Test Samples of Reinforced Concrete Are Often Too Small

Localized pitting corrosion is often found in the steel reinforcements within concrete, according to an ETH Zürich research team. Photo by Ueli Angst, ETH Zürich.

A recent study from  a European university suggests that samples of reinforced concrete need to be significantly larger to ensure accuracy in corrosion detection.

Ueli Angst and Bernhard Elsener, professors at science, technology, engineering and mathematics university ETH Zürich (Zürich, Switzerland), released their findings1 on the size effect in measuring laboratory corrosion rates and the potential shortcomings when compared to real-world conditions.

“In our research project, we examined reinforced concrete specimens of various sizes and discovered that corrosive chloride concentration was far more apparent in smaller samples and subject to larger fluctuations than in larger specimens,” Angst says in a news release.2

Chloride Threshold to Initiate Corrosion

According to Angst, chloride-induced corrosion of the reinforcing steel in concrete is the most prominent cause for corrosion in concrete infrastructure. Localized pitting and crevice corrosion are particularly prevalent forms, often owing to harmful environmental influences such as carbon dioxide (CO2) in the atmosphere and deicing salts. Over the years, chloride from the salt penetrates the concrete until it eventually reaches the steel reinforcement.

Traditionally, scientists have tried to evaluate corrosion rates by relying on the determination of a chloride concentration level needed to initiate corrosion. “All state-of-the-art models for forecasting chloride-induced steel corrosion in concrete are based on this concept,” the professors write.

Along with visual appraisals and the use of nondestructive testing methods, the determination is often made via the extraction of concrete samples, which are then tested.

“The chloride concentration in the samples is calculated in the laboratory,” Elsener says. “If the sample exceeds the critical threshold of 0.4% relative to cement weight, not just near the surface but in the deeper levels, the assumption to date has been that corrosion could soon set in and that repairs were required.”

However, the professors believe that the accuracy of those tests is strongly correlated with the size of the exposed steel surface area, with samples that are too small leading to too much variance for the results to be valid. They say typical concrete samples measured today range from about 50 to 200 mm.

Small vs. Large Specimens

In their experiment, the professors found that for concrete sample specimens with 100-mm of exposed reinforcing steel length, the observed chloride threshold values for corrosion to initiate ranged from 0.9% to 2.1% chloride by mass of cement. In the smallest specimen studied (10 mm), no corrosion was observed during the test period, and the corresponding chloride concentration at the steel depth was determined at 2.4% by mass of cement. For 1,000-mm long specimens extracted from the same site, chloride concentrations of only 0.6% to 1.2% by mass of cement were found to be sufficient for corrosion initiation.

According to the professors, the reason for the discrepancy traces back to the inhomogeneous nature of concrete as a composite material. They explain that the steel-concrete interface in engineered structures may show significant variability in its local properties that  arise from the localized presence or absence of concrete cracks, honeycombs, voids of different origins, spacers, tie wires, mill scale, native rust layers, and steel surface contaminants. Many of these characteristics can act as weak spots in the concrete and are generally thought to initiate corrosion, such as air voids or crevices in the mill scale.

“Concrete is not homogeneous,” Angst says. “The size effect of corrosion can be directly accounted for by these differences. Only the analysis of a larger specimen, say a meter long, will allow a realistic assessment of the condition.”

Unfortunately, such larger specimens are not always practical to extract or bring to the laboratory. As a result, the professors suggest using a mathematical formula to account for the size discrepancy, rather than simply sticking to the fixed threshold of 0.4%.

For example, if a 1,000-mm (e.g., 1-m) specimen cannot be extracted, they suggest taking a combination of smaller specimens to reach that same size, such as 10 specimens that are 100 mm in size. Then, the chloride concentration at which corrosion initiates in the minimum of the 10 smaller specimens could be used as the basis for calculations. The professors refer to this dynamic as the “weakest link theory.”

“Ultimately, application of the weakest link theory to chloride-induced corrosion in concrete may significantly contribute to successful translation of laboratory results to engineering structures,” they write.

Findings Applicable to Sensors

The professors explain that their findings are also relevant to the use of sensors that are often built into reinforced concrete structures to monitor corrosion. Typically, they say, the sensor-based approaches are based on carbon steel specimens embedded at increasing depths within the concrete cover. They are monitored until corrosion initiates at a probe at a certain depth.

However, the professors say the surface area of most probes used in practice ranges from a fraction of 100 mm2 up to ~5,000 mm2. As a result, the sensor sizes include a surface area that is smaller than the recommended 1-m size found through their study.

According to the professors, this can be counteracted by either increasing the number of sensors at a given depth or by applying the weakest link theory for the needed size translation.

Going forward, the professors say further research is needed to better quantify the size effect on real conditions, such as testing reinforcing steel samples using conditions found at construction sites and embedded in mature concrete. The professors say laboratory studies like theirs are often limited because corrosion is investigated under conditions that may not be representative of engineered structures—where there can be even greater variations in properties at the steel-concrete interface.

“The size effect may tend to be more significant in these structures,” the professors write. “We consider it important to experimentally validate this and to quantify the size effect for practical conditions.”

Source: ETH Zürich, www.ethz.ch/en. Contact Ueli Angst, ETH Zürich—email: ueli.angst@ifb.baug.ethz.ch.


1 U. Angst, B. Elsener, “The Size Effect in Corrosion Greatly Influences the Predicted Life Span of Concrete Infrastructures,” Science Advances, Aug. 2, 2017, http://advances.sciencemag.org/content/3/8/e1700751.full (Dec. 13, 2017).

2 “When Time Ravages from Within,” ETH News, Aug. 3, 2017, https://www.ethz.ch/en/news-and-events/eth-news/news/2017/08/when-time-ravages-from-within.html (Dec. 13, 2017).

Related Articles