See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/364344055

Corrosion-induced damages and failures of posttensioned bridges: A

literature review

Article  in  Structural Concrete · October 2022

DOI: 10.1002/suco.202200297

CITATIONS READS

4 137

6 authors, including:

Antonia Menga Terje Kanstad

Norwegian University of Science and Technology Norwegian University of Science and Technology
6 PUBLICATIONS   6 CITATIONS    78 PUBLICATIONS   573 CITATIONS   

SEE PROFILE SEE PROFILE

Daniel Cantero Lise Bathen

Norwegian University of Science and Technology Norwegian Public Roads Administration
74 PUBLICATIONS   1,222 CITATIONS    2 PUBLICATIONS   4 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Evaluation and Improvement of Crack Width Calculation Methods for Large-Scale Concrete Structures View project

Special Session on Bridge Weigh-in-Motion at IALCCE2023 View project

All content following this page was uploaded by Antonia Menga on 16 November 2022.

The user has requested enhancement of the downloaded file.

 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Received: 31 March 2022 Revised: 11 June 2022 Accepted: 27 August 2022
DOI: 10.1002/suco.202200297

ARTICLE

Corrosion-induced damages and failures of posttensioned

bridges: A literature review

Antonia Menga 1 | Terje Kanstad 1 | Daniel Cantero 1 | Lise Bathen 2 |

Karla Hornbostel 2 | Anja Klausen 1

1
Department of Structural Engineering,
Norwegian University of Science and Abstract
Technology (NTNU), Trondheim, Norway Corrosion of posttensioned bridges raises great concern, since the only way to
2
Norwegian Public Road Administration safely assess the condition of the tendons is through destructive evaluation. In
(NPRA), Oslo, Norway
this scenario, this paper presents the results of an extensive literature review
Correspondence on reported cases of corrosion of posttensioned bridges and few prestressed
Antonia Menga, Department of Structural structures. The aim is to increase knowledge about the structural consequences
Engineering, NTNU, Materialteknisk,
3-42, Gløshaugen, Richard Birkelands vei
of possible corrosion-induced failure mechanisms. The cases were rated
1a, 7034, Trondheim, Norway. according to the increasing severity of the observed damage and consequent
Email: [email protected] interventions as follows: ordinary maintenance is effective, extraordinary
Funding information maintenance is necessary, and maintenance is insufficient or absent. It was
NPRA found that most corrosion-sensitive structures were segmental and box
section beam bridges. Most cases required extraordinary maintenance. Warn-
ing signs were often observed. The damage was mainly ascribed to design and
execution mistakes that facilitated the ingress of external chlorides causing
corrosion. The paper concludes with a discussion of the consequences for
structural safety.

KEYWORDS
chlorides, corrosion, deficient grout, failure, literature survey, post-tensioned bridges,
warning signs

1 | INTRODUCTION lives and traffic disruption problems. Numerous investi-

gations have been conducted to evaluate the health of
When a bridge failure occurs, the loss of the structure bridges and the causes of their failure in various
can result in much greater economic consequences than countries.
the value of the asset itself,1 for example, loss of human In the period 1977–1981, a study on 150 damaged and
collapsed structures around the world was carried out by
Hadipriono.2 More than one-third of the analyzed struc-
Discussion on this paper must be submitted within two months of the
print publication. The discussion will then be published in print, along
tures were bridges, the degradation of which was mostly
with the authors’ closure, if any, approximately nine months after the attributable to external events (like lateral impact forces,
print publication. unexpected live loads) and construction deficiencies

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2022 The Authors. Structural Concrete published by John Wiley & Sons Ltd on behalf of International Federation for Structural Concrete.

Structural Concrete. 2022;1–16. wileyonlinelibrary.com/journal/suco 1

 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 MENGA ET AL.

(e.g., falsework and concreting faults, lack of knowledge construction deficiencies, corrosion) and are strongly
in long-term creep and shrinkage effects on concrete). influenced by the characteristics of the bridge (like its
Forty point eight percentage of the bridges present in location, environment, age, structural typology). Conse-
the United States in 1988 were rated substandard by the quently, each country tends to orient research according
Federal Highway Administration (FHWA).3 One hun- to the characteristics and conditions relevant to the
dred and fourteen bridges failed in the period 1951– bridges in their territory.
1988,3 and over 500 bridges failed between 1989 and In this context, in 2017 the Norwegian Public Road
2000.4 The outcome of the investigations was that the Administration (NPRA) promoted the “Better Bridge
most frequent reasons of failure were not due to design Maintenance’ research and development program. The
and construction fault, but due to accidents (e.g., cars or aim was to reduce the deterioration of national bridges
trucks colliding with the structure) and natural catastro- by finding ways of assessing the state of the structures
phes like scour, earthquake, flood.3,4 and identifying the most favorable maintenance
In the early 1990s, the American Segmental Bridge methods.
Institute (ASBI) promoted a durability survey in the According to the Bridge Management System of the
United States and Canada. The inspection included NPRA, concrete bridges constitute the majority of the
96 bridges with completion dates ranging from 1966 to 18,199 bridges on national and county roads in Norway,
1993, and highlighted that segmental constructions per- and a significant number of them were built with pre-
formed well over time.5 stressing tendons. Many of the bridges are in or near
Another investigation was conducted by the FHWA coastal regions with varying exposure to sea water, while
in 2001, reporting that nearly 30% of bridges in the nearly all of them are subject to de-icing salts during the
United States were rated as deficient.6 Moreover, since winter. In this chloride-contaminated environment, the
2008 the FHWA has been supporting a research project risk of corrosion increases, depending on age, exposure,
at the Multidisciplinary Center for Earthquake Engineer- and detailing.10
ing Research (MCEER) with the aim of establishing rea- Corrosion (and in general damage) of bridges is very
sonable bridge damage/failure models.1 concerning since, in an increasingly connected world, the
Between 1979 and 1992 the UK Standing Committee cost of damaged infrastructures represents a significant
on Structural Safety published annual reports, which part of the gross domestic product (GDP) of a country. It
included reports on suspected deficiencies of grouting of is extremely difficult to estimate the indirect cost of cor-
posttensioning tendons. Then, the Highway Agency in rosion (e.g., cost of loss of productivity, compensation for
1994 started a significant systematic series of inspections causalities and environmental pollution). However, the
of their bridges. These activities gradually provided evi- direct economic cost (i.e., cost of the application, opera-
dence of a growing problem with some posttensioned tion, and maintenance of anti-corrosion technologies) is
structures.7 deemed calculable.11 The global cost of corrosion was
Between 1995 and 1998, a study on the durability of first estimated to be 2.5 trillion US dollars (3.4% of GDP)
prestressed concrete bridges was carried out in in 2013 by the National Association of Corrosion Engi-
Switzerland.8 Some cases of significant corrosion dam- neers (NACE International),12 but it was also estimated
ages were identified. However, the study showed that that “savings of between 15 and 35% of the cost of corro-
prestressed concrete bridges generally behave very satis- sion could be realized”12 by using available corrosion
factorily and fulfill the durability requirements, especially control practices.
if they are properly maintained. Corrosion is an electrochemical reaction where the
In Germany, the main part of the bridges was built steel in direct contact with oxygen (present in the air
between the 1960s and the 1980s, when the current or dissolved in water) tends to return to its original
knowledge about the durability of structures was not form (ore) forming iron oxides on the surface (rust).13
available. Therefore, despite the design lifespan of more The volume of rust is from two to six times greater than
than 70 years, bridges showed major damage after a steel,14 and this causes the development of large tensile
period of 30–40 years.9 stresses in the surrounding concrete area resulting in
The results of the investigations summarized above cracking and spalling of the concrete, which can influ-
show that bridges in general tend to perform well over ence various characteristics such as the mechanical
time, that is, if accurate durability-based design and performance and load capacity of the concrete struc-
appropriate maintenance are conducted, bridges tend tures.15 In specific, corrosion induces the loss of bar
to meet their expected service life. Nevertheless, many section, loss of concrete section (due to cracking and
failures of bridges are also reported, and their causes spalling), and a reduction of the bond between rein-
vary (accidents, natural catastrophes, external events, forcement and concrete.16
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
MENGA ET AL. 3

