Keywords

1 Introduction

Understanding the rate of material degradation can be critical in preserving the fabric of historic concrete buildings. Durability models can be utilized for predicted performance of specified materials within their given environment. This allows for the determination of remaining service life. By forecasting the probabilistic long-term behavior of corrosion related degradation, preventative maintenance and proactive repairs can be planned.

Durability modeling is based on a detailed material assessment that focuses on structure specific conditions. The assessment identifies mechanisms of deterioration, including physical, chemical, structural, mechanical and electrochemical damage, which is incorporated into mathematical models. The models are based on predetermined degradation thresholds. Long-term monitoring systems can be installed to provide extensive data ranges and higher statistical reliability for the analysis. This approach allows owners to make informed decisions on the best repair choice for the life extension of the buildings and can also be successfully used to determine how repairs perform.

2 Historic Concrete and Deterioration

2.1 Preserving Concrete

Reinforced concrete revolutionized construction of the 20th century. Design, construction methods, materials, and standards of workmanship will vary enormously according to the date of construction. In addition, location and climate will play a part in the deterioration. All these factors will affect the durability of the structure [1].

As concrete buildings begin to age, those deemed architecturally significant will require material conservation. Challenges exist in the assessment and repair of historic concrete buildings, however an appropriate repair program addresses existing distress and reduces the rate of future deterioration [2]. Related disciplines such as construction history, conservation philosophy, concrete durability and concrete repair [3] must be integrated to properly assess performance.

2.2 Establishing Condition States

To understand durability, a condition state rating system and service life analysis will define the current condition of the structure. Visible defects are identified as known conditions. The unknown conditions, or the areas that are not yet visibly defective, is where durability modeling and a service life analysis is required. Data collection and subsequent analytical procedures provide timeframes on deterioration critical to future performance. The service life assessment should provide a time line of when thresholds are reached in the structure and when obsolescence will occur [4]. When studying behavior of historic concrete, condition state parameters should include carbonation, temperature, relative humidity and corrosion.

2.3 Corrosion Deterioration

Corrosion of reinforcing steel is arguably the most detrimental deterioration mechanism to concrete. It is not a question of if a concrete structure will corrode; it is a question of when it will corrode. Traditional corrosion models established by Tuutti [5] define three modes of time based deterioration in reinforced concrete structures: ti, tp, tf (see Fig. 1).

Fig. 1.
figure 1

Modified Tuutti model

Full size image

Where:

ti = Time to Corrosion Initiation

tp = Time to Corrosion Propagation

tf = Time to Failure.

The corrosion process for steel reinforced concrete can be simplified into a two-stage process, namely the ‘initiation phase’ and the ‘propagation phase’. The initiation phase is the time taken for conditions to become conducive to corrosion through carbonation or chloride ingress. The propagation phase is the period in which the accelerated corrosion of the steel reinforcement ultimately leads to rust staining, cracking and spalling of the cover concrete [6].

When corrosion of steel is concerned, the service life [tl] is assumed as the sum of an initiation period and a propagation period: tl = ti + tp [5]. Initiation of corrosion is usually chosen as limit state for chloride-induced corrosion [or carbonation]… As a consequence, the service life [of the reinforced concrete structure] is assumed to be coincident to the initiation time (i.e., tl = ti) [7]. In chloride rich environments, the service life or tl has proven to be very short in some instances and corrosion is certainly anticipated on structures within 20 years or less.

Carbonation induced corrosion, which tends to initiate at cracks and defects, typically has an extended ti. Rates of carbonation are variable, and depend upon material characteristics of the concrete, such as:

  1. (1)

    the level of pore water, i.e. relative humidity;

  2. (2)

    grade of concrete;

  3. (3)

    permeability of concrete;

  4. (4)

    whether the concrete is protected or not;

  5. (5)

    depth of cover; and

  6. (6)

    time [8].

Typically, tl for historic concrete buildings subject to carbonation is longer than structures subject to high chloride loads (i.e. bridges, parking garages, marine structures).

3 Establishing the Parameters

3.1 Establishing a Carbonation Protocol

Concrete is highly alkaline in nature. Carbon steel in alkaline concrete will have negligible rates of corrosion while the Fe3O4 (passive layer) is intact. Carbonation is the reaction between the cement hydration products in the concrete and atmospheric carbon dioxide (CO2), which leaves the steel vulnerable to corrosion [9]. The resulting drop in pH leaves the steel unstable, thus subject to corrosion in the presence of oxygen and moisture. Carbonation rates generally follow parabolic kinetics where the depth of the reaction on a sample or structure is proportional to the square root of time (Eq. 1):

$$ d = At^{0.5} $$
(1)

where:

  • d = carbonation depth,

  • t = time and

  • A = constant, generally of the order 0.25 to 1.0 mm.year−0.5 [10].

