aggressive than in grout even though high moisture and sulfate content in alkaline conditions were generally observed in the latter. For Case 4, with the highest level of externally admixed sulfate, a feature of the polarization graph showed an abrupt increase in anodic current at ∼100 mVSCE for steel in the upper portions of the tendon. Transpassive behavior for the given pH ∼ 12 was not expected and that behavior was not observed for any other test cases. If conditions were able to cause significant anodic polarization near this range for breakdown of passive-like behavior, then larger anodic currents may result. Nevertheless, even though the expected Tafel-like behavior was not well manifested in grout environments, the visual observations of surface rusting of steel in the deficient grout within the 2 y of the testing are indicative of adverse corrosion conditions. In part because of the variability of grout deficiency even within the top 1 ft (0.3 m) of the tendon, the severity of rust formation differed between the steel probes and the steel strand. The heavier corrosion of the steel strand may also be related to adverse galvanic coupling in the dissimilar segregated grout.

FIGURE 11.

Anodic polarization of steel probes embedded in the deficient grout at (a) higher and (b) lower elevation of the tendons.

FIGURE 11.

Anodic polarization of steel probes embedded in the deficient grout at (a) higher and (b) lower elevation of the tendons.

Close modal

Characterization of Steel Corrosion Products

Figure 12 shows the visual appearance of the embedded 7-wire strand placed along the length of each tendon (at higher elevations) ∼2 y after casting. Significant corrosion was observed on the section of the strand in contact with segregated grout at the upper elevations of the tendons in Cases 1, 3, and 4. The same strand in each sample did not exhibit corrosion development when embedded in the hardened grout at lower elevations. Surface rusting to moderate corrosion on the corrosion probes was observed on sensors embedded in the upper elevations of the tendon where grout segregation was prevalent (Figure 10). Steel probes in the lower elevations of the tendons were generally rust free. For steel probes simulating crevice conditions, localized discoloration under where the rubber gasket was placed was evident for all samples and at all locations.

FIGURE 12.

Photos of embedded strands in MIT samples at higher elevation.

FIGURE 12.

Photos of embedded strands in MIT samples at higher elevation.

Close modal

XRD data of the corrosion products of the embedded strand (Cases 1 and 3) are visualized in Figure 13. For comparison purposes, XRD data of corrosion products from a strand from a Florida bridge and from the steel electrodes tested in sulfate alkaline solutions (Part 1 paper11 ) are also depicted (Figure 13). Results did not show strongly-defined diffraction peaks, and the relative intensity of the peaks did not scale significantly above background noise. The rust product was identified as goethite. Calcium carbonate, presumably from grout residue, was detected in the grout test samples. There were indications that sulfur-bearing iron corrosion products were present but their presence did not appear to be very strong.

FIGURE 13.

XRD diffractograms of corrosion products.

FIGURE 13.

XRD diffractograms of corrosion products.

Close modal
  • Findings verified that the physical grout deficiencies that form in MIT testing have adverse chemical characteristics. The most severe physical and chemical grout segregation was localized to the tendon high points because of the accumulation of moisture and transport of ionic species. The sulfate level was high in the segregated grout regardless of the level of external sulfate additions. This observation for the base case verified that that vestigial sulfate content in the grout can have significant accumulation as a result of the grout segregation.

  • Distinctly enhanced but marginal corrosion current densities occurred in the deficient grout created with expired grout and excessive mix water at elevated portions of the tendon when compared to hardened grout at lower elevations. The difference in Ecorr for the steel probes located in deficient and hardened grout was as much as 100 mV, which would allow some level of polarization for coupled electrodes. Anodic polarization in deficient grout materials formed in MIT did not exhibit strong passive to active transitions. If the severity of grout segregation is advanced where corrosion kinetics in alkaline sulfate solutions (Part 1) is more representative, then the macrocell coupling can lead to the adverse galvanic coupling promoting the accelerated corrosion rates observed in the Florida bridges.

This investigation was supported by the Florida Department of Transportation (FDOT). The opinions, findings, and conclusions expressed here are those of the authors and not necessarily those of the FDOT or the U.S. Department of Transportation. Support from the FDOT State Materials Office is acknowledged here. The contributions by Md. Ahsan Sabbir and Roberto Rodriguez are acknowledged here, as is the support and assistance from the FDOT State Materials Office and H.R. Hamilton and Mario Paredes.

Trade name.