F I G U R E 1 Schematic representation of the pretensioning and posttensioning methods.20 In pretensioning (left side), the tension is
applied to the tendons before the casting of the concrete. The prestress is applied to the concrete element through bond between the
concrete and the tendons. In posttensioning (right side), the tension is applied to the tendons after the casting of the concrete. The tendons
are enclosed in ducts, generally filled with grout after the application of the prestress. Hence, the prestress is applied to the concrete element
by means of wedge actions occurring at the anchorages of the tendons, and not by bond

Corrosion can appear with two different morphol- Corrosion has more severe effects in prestressed struc-
ogies: uniform and localized (pitting) corrosion. Uniform tures, where the combination of the prestress with the
corrosion produces a material loss distributed uniformly cross-sectional loss of the strands due to corrosion could
over the entire surface of steel exposed to the corrosive enhance the risk of brittle failure.19
environment.17 It is caused by the carbonation of con- The two prestressing methods (pretensioning and
crete. Pitting corrosion produces a located material loss posttensioning, Figure 1) used worldwide can lead to dif-
on the surface of steel creating small cavities (pits) that ferent strand conditions. In the case of pretensioned brid-
leads to an increase in stress concentration.13 It is caused ges, the strands are placed in concrete without ducts. The
by chlorides partaking in the electrochemical reaction.18 prestress is hence transmitted through bond between the
The concrete environment initially protects the steel rein- strands and the concrete. In contrast, strands in postten-
forcement acting as a physical barrier with the environ- sioned bridges are located inside ducts. The prestress is
ment because of the high alkalinity of concrete hence transmitted by means of wedge actions occurring
(pH  12.5), which creates a protective thin film at the at the anchorages in the end zones of the concrete
surface of the steel bar.13 Carbonation of concrete pro- element.
duces a decrease of pH in the concrete. The drop in pH at The presence of ducts affect the reliability and perfor-
the surface of the steel bar causes the disruption of the mance of nondestructive evaluation (NDE) methods
protective film and the start of the electrochemical pro- developed for ordinary reinforced concrete.21 In fact,
cesses of corrosion. The corrosion reaction is accelerated NDE can be used to detect corrosion of strands only if
if chloride ions are present. The chlorides react with the they are enclosed in a nonmetallic duct. Even if NDE
protective film, dissolving it locally.18 Then, the dissolved methods can also be used to detect voids in the grout
iron accumulated in the pit reacts with chloride and (where corrosion usually occurs) regardless of the duct
water producing hydrochloric acid, which further accel- type, the presence of voids does not always imply the
erates the reaction. presence of corrosion. This means that the only way to
Pitting attack is recognized as the most severe type of safely assess the presence and degree of corrosion in post-
steel corrosion for two reasons. First, the localized nature tensioning strands is through destructive testing
of attack means that substantial section loss may occur (e.g., local removal of concrete and duct), which could
prior to warning signs.16 Second, pitting corrosion affects compromise the capacity of the bridge. As a result, sur-
not only strength, but also ductility substantially reduc- veys of posttensioned bridges are often limited to visual
ing it.19 This is worrisome since the current design and inspections. However, corrosion damage in posttensioned
assessment rules are based on an assurance of adequate elements has been found even in situations where no
ductility that may no longer be applicable under pitting external indications of the problem were apparent,22
corrosion, and could thus be unsafe.16 undermining the reliability of visual inspections.
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 MENGA ET AL.

T A B L E 1 Template table of a case study description, including the description of the structure and the characterization of the failure:
example for case ID-723

ID-7 Mid-Bay bridge. 1992–1993 (built) – beginning of 2000 (inspection)

Location: Destin, Florida, USA Reference: Venugopalan and Powers (2003)

Description Failure characterization Notes
• 141 span structure; two-lane undivided • Failure of individual wires was primarily • Beginning of 2000: replacing of 11
highway with shoulders on both sides. due to corrosion and subsequent elevated tendons after disclosing of a box girder.
• Segmental precast concrete box girders tensile stress that resulted from the • Tendon corrosion associated with grout
held together by six posttensioning reduced cross-section. voids and bleed water is more likely to
tendons (three on each side). The • Visual inspection of the tendon ends occur along the inclines rather than
tendons span eight to nine segments and highlighted the presence of acid horizontal runs.
terminate at a metal trumpet type corrosion products, meaning active First, since the latter location are at a
anchorage assembly. corrosion was ongoing. lower elevation and as such exhibit
• Each tendon is comprised of 19 spirally • Cracking of the polyethylene ducts is fewer void.
wound 5/8 in. diameter seven-wire extensive, but it has not been a primary Second, because the incline from which
strands with a grouted 4 in. diameter cause of tendon corrosion problems. mortar was not pumped should be
polyethylene duct. However, if left unrepaired, the cracks particularly likely to contain both voids
will contribute in the long-term to and bleed water.
further tendon corrosion.