The progress of the carbonation front is the most common means of expressing the extent to which carbonation has occurred [9]. Fridh and Lars-Olof suggest that carbonation is a moving boundary where the rate is dependent upon the amount of CO2 in the air, the diffusion coefficient of the carbonated concrete (established by the NT 492 Test), as well as the amount of CaO which can be carbonated [11]. Simple phenolphthalein tests which are spray or brush applied indicate when pH has dropped below 8.3, however the loss of the passivity usually occurs around pH 11 [8] to 11.5 [9]. A carbonation front can be 5 mm ahead of the delineator. Therefore, phenolphthalein test results are not sufficient to gauge actual carbonation depths. Once carbonation occurs at the steel corrosion may ensue.

3.2 Temperature and Relative Humidity

Corrosion kinetics are influenced by oxygen availability, temperature and moisture. The three environmental factors that have the largest impact on the rate of carbonation are CO2 concentration, relative humidity [RH] and temperature [9]. Therefore, these factors are to be considered when modeling future carbonation progress and corrosion risk. The overall effect is that the optimum rate of carbonation occurs at 55% RH [9]. RH values greater than 75% will provide sufficient moisture for corrosion to initiate in carbonated concrete. The most aggressive corrosion activity will occur in carbonated concrete with RH values between 85% and 95%, at temperatures above 20 °C [12].

Since the deterioration of concrete is based on diffusion or chemical reactions, the deterioration progress can also be considered as having temperature dependency. Generally, the rate of material diffusion and the chemical reactions rise with increases in temperature [13].

3.3 Site Specific Corrosion Data

To establish site specific condition states from corrosion activity, Linear Polarization Resistance (LPR) testing is required. In any real corrosion system, corrosion current (Icorr) and corrosion rate are a function of many system parameters, including type of metal, composition of the electrolyte solution, concrete type, temperature, movement of the solution, material resistivity, etc. LPR measures both Half-cell potential (Ecorr) and Icorr. Icorr is the loss of electrons occurring at the anodic site of the corrosion cell and is measured in microAmps per square centimeter (µA/cm2) [14].

The values of Icorr, are used to then calculate steel section loss (corrosion rates) using Faraday’s Law of Metal Loss, and measures in microns per year (µmyr−1). The rate of degradation of the steel and the predictive condition state can be established, as seen in Table 1. As LPR is variable, more accurate models can be generated if monitoring probes can be installed. Otherwise, seasonal measurements should be acquired to understand Icorr values. In buildings where RH is 85%, corrosion rates of 3 µmyr−1 would be typical, at 90% 12 µmyr−1 and at 95% 50 respectively [12]. These values drop to 2 µmyr−1 or less in saturated concrete.

Table 1. Corrosion rates and condition states [14]
Full size table

4 Case Study

4.1 Concrete Building Case Study

A reinforced concrete museum in a metropolitan area, constructed in 1954, underwent a major repair campaign in 2007–2008. The building was identified to be suffering mildly from carbonation induced corrosion. The walls comprise 12.7 cm thick gunite which are rigidly connected to a ramp slab (floor) and web walls. Original construction records indicate that the walls were reinforced with two layers of 50 mm (2 × 2) mesh, plus 2 layers of No. 3 bars at 30 cm centers horizontally, 2 layers of No. 4 bars vertically, and top and bottom continuous No. 4 bars—all secured to a framework of 38 mm ‘tees’ which are in turn tied into the structure. The outside surfaces were smoothed after the forms were removed, but while the gunite was still green [15]. Soon after construction the exterior walls were painted with a marine coating.

4.2 Verification of Conditions

Prior to the corrosion monitoring program, a full condition assessment and structural analysis were performed. The review indicated carbonation had diffused between 1 and 6 mm into the surface of sound concrete from the exterior and up to 17 mm from the interior face of the gunite wall. Corrosion had been documented along the cracks which were parallel with the reinforcing steel of the rotunda dating back to 1989, thirty years after completion. Reports up to 2006 confirmed visual correlation between hairline cracking, and minor corrosion activity at the reinforcement. Reinforcing steel cover varied, however low cover was approximately 25 mm.