1.
K.
Lau
,
I.
Lasa
,
M.
Paredes
,
“Corrosion Failure of Post-Tensioned Tendons in Presence of Deficient Grout,”
CORROSION 2013
,
paper no. 2600
(
Houston, TX
:
NACE International
,
2013
).
2.
K.
Lau
,
I.
Lasa
,
M.
Paredes
,
“Update on Corrosion of Post-Tensioned Tendons with Deficient Grout,”
CORROSION 2014
,
paper no. 4225
(
Houston, TX
:
NACE
,
2014
).
3.
K.
Lau
,
“Corrosion of Pre-Stress and Post-Tension Reinforced Concrete Bridges,”
in
Corrosion of Steel in Concrete Structures
,
ed.
A.
Poursaee
(
Cambridge, United Kingdom
:
Woodhead Publishing
,
2016
),
p
.
37
57
.
4.
K.
Lau
,
I.
Lasa
,
M.
Paredes
,
“Corrosion Development of PT Tendons with Deficient Grout: Corrosion Failure in Ringling Causeway Bridge,”
Florida Department of Transportation State Materials Office
,
Report
,
2011
.
5.
H.R.
Hamilton
,
A.
Piper
,
A.
Randell
,
B.
Brunner
,
“Simulation of Prepackaged Grout Bleed Under Field Conditions,”
Final Report to Florida Department of Transportation
,
Contract No. BDK75 977–59
,
2014
.
6.
L.
Taerwe
,
ed.
,
“Durability of Post-Tensioning Tendons,”
FIB Fédération internationale du béton, Proceedings of a Workshop held at Ghent University on 15–16 November 2001
,
technical report, bulletin 15, November
2001
.
7.
L.
Bertolini
,
M.
Carsana
,
“High pH Corrosion of Prestressing Steel in Segregated Grout,”
in
Modeling of Corroding Concrete Structures
,
eds.
C.
Andrade
,
G.
Mancini
,
RILEM Book Series 5
(
Paris, France
:
RILEM
,
2011
),
p
.
147
158
.
8.
M.
Carsana
,
L.
Bertolini
,
Struct. Infrastruct. Eng.
11
,
3
(
2015
):
p
.
402
.
9.
M.
Carsana
,
L.
Bertolini
,
J. Mater. Civil Eng.
28
,
6
(
2016
):
p
.
1
.
10.
C.L.
Page
,
O.
Vennesland
,
Matériaux et Constructions
16
,
1
(
1983
):
p
.
19
.
11.
K.K.
Krishna Vigneshwaran
,
S.
Permeh
,
M.
Echeverría
,
K.
Lau
,
I.
Lasa
,
Corrosion
74
,
2
(
2018
):
in press
.
12.
M.
O’Reilly
,
D.
Darwin
,
J.
Browning
,
“Corrosion Performance of Prestressing Strands in Contact with Dissimilar Grouts,”
University of Kansas Center for Research, Inc.
,
technical report SL Report 12-1
,
2012
.
13.
S.K.
Lee
,
J.
Zielske
,
“An FHWA Special Study: Post-Tensioning Tendon Grout Chloride Thresholds,”
U.S. Department of Transportation, Federal Highway Administration, report no. FHWA-HRT-14-039
,
May
2014
.
14.
K.K.
Krishna Vigneshwaran
,
S.
Permeh
,
I.
Lasa
,
K.
Lau
,
“Corrosion of Steel in Deficient Grout with Enhanced Sulfate Content,”
NACE Concrete Service Life Extension Conference 2015
,
paper no. CCC15-6942
(
Houston, TX
:
NACE
,
2015
).
15.
S.
Permeh
,
K.K.
Krishna Vigneshwaran
,
K.
Lau
,
I.
Lasa
,
M.
Paredes
,
“Material and Corrosion Evaluation of Deficient PT Grout with Enhanced Sulfate Concentrations,”
CORROSION 2013
,
paper no. 5828
(
Houston, TX
:
NACE
,
2015
).
16.
V.K.
Gouda
,
W.Y.
Halaka
,
Br. Corros. J.
5
,
5
(
1970
):
p
.
204
.
17.
A.J.
Al-Tayyib
,
M.S.
Khan
,
Cem. Concr. Compos.
13
(
1991
):
p
.
123
.
18.
ASTM C566-13
,
“Standard Test Method for Total Evaporable Moisture Content of Aggregate by Drying”
(
West Conshohocken, PA
:
ASTM International
,
2013
).
19.
W.H.
Hartt
,
S.-K.
Lee
,
Corrosion
72
,
8
(
2016
):
p
.
991
.