Since visual inspections convey unreliable informa- possible failure mechanisms induced by corrosion in the
tion about corrosion of posttensioned bridges, experience posttensioned system. The aim was to increase knowl-
from past failures or damage reports may help in per- edge about the structural consequences of these failures
forming a better evaluation based on the type of struc- beyond what can be acquired through merely visual
ture, exposure, etc.1,6 Unfortunately, it is not possible to inspections. For this purpose, the following problems
obtain precise numbers on the incidence of serious corro- were addressed:
sion damage in prestressed concrete structures. This is
mainly because many cases are not reported and some a. Is there any way to improve visual inspections to
occurrences of corrosion have not yet been detected.22 assess the development of corrosion in posttensioned
Moreover, past surveys carried out in various countries structures? In other words, can the presence of warn-
have mainly studied the overall health of the bridges. ing signs be used to anticipate corrosion-induced
Even when the surveys focused on corrosion-induced failures?
damage, research was limited in space (i.e., only bridges b. How does location (climate conditions, proximity to
in a determined area were studied) or to a specific topic the sea, or exposure to pollutants) affect the failure?
(e.g., grouting conditions). c. For the different types of posttensioned structures:
This being the case, the lack of knowledge about  To what corrosion-induced problem are they most
corrosion-induced failures and serious damage in post- sensitive?
tensioning tendons is alarming, mainly because it is  How severe is generally the damage when it is
almost impossible to anticipate these kinds of failures. detected?
The main aim of this paper is therefore to present the
results of an extensive literature review about corrosion-
induced failures of posttensioned bridges conducted by 2 | METHODOLOG Y
the authors.23 This paper is meant to be a more approach-
able and concise summary of the information and com- To answer the above questions, a literature survey of
ments reported in reference 23, with the aim of reaching corrosion-induced damages and failures in prestressed
more people. Therefore, the details about the structures structures was carried out.23 The survey consisted in the
and the analyses here presented can be found in refer- analysis of all the publicly available documents on the
ence 23, which includes three appendices containing topic to the authors' knowledge (papers and case-studies).
summary tables of the data and pictures collected for A total of 52 cases where corrosion was the only cause or
each structure in the literature review. one of the main causes producing the damage or failure,
A number of representative posttensioned structures were reported in the reviewed literature and all of them
(especially bridges) were investigated considering were included in this study. They mostly involved
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
MENGA ET AL. 5

T A B L E 2 Template of summary table containing the most important details for a case study (i.e., structure and tendons description,
presence of warning signs, corrosion causes and products, failure mechanism and damage description): example for case ID-723

ID-7 Mid-Bay bridge

Prestressing
Age at failure: system: External
7 years Location: Destin, Florida posttension Number of spans: 141
Structure Tendon and grout Warning signs: Corrosion causes Failure mechanism: Damage description:
description: description: Each Cracking of the and products: Failure of Tendon corrosion
Segmental tendon was polyethylene Grout voids, bleed individual wires mostly along the
precast box constituted by 19 ducts. water and soft primarily due to inclines,
girders held spirally wound chalky grout in corrosion and associated with
together by six 5/8 in. diameter the affected areas. subsequent grout voids and
posttensioning seven-wire strands A variety of factors elevated tensile bleed water.
tendons (three within a grouted led to the failures stresses that
on each side). 4 in. diameter of external resulted from the
polyethylene duct. tendons, reduced cross-
The center lengths including the section of the
of the tendons penetration of salt wires.
were draped water into the
downward from external tendons
the anchorage and the
through deviation preponderance of
blocks along the grout voids. Acid
length of the span. corrosion
products.

posttensioned bridges, but also comprised a few cases of to make graphical analyses of the case studies (see
pretensioned bridges and posttensioned structures such Section 3).
as slabs or cylindrical containers. This was done to make Finally, the cases were classified according to the
it possible to evaluate whether considerations made for increasing severity of the observed damage and conse-
structures potentially easier to assess can be applied to quent interventions. Not all the papers and case-studies
posttensioned bridges. The arguments can be listed as: consulted during the literature review provided specific
information about how the corrosion damage of the ten-
• Pretensioned bridges can be assessed through NDE dons was assessed (e.g., mass-loss, cross-sectional loss—
methods. for more information see Appendix A in reference 23).
• The assessment of simpler posttensioned structures Therefore, it was decided not to classify the damage
(i.e., when compared to bridges, with shorter design according to the degree of corrosion, but to include in the
working life, with smaller dimensions, and subject to lowest level all the cases in which, regardless of its sever-
smaller loads) such as slabs requires less effort than ity, corrosion did not cause the failure of the tendons, the
the assessment of bridges. collapse of the structure, or the need for its demolition.
From lowest to highest, the levels of damage were:
To manage and analyze the considerable amount of
data systematically, every case was given an identification • Maintenance. Moderate damage (concrete cracking,
number (denoted by the symbol ID-xx). Each case was concrete cover spalling, loss of tendon cross-section)
described following the same template table (see Appen- that needs to be repaired with ordinary maintenance
dix A in reference 23 for all the structures and Table 1 for activities.
an example), where characteristics about age, location, • Tendon failure. The damage included breakage of the
structure type, failure mechanism, and failure/damage tendons, potentially affecting the overall safety of the
causes were reported together with additional notes. The structure. Extraordinary operations like tendon substi-
information gathered for each case was then summarized tution were necessary.
in another table (example shown in Table 2, complete • Demolition. Serious damage affecting the overall safety
data in Appendix B of reference 23) together with rele- of the structure and leading to demolition (extraordi-
vant keywords for every entry. The keywords were used nary intervention).
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 MENGA ET AL.

F I G U R E 2 Illustration of typical
relationships between how early the
damage was detected, the typology and
suitability of the interventions carried
out, and the structural consequences in
terms of increasing severity

• Collapse. Condition of total or partial collapse due to

inadequate or absent interventions, often resulting from
absence of warning signs (i.e., no detection of damage).

The adopted classification encapsulates several

aspects of the problem into four categories characterized
by increasing severity of the observed damage. The
names chosen for the categories are not on the same level
of evaluation (i.e., “maintenance” and “demolition” are
consequences of the damage, while “tendon failure” and
“collapse” are different types of damage). Nonetheless,
the names have been chosen for their simplicity (useful FIGURE 3 Geographical location of the case studies.
in the graphical representations of the analyses) and
because they are evocative of the level of damage (e.g., a
damage associated with the word “maintenance” seems
less severe than one associated to “tendon failure”). This
system inherently also includes considerations regarding
when (and if) the damage was detected and what sort of
action it required. Damage could have been detected dur-
ing planned inspections (“early”), when the damage had
severely corrupted parts of the posttension system
(“late”), when the damage was so severe that no interven-
tion could save the structure (“too late”), or the damage
was not detected in time (“not on time”). Correspond-
ingly, each damage detection time led to intervention
that could be classified as “ordinary,” “extensive,” FIGURE 4 Types of studied structures
“extraordinary,” and “inadequate or absent.” Figure 2
provides an overview of how the adopted classification
relates when the damage was detected, what type of structures were classified according to their typology.
intervention was required, and the structural conse- Then, the damage was characterized following the
quence (i.e., level of damage). scheme introduced in Figure 2. Finally, the failure and
damage causes were identified and investigated.

3 | R E SUL T S
3.1 | Structure description
This section reports the graphical representations of the
data analyzed in this work and pertaining to the 52 case The selected structures were located in various countries
studies identified during the literature review. First, the (Figure 3), often close to the sea or to heavily polluted
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
MENGA ET AL. 7

F I G U R E 5 Schematic representation of a segmentally mounted bridge24 (left side) and of a beam bridge25 (right side). A segmental
bridge is made of short transversal segments of the deck, assembled and then kept in place by posttensioned tendons. In a beam bridge, the
deck is made by beams supported by an abutment or pillar at each end. The beams can be pretensioned or posttensioned

F I G U R E 6 Polcevera Bridge, Ligury, Italy.26 The peculiarity of this bridge consists in the balanced system (in green). The system
comprises: the pylons from which two A-shaped structures stand (the tower); the main deck, supported at four locations by the pylons, and
with no connection to the tower; four transverse link girders connecting stays and pier trusses to the deck; four concrete cable stays covering
two sets of steel cables. The balanced systems are connected by two simply supported Gerber beam spans (in red)

and cast in situ) and box section beam bridges (Figure 5).
Some “Innovative structures” were also included in
Figure 4. They are structures with a unique design, which
has rarely or never been adopted elsewhere
(e.g., Polcevera Bridge, Figure 6, and Carpineto Viaduct
in Italy, in Ligury and Basilicata, respectively).
The selected structures were also classified according
to the type of prestressing system. There were:

• 44 posttensioned structures, of which:

 5 internally posttensioned structures different from
bridges;
 18 internally posttensioned bridges;
 21 externally posttensioned bridges.
F I G U R E 7 Types of studied structures divided by type of • 3 pretensioned beam bridges;
prestressing system • 1 bridge with some pretensioned spans and some post-
tensioned spans (F. G. Gardiner Expressway, Toronto,
environment (more details in Appendix A of refer- Canada);
ence 23). • 3 cable-stayed bridges;
The analyses showed that most corrosion-sensitive • 1 suspension bridge (Williamsburg Bridge, Williams-
structures (Figure 4) were segmentally mounted (precast burg, Virginia, USA).
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 MENGA ET AL.

F I G U R E 8 Type of ducts in the studied structures.

Pretensioned structures are not included in the figure
F I G U R E 1 0 Type of ducts in the studied structures divided by
type of prestressing system. Pretensioned structures are not
included in the figure

F I G U R E 9 Type of duct filling in the studied structures.

Pretensioned structures are not included in the figure

For each structure, only the most important category

was assigned. For example, Polcevera Bridge was a cable-
F I G U R E 1 1 Type of duct filling in the studied structures
stayed bridge with a posttensioned deck, hence it could divided by type of prestressing system. Pretensioned structures are
have been enlisted as a posttensioned or a cable-stayed not included in the figure.
bridge. However, since the cables were the ones to fail, it
was considered as a cable-stayed bridge. Only F. G. Gar- While in pretensioning systems the tendons are
diner Expressway presented failures both in pretensioned bonded directly to the concrete, in posttensioning sys-
and posttensioned spans, so both the categories “preten- tems, both internal and external, the tendons are
sion” and “posttension” were indicated, considering the enclosed in ducts generally filled with grout or grease.
bridge twice. This is why, in this paper considerations Figure 8 and Figure 9 report the types of duct and fill-
regarding the type of prestressing are always related to ing for the studied structures, respectively. These figures
53 structures instead of 52. do not include pretensioned structures, as they do not
Detailed information about the type of tendons con- have ducts. They also show that in most of the cases, the
stituting the prestressing systems was found in the litera- information included in the reviewed documents were
ture only for 25 out of 52 cases. Most of them consisted of insufficient to determine the type of duct, the type of fill-
spirally wound seven-wire strands with diameters rang- ing, or both. This explains the existence and the differ-
ing between 12 and 20 mm. In some cases, the strands ence in numbers in the entry “not identified” in the
were made with 12, 18, 19, or 27-twisted wires, while in figures. Nonetheless, Figure 8 shows that most of the
one single case the strands were made of 12 straight wires ducts were made of plastic, polyethylene to be precise.
each of 5 mm diameter. There were also a few cases in which the duct was made
After describing the types of structures and the types of precast posttensioned concrete blocks27 or cast in situ
of prestressing system of the studied cases, a relationship mortar boxes.24 Figure 9 shows that the majority of duct
between the two categories is shown in Figure 7. fillings in this study consisted of cementitious grout.
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
MENGA ET AL. 9

The type of ducts and duct fillings are related to the • most of the external tendons were enclosed in plastic
type of prestressing system in Figure 10 and Figure 11, ducts filled with cementitious grout;
respectively. Excluding the “not identified” cases, the fig- • internally posttensioned tendons were equally
ures show that: enclosed in metallic and plastic ducts filled with grout
or were not provided with ducts using alternative solu-
tions (e.g., inflatable ducts removed after grouting28).

3.2 | Damage characterization

The reported cases were classified according to the

increasing severity of the observed damage and conse-
quent interventions (Figure 2).
The damage was mostly detected late, yielding to
extensive and extraordinary operations like tendon sub-
stitution and demolition of the structure. Figure 12 high-
FIGURE 12 Level of damage of the studied structures lights that the number of cases in which the structure
could be brought back to safety was also significant. Still

F I G U R E 1 3 Level of damage in the studied structures divided F I G U R E 1 5 Age at damage detection of the studied structures
by type of prestressing system divided by level of damage

F I G U R E 1 4 Warning signs detected in a posttensioned bridge in the Midwest of the USA.29 Presence of efflorescence, delamination,
and spall observed on a posttensioned box girder (left side). Voids and rust at different sections of the tendon (right side)
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 MENGA ET AL.

F I G U R E 1 6 Damage location in the studied structures. The

locations have been collected in the macrocategories F I G U R E 1 8 Major damage and failure causes of the studied
“superstructure,” “tendons,” “miscellaneous,” and “not identified,” structures divided by level of damage
reported in yellow, blue, green, and brown, respectively. The
entries included in the macro-categories, are reported in lighter • maintenance measures should be implemented as
shades of the color corresponding to the appurtenant macro- planned.
category

To fulfill the previous conditions, it is necessary to

improve not only the frequency, but more importantly
the quality of the inspections. In fact, there have been
cases where a bridge collapsed even if no worrisome
damage was detected during the inspections
(e.g., Fossano Bridge29 collapsed after a positive inspec-
tion occurred few hours before the failure). The problem
is that, if no warning signs are reported, a bridge is not
usually included in the plans for accurate and detailed
evaluation. This is why in Figure 2 the first column is
named “time of detection” instead of, for example, “fre-
quency of inspections.” Nevertheless, in many cases fail-
ures such as broken tendons (five cases) or the collapse
FIGURE 17 Major damage and failure causes of the studied of the structure (three cases) occurred without warning
structures signs being detected. In some other cases (i.e., two cases
of collapsed structures and five cases of tendon failures)
more concerning were the cases in which the collapse of presence of warning signs were not reported in the litera-
the structure could not be foreseen. As shown in ture. This considerations highlight that the absence of
Figure 13, these cases mostly occurred in internally post- warning signs does not guarantee the safety of the
tensioned structures. While tendon failures were mostly structure.
detected in external posttension systems. When it comes to the age at damage detection of the
Warning signs such as small and large longitudinal structures, Figure 15 highlights that:
cracks, corrosion staining, concrete cover spalling, were
detected in half of the case studies (example of warning • tendon failures tended to occur during the first years
signs in Figure 14). For the other half, they were partly from construction;
not detected or no information about them was reported • collapses and demolitions occurred mainly 20 years
in the literature. The analyses conducted by the authors23 after construction;
indicate that detection of warning signs can help limit • damage needing ordinary maintenance was present
and reduce the damage, provided some conditions are during the entire life span of the structures.
fulfilled:
A discussion with proposed explanations about the
• the damage needs to be detected early enough for the age of the structures and the related damage is included
repair measures to be effective; in Section 4.
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
MENGA ET AL. 11

F I G U R E 1 9 Examples of execution and conceptual design mistakes found in Ynis-y-Gwas bridge.32 On the left side, void in the centre
of a longitudinal tendon, due to execution mistakes. Helical coil was used to space the wires apart in the image. On the right side, cardboard
tube across a transverse joint. This represents a conceptual design mistake because the joint (i.e., a region in which the tendons are not
protected by the concrete against the external environment) is protected by mere cardboard, which is not even waterproof

occur, and their frequency for the case studies. Contrary

to previous analyses (concerning, e.g., type of structures,
type of ducts, presence of warning signs), this figure
includes up to three entries for each structure. This is
because past investigations (e.g.,31) highlighted how it is
unlikely that damage, but especially failure can be caused
by one of the insufficiencies alone. In fact, failures are
usually the result of a combination of factors that arise
over time.
To better understand how the causes produced the
observed damage and failure, they are reported in
Figure 18, classified according to the level of damage.
F I G U R E 2 0 Major corrosion causes of the studied structures. This figure demonstrates the importance of accurate
The entry “external chlorides” (in green) is the sum of the sub- design and careful execution in preventing failure
entries “road salt,” “liquid sea water,” and “air-form sea water” (Figure 19). It also highlights that tendon failures were
(in a lighter shade of green). The sub-entries specify the origin of caused mostly by grouting problems. This information is
the external chlorides valuable because, as previously discussed (Figure 13),
tendon failure occurred in almost all the externally post-
tensioned bridges included in the survey. This means that
The locations where the damage associated to the fail- those bridges were extremely sensitive to corrosion due
ures was observed are listed in Figure 16. The damage to grouting problems, more than to execution or design
was mainly observed along the tendons and in the bridge mistakes.
superstructure at joints or anchorages, where water could In addition, Figure 18 shows that cracks have some
penetrate. importance in cases where monitoring and ordinary
maintenance could be performed. Cracks can be consid-
ered as a result of corrosion (due to the stress generated
3.3 | Damage and failure causes by the presence of corrosion products with increased vol-
ume compared to uncorroded steel). Taking this into
All the cases included in this work presented damage account, some studies have also started to use cracks as
and/or failure caused by corrosion alone, or by a com- an indicator for the degradation of bond and load-bearing
bined effect of corrosion and other factors resistance due to corrosion of reinforcement.33–35 How-
(e.g., overloading, fatigue13,30). ever, cracks are also a cause of corrosion. This can be
Figure 17 presents a list of the major damage and fail- considered when cracks are produced not only by design
ure causes which created the conditions for corrosion to and execution mistakes, but also by various other factors
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 MENGA ET AL.

or worse, through openings at joints, anchorages, or ten-

don deviation points not properly designed or con-
structed. Moreover, the above reported analysis showed
that other sources of external chlorides are sea water in
the liquid and air-form state (see Figure 20).
Most of Norwegian bridges are located in coastal
regions, and climate in Norway is notoriously cold. Tak-
ing this into consideration, the simultaneous presence of
both road salt and external chlorides coming from the sea
F I G U R E 2 1 Example of whitish segregated grout embedding
aerosol raises great concern. Even more since chlorides
corroding strands in an Italian bridge37
generally induce pitting corrosion. In fact, pitting corro-
sion creates localized holes and cavities in steel tendons,
like for instance, shrinkage and thermal effects. This con- it is more destructive than uniform corrosion, and results
firms that corrosion is induced not only by issues in the in reduced capacity, especially under repeated loads.18,19
planning and construction phases, but also by events dur- It is also more difficult to detect, since its occurrence is
ing the structure's service life. local, subjecting the tendons to high levels of stress con-
To cause the chemical reaction called corrosion the centration.36 Therefore, a more detailed analysis on cor-
presence of determinate elements (namely water, oxygen, rosion induced by road salt and external chlorides is
and chlorides—this last is not required in case of needed.
carbonation-induced corrosion) is needed. In this paper The analyses currently conducted23 showed that most
the words “corrosion causes” are used to indicate the failures were attributed to design and execution mistakes
sources of the aforementioned elements. Both corrosion that facilitated the ingress of external chlorides leading to
causes and damage and failure causes (see Figure 17) corrosion damage. Since chlorides usually need time to
contributed to the onset of corrosion in the studied cases. penetrate the concrete cover, these failures generally
Specifically, damage and failure causes created the condi- occurred after the structure was at least 10 years old.
tions (e.g., presence of cracks, weak points) and corrosion Design mistakes associated with these kinds of failure
causes provided the elements for corrosion to occur mostly concerned the inadequacy of the drainage system,
(i.e., water, oxygen, chlorides). the joints, and the position of the ducts' vents or of the
Figure 20 lists the major corrosion causes detected in anchorages, allowing water infiltrations inside the ducts.
the survey. Only one corrosion cause (i.e., the one that On the other hand, execution mistakes induced mainly
seemed to be the most relevant in enabling corrosion) is incorrect grouting operations, which resulted in the pres-
associated to each structure. External chlorides, mainly ence of voids in the ducts and/or vents not fully sealed.
coming from de-icing salts, were the main corrosion Consequently, design and execution mistakes
cause in the analyzed structures. Other origins of external resulted in:
chlorides were liquid and air-form sea water.
• tendons that were not adequately protected by the
grout-duct system;
4 | M AJ OR C AUSES FOR • ducts not adequately protected by concrete in sensitive
CORROSION locations like joints, anchorages, and tendon deviators.