4.3 The Condition Survey

A corrosion condition assessment was performed with half-cell (Ecorr) and corrosion rate (Icorr) values measured and mapped on two areas of the building. This included the 6th floor rotunda (external), and the 2nd floor gallery wall (internal). The analysis of half-cell potentials values, and potential gradient mapping provided visual support of corrosion activity. The corrosion assessment indicated that only 4% of the area tested exhibited corrosion rates greater than 2.2 µmyr−1 (Condition State 3). This would be considered moderate [16] in most structures. However, higher rates measured during the testing program from Condition State 2 (4 µmyr−1) indicated ongoing damage in the form of cracking within 10 years at areas of low cover. This level of degradation is significant for a landmark concrete structure. The client required on-going and continued use of the facility. As the data from the survey indicated future risk, a long-term corrosion monitoring system was installed to better understand the performance of the structure, and post-repair conditions.

4.4 Monitoring System

The subject areas of the monitoring program included the survey areas at 6th floor building envelope walls (Rotunda) and the 2nd floor gallery walls. The system consisted of embedded temperature and humidity probes and LPR probes comprised of Silver/Silver Potassium Chloride (0.5 M) reference electrodes, each with an auxiliary Mixed Metal Oxide Titanium electrode. Both half-cell potential and corrosion rate values could be collected and measured from the same probe. Probes were installed at high risk locations with two controls at an internal web wall.

The locations deemed high risk for the 6th floor Rotunda were in two ‘Bays’ identified as 62 and 67. 62 is a West facing bay with a neighboring adjacent high-rise building, and Bay 67 is south facing, where the highest environmental exposure was noted. LPR Probes were placed to measure the performance of the following steel components: the Web Wall Double ‘Ts’ i.e. the internal steel connection angles at transverse web wall and exterior wall; Intermediate ‘Ts’; Inner Rebar, Outer Rebar; Inner Mesh, Outer Mesh. Data for the initial year is provided in Figs. 2 and 3, with statistic of corrosion rate, temperature and humidity in Tables 2, 3, 4, 5 and 6.

Fig. 2.
figure 2

Bay 62 LPR monitoring data over 1 year

Full size image
Fig. 3.
figure 3

Bay 67 LPR monitoring data over 1 year

Full size image
Table 2. Statistical data corrosion current Bays 62
Full size table
Table 3. Statistical data corrosion rate Bay 62
Full size table
Table 4. Statistical data corrosion current Bays 67
Full size table
Table 5. Statistical data corrosion rate Bay 67
Full size table
Table 6. Embedded temperature and humidity data Bay 62 and 67
Full size table

Temperature and humidity probes were installed in three ‘spaces’ within the two bays. The probe placement was (1) embedded within the walls; (2) within the interstitial space at the internal wall face enclosed within a protective box, and (3) within the insulation under the air barrier for internal ambient. The data has been collected over an eight-year period, however for the purposes of brevity the first year of corrosion data is being reviewed.

5 Conclusions

The long-term data acquired from the monitoring system illustrated the improved performance of the building envelope. The values of the data collected from the LPR readings indicates that over the course of the first year of monitoring, the building shifted corrosion risk categories. Prior to the repairs, the average values were 2.2 microns. After the monitoring system was installed, the average corrosion rates were measured at less than 0.106 micro/Amps per square centimeter (<0.1 µA/cm2) or 1.24 microns (m−yr) section loss for Bay 62 and 0.08 micro/Amps per square centimeter (<0.1 µA/cm2) or 0.98 microns (m−yr) section loss. This data indicates that the reinforcing steel in the location of the probes to be a low risk of corrosion. Using the new inputs in the crack propagation models, anticipated damages were predicted to occur >25 years. The decrease in the wall RH from 100% at the time the coating was applied, to values less between 76% and 65% over a 1-year period indicated that corrosion condition states were decreasing and building envelope conditions had improved. The information indicated a reduction in deleterious conditions impacting corrosion.

The general assumption made by the various system reviews supported that the stable conditions remain and advancing conditions states were not posing a risk to the building elements which were monitored. These assumptions can be based on reducing the movement of the carbonation front, reducing moisture and measuring on going corrosion activity. In this instance, coating integrity is required to minimize further corrosion propagation. The monitoring allowed for a determination of repair performance to be assessed and to scientifically gauge whether further interventions were required.

The comprehensive approach can provide a timeline for degradation, budgeting and interventions to landmark concrete structures prior to damage occurring on the structure. In this instance, understanding the risks affecting advancing condition states, or a reduction, allowed the team assurance in their chosen repair. The work demonstrated that the conditions on the structure stabilized over time, further reducing moisture, and subsequent corrosion in the building envelope. By understanding the reactions driving corrosion kinetics and how active corrosion will become once initiated, allows time frames for intervention to be developed.