Flawed Freeze-Thaw Testing Standards Unreliable for Predicting Concrete Tie Failures

The scientific world has achieved remarkable discoveries and innovations responsible for the modern technology we rely on today. Despite our scientific sophistication, the ability to manipulate time—speed it up, slow it down, reverse it—remains largely a figment of our collective imagination, or a plot vehicle in a science fiction novel. This is not a ground-breaking revelation, but you may be surprised at the ability of scientists and engineers to “predict the future” through the use of accelerated tests performed in the laboratory. These include product testing to determine anticipated life, susceptibility of materials to environmental degradation, or remaining life of damaged parts in a particular environment or application. If you want to know if a product is going to last 30 years, running a 30-year test is not practical and will not provide manufacturers or regulators the ability to anticipate premature failures or affect timely course-corrections to avoid such problems.

Sample bar cut according to freeze-thaw testing specifications

A sample bar cut according to freeze-thaw testing specifications. The arrows point to a crack formed during cutting of the bar from a high performance concrete railroad tie. Depending on location and size of the sample, cracks may form from mechanical stresses that can give inaccurate results that are not indicative of the performance of the engineered end-product.

The ability to perform accelerated laboratory evaluations is predicated on knowledge of the service environment being simulated and understanding the degradation mechanisms normally encountered in that service environment. Acceleration is made possible by