This section presents the discussion on the results pre- Voids in the duct mean that tendons are surrounded
sented in Section 3. The discussion focuses on the two by air instead of grout. Hence, once chlorides penetrate
more common and dangerous causes of corrosion inside the duct, they react with the air and trigger corro-
encountered in this study: external chlorides penetration sion of the tendons. Since air is lighter than grout, voids
and presence of deficient grout. tend to be situated at higher locations inside the duct.
For this reason, the strands generally corroded more
severely along the top and sides.28
4.1 | External chlorides penetration In pretensioned bridges, the technology to apply pre-
stress forces does not involve the grout-duct system. The
In places with cold climates it is common practice to absence of this system:
spread salt on the roads to prevent them from icing.
However, chlorides contained in de-icing salts could pen- • reduces possible mistakes related to grouting
etrate inside the structure through pores in the concrete operations;
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
MENGA ET AL. 13

• makes NDE possible because there is no duct interfer- evaluate the performance of grouts for posttensioned
ing with the evaluation. bridge structures. The study demonstrated that conven-
tional grout was susceptible to develop significant levels
However, our analyses confirmed that, as with inter- of bleed-water under normal handling conditions. It was
nally posttensioned bridges, attention needs to be paid to also shown that, although the bleed-water eventually dis-
joints, anchorage regions, and other locations sensitive to sipated, significant corrosion developed on the strand at
mistakes in both phases of design and execution in pre- the grout-water interface. Moreover, one particular com-
tensioned bridges as well. In this research only a reduced monly used grout, containing an aluminium based
number of pretensioned cases were explored, so for fur- expanding admixture, experienced the highest amounts
ther detail on durability of pretension structures please of bleed-water development and subsequent corrosion.39
refer to other studies, for instance that by Osmolska In France, controls on grouting began in 1995. Labo-
et al.10 ratory testing in tilted transparent tubes showed a phe-
nomenon of exudation combined with a settlement of the
grout. The exudation phenomenon (i.e., the discharge of
4.2 | Presence of deficient grout a liquid or viscous gel-like material through a pore, crack,
or opening in the surface of concrete) was amplified by
The current study showed that in structures that exhib- the presence of a superplasticizer. At the end of the test,
ited failure of tendons, corrosion was mostly enabled by in the higher part of the tube, a layer of whitish paste
problems in the grout. In all these cases, the grout topped by a yellowish liquid was produced. The liquid
remained unhardened, with a whitish and soft appear- itself was topped by a space filled with air. Mineralogical
ance (Figure 21). High pH, low levels of chlorides, and analyses carried out at the “Laboratoire Central des ponts
high levels of sulfates were detected inside the chalky et chaussées” (LCPC) on the various products reported
grout, while corrosion would manifest as “deep and local- that40:
ized attacks that resembled the form of pitting attacks.”37
It is important to highlight that no corrosion attacks were • The whitish paste, which hardened quickly in contact
found in the areas where the tendons were embedded in with the air, was primarily made up of ettringite (40%),
a regular hardened cement grout.38 portlandite (20%), and calcite (20%). Calcite was com-
These grouting problems mean that the usual causes ing mainly from the carbonation of the portlandite.
of steel corrosion in concrete (chlorides and/or carbon- The remainder of the paste presented an enrichment
ation) could not be responsible for the corrosion attacks in admixtures and sulfates.
in the prestressing tendons: • The yellowish liquid presented a composition close to
the interstitial solution of a cement paste with a very
• high pH excludes the possibility of carbonation- strong alkalinity (pH of 13.8). This admixture could
induced corrosion; present in some cases an incompatibility with the
• low levels of chlorides exclude the possibility of pitting cement, causing an instability of the grout during the
corrosion; dormant phase of the set.
• plastic ducts exclude the possibility of water and air • The air was saturated with water (100% of relative
infiltration (in the absence of cracks due to design or humidity).
execution defects).
Under these conditions, combined with the effect of
The typical grout used through the 1990s was made of variations in temperature, a pure water condensation
cement, water, and an expansion agent, and in some could occur. The water would condense either on the
cases a gelling agent that made the grout thixotropic39 interior wall of the duct, or on the parts of the prestres-
(i.e., a property of certain gels and emulsions of becoming sing wires exposed to this air saturated with water, trig-
fluid when agitated and then setting again when left gering the onset of corrosion. Moreover, the high levels
at rest). of sulfates detected in the chalky grout might destabilize
When corrosion of tendons coincided with the pres- the growth of the passive film around the steel facilitating
ence of voids, the corrosion cause was generally identi- its corrosion.
fied as external chlorides penetration through Therefore, under these conditions, the cause of corro-
inadequately protected anchorages, grout, and vents. sion was identified with the processes of exudation,
However, in some cases the observed corrosion was due bleeding, and segregation due to variation in tempera-
to excessive accumulation of bleed water.39 Hence, in ture. This leads to rapid formation and diffusion of corro-
1989 the FHWA (United States) commissioned a study to sion, which explains why tendon failures have been
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
14 MENGA ET AL.

reported in the first years from the construction of the hand, frequent inspections and easy access mean
structure. warning signs are usually observed earlier in exter-
nal tendons than in internal tendons.
 In pretensioned bridges, damage mostly involves
5 | C O N C L U D IN G R E M A R K S the superstructure, particularly affecting joints, ten-
don deviation points, and anchorages.
This paper presents the results of an extensive literature
review on reported cases of corrosion of posttensioned To conclude, the conducted review23 showed that pre-
tendons in bridges. tensioned bridges and internally posttensioned bridges
The questions raised in Section 1 can be answered share some similarities related to where the corrosion-
based on the review: induced damage occurs (e.g., joints, tendon anchorages,
and deviation points). This observation highlights once
a. Detection of warning signs can help limit and reduce more the importance of correct design and execution,
the damage, provided that these conditions are especially in these specific locations. In fact, joints, ten-
fulfilled: don anchorages, and deviation points are known to be
 the damage needs to be detected early enough for vulnerable to corrosion attacks since they represent the
the repair measures to be effective; place where the continuity of the element is broken. This
 maintenance measures should be implemented as is particularly true for pretensioned bridges, generally
planned. consisting of simply supported precast prestressed beams
Nevertheless, the absence of warning signs does not separated by joints. On the other hand, in case of inter-
guarantee the safety of the structure. An accurate nally posttensioned bridges the tendons and the ducts
program of inspection and maintenance activities continue throughout the joint between two adjacent seg-
throughout all the life span of the structure is neces- ments. However, at joint locations we expect voids in the
sary. Particular attention should be paid to the quality, grout due to the grout composition, and corrosion of
more than to the frequency of the inspections, even in internal tendons is often associated with voids in the
the absence of warning signs. In this regard, the grout.
authors suggest developing a classification system for
the all the bridges in the network to assess their ACKNOWLEDGMENTS
robustness against tendon failures. The present research project was mainly conducted
b. Since all the cases with chalky grout problems were between September and November 2020, at the Depart-
observed in places with high-temperature climates ment of Structural Engineering at NTNU in Trondheim,
(e.g., Florida, Italy), it is the authors' opinion that Norway. The study was supported by the 5-year research
grouting-induced failures are more probable in loca- and development program for bridges and ferry quays of
tions where temperatures of 30 C are commonly steel and concrete “Better Bridge Maintenance,” estab-
reached. On the other hand, the analyses confirmed lished by the Norwegian Public Road Administration
that in cold climates the structures are more sensitive (NPRA) in February 2017.
to the penetration of external chlorides (mainly com-
ing from de-icing salts). DA TA AVAI LA BI LI TY S T ATE ME NT
c. Various types of structures were analyzed in the cur- The data that support the findings of this study are avail-
rent work, showing that: able from the corresponding author upon reasonable
 Internally posttensioned bridges can show all levels request.
of damage, and that they appear in the superstruc-
ORCID
ture at various locations (joints, anchorages, and
other structural details). Warning signs are not Antonia Menga https://orcid.org/0000-0001-7349-1726
always present. On the contrary, damage and fail- Anja Klausen https://orcid.org/0000-0002-0888-5769
ure may occur without showing any warning signs.
 Internally posttensioned slabs show damage mainly RE FER EN CES
at tendon anchorages. 1. Lee GC, Mohan S, Huang C, Fard BN. A study of US bridge
failures (1980–2012). Buffalo, NY: MCEER; 2013.
 In externally posttensioned bridges, the damage
2. Hadipriono FC. Analysis of events in recent structural failures.
usually consists in failure of external tendons. In J Struct Eng. 1985;111(7):1468–81.
cable-stayed/suspension bridges, the damage can be 3. Harik I, Shaaban A, Gesund H, Valli G, Wang S. United States
due to the exposure of the external tendons/cables, bridge failures, 1951–1988. J Perform Constr Facil. 1990;4(4):
which carry the main load globally. On the other 272–7.
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
MENGA ET AL. 15

4. Wardhana K, Hadipriono FC. Analysis of recent bridge failures 23. Menga A, Kanstad T, Cantero D. Corrosion induced failures of
in the United States. J Perform Constr Facil. 2003;17(3):144–50. post-tensioned bridges. Report. Norway: Department of Struc-
5. Miller MD. Durability survey of segmental concrete bridges. tural Engineering, NTNU; 2022. ISBN:978-82-7482-200-9.
PCI J. 1995;40(3):110–23. 24. Cousin B, Buchin-Roulie V, Vandevoorde C, Fabry N, editors.
6. Choudhury JR, Hasnat A, editors. Bridge collapses around the Hammersmith flyover: a complete innovative renovation. Proc. Int.
world: causes and mechanisms. IABSE-JSCE Joint conference Conf. on ultra-high performance fiber reinforced concrete; 2017.
on advances in bridge engineering-III; 2015. 25. Available from: https://www.allplan.com/products/allplan-
7. Clark GM. Post-tensioned structures—improved standards. bridge-2022-features/.
Proc Inst Civ Eng Forensic Eng. 2013;166(4):171–9. 26. Bazzucchi F, Restuccia L, Ferro G. Considerations over the Ital-
8. Matt P, Hunkeler F, Ungricht H, editors. Durability of pre- ian road bridge infrastructure safety after the Polcevera viaduct
stressed concrete bridges in Switzerland. IABSE congress collapse: past errors and future perspectives. Frat Integrita
report. Washington, DC: International Association for Bridge Strut. 2018;12:400–21.
and Structural Engineering; 2000. 27. Domaneschi M, Pellecchia C, De Iuliis E, Cimellaro G,
9. Haardt P, Holst R. The German approach to bridge manage- Morgese M, Khalil A, et al. Collapse analysis of the Polcevera
ment. International bridge and structure management. Tenth viaduct by the applied element method. Eng Struct. 2020;214:
international conference on bridge and structure management. 110659.
2008:3. 28. Pape TM, Melchers RE. The effects of corrosion on 45-year-old
10. Osmolska MJ, Kanstad T, Hendriks MA, Hornbostel K, pre-stressed concrete bridge beams. Struct Infrastruct Eng.
Markeset G. Durability of pretensioned concrete girders in 2011;7(1–2):101–8.
coastal climate bridges: basis for better maintenance and future 29. Venugopalan S. Corrosion evaluation of post-tensioned ten-
design. Struct Concr. 2019;20(6):2256–71. dons in a box girder bridge. International bridge and structure
11. Hou B, Li X, Ma X, Du C, Zhang D, Zheng M, et al. The cost of management. Tenth international conference on bridge and
corrosion in China. NPJ Mater Degrad. 2017;1(1):4. structure management 2008:286.
12. Koch G, Varney J, Thompson N, Moghissi O, Gould M, 30. Kashani MM, Crewe AJ, Alexander NA. Damage propagation
Payer J. International measures of prevention, application and in corroded reinforcing bars with the effect of inelastic buck-
economics of corrosion technologies study. NACE interna- ling under low-cycle fatigue loading. Life-cycle of engineering
tional IMPACT report, Houston: Nace International Impact. systems: proceedings of the fifth international symposium on
http://impact.nace.org/documents/Nace-International-Report. life-cycle civil engineering (IALCCE 2016), 16–19 October
pdf. 2016. 2016, Delft, The Netherlands. Life-cycle of civil engineering
13. Apostolopoulos C, Konstantopoulos G, Koulouris K. Seismic systems. London: CRC Press; 2016. p. 1996–2002. https://doi.
resistance prediction of corroded S400 (BSt420) reinforcing org/10.1201/9781315375175-293
bars. Int J Struct Integr. 2018;9:119–38. 31. Helmerich R, Zunkel A. Partial collapse of the Berlin Congress
14. Broomfield J. Corrosion of steel in concrete: understanding, Hall on May 21st, 1980. Eng Fail Anal. 2014;43:107–19.
investigation and repair. London; CRC Press; 2003. 32. Woodward RJ, Williams FW. Collapse of Yns-y-Gwas bridge,
15. Zhu W, François R, Coronelli D, Cleland D. Effect of corrosion Glamorgan. Proc Inst Civ Eng. 1988;84(4):635–69.
of reinforcement on the mechanical behaviour of highly cor- 33. Fischer C, Ozbolt J. An appropriate indicator for bond strength
roded RC beams. Eng Struct. 2013;56:544–54. degradation due to reinforcement corrosion. Proceedings of the
16. Cairns J, Plizzari GA, Du Y, Law DW, Franzoni C. Mechanical 8th international conference on fracture mechanics of concrete
properties of corrosion-damaged reinforcement. ACI Mater J. and concrete structures (FraMCoS), Toledo, Spain, 11–14
2005;102(4):256. March 2013; International Center for Numerical Methods in
17. Landolt D. Corrosion and surface chemistry of metals. New Engineering, 2013. p. 1828–35.
York: EPFL Press; 2007. 34. Apostolopoulos CA, Koulouris KF, Apostolopoulos AC. Corre-
18. Imperatore S, Rinaldi Z, Drago C. Degradation relationships lation of surface cracks of concrete due to corrosion and bond
for the mechanical properties of corroded steel rebars. Constr strength (between steel bar and concrete). Adv Civ Eng. 2019;
Build Mater. 2017;148:219–30. 2019:1–12.
19. Kioumarsi M, Benenato A, Ferracuti B, Imperatore S. Residual 35. Tahershamsi M, Fernandez I, Lundgren K, Zandi K. Investigat-
flexural capacity of corroded prestressed reinforced concrete ing correlations between crack width, corrosion level and
beams. Metals. 2021;11:442. anchorage capacity. Struct Infrastruct Eng. 2016;13:1294–307.
20. Available from: https://allaboutcivil.org/pre-stressed-concrete- 36. Morgese M, Ansari F, Domaneschi M, Cimellaro GP. Post-
types-of-pre-stressed-concrete-systems/ collapse analysis of Morandi's Polcevera viaduct in Genoa,
21. Hurlebaus S, Hueste MBD, Karthik MM, Terzioglu T. Inspec- Italy. J Civ Struct Heal Monit. 2020;10(1):69–85.
tion guidelines for bridge post-tensioning and stay cable sys- 37. Bertolini L, Carsana M. High pH corrosion of prestressing steel
tems using NDE methods. NCHRP report 848. Washington, in segregated grout. In: Andrade C, Mancini G, editors. Model-
DC: The National Academies Press; 2017. https://doi.org/10. ling of corroding concrete structures. Dordrecht: Rilem Publi-
17226/24779 cations, Springer; 2011. p. 147–58. https://doi.org/10.1007/978-
22. West JS, Larosche C, Koester B, Breen J, Kreger M. State-of- 94-007-0677-4_10 Joint Research Centre. JRC science and polit-
the-art report about durability of post-tensioned bridge sub- ical report. New European technical rules for the assessment
structures. Austin: University of Texas at Austin, Center for and retrofitting of existing structures. Luxembourg: Publica-
Transportation Research; 1999. tions Office of the European Union; 2015.
 17517648, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/suco.202200297 by NTNU Norwegian University Of Science & Technology/Library, Wiley Online Library on [12/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
16 MENGA ET AL.

38. Carsana M, Bertolini L. Corrosion failure of post-tensioning

tendons in alkaline and chloride-free segregated grout: a case Lise Bathen
study. Struct Infrastruct Eng. 2015;11(3):402–11. Chief Engineer, Norwegian Public
39. Ahern M, Yousef A, Xia Z, Lewis R, Poston R, editors. Anat- Road Administration (NPRA), Oslo,
omy of grouted post-tension tendon failure. Proceedings of the Norway
5th fib congress Fédération internationale du beton (fib), Lau- Email: [email protected]
sanne, Switzerland; 2018.
40. Godart B, editor. Durability of post-tensioning tendons: techni-
cal report. Proceedings of a workshop held at Ghent University
on 15–16 November 2001; fib Fédération internationale du
béton. 2001: in France.
Karla Hornbostel
Chief Engineer, Norwegian Public
AUTHOR BIOGRAPHIES Road Administration (NPRA),
Trondheim, Norway
Antonia Menga Email: karla.hornbostel@
Ph.D. Student, Department of Struc- vegvesen.no
tural Engineering, Norwegian Uni-
versity of Science and Technology
(NTNU), Trondheim, Norway Anja Birgitta Estensen Klausen
Email: [email protected] Associate Professor, Department of
Structural Engineering, Norwegian
University of Science and Technol-
Terje Kanstad ogy (NTNU), Trondheim, Norway
Professor, Department of Structural Email: [email protected]
Engineering, Norwegian University
of Science and Technology (NTNU),
Trondheim, Norway
Email: [email protected]

How to cite this article: Menga A, Kanstad T,

Cantero D, Bathen L, Hornbostel K, Klausen A.
Daniel Cantero Corrosion-induced damages and failures of
Associate Professor, Department of posttensioned bridges: A literature review.
Structural Engineering, Norwegian Structural Concrete. 2022. https://doi.org/10.1002/
University of Science and Technol- suco.202200297
ogy (NTNU), Trondheim, Norway
Email: [email protected]

View publication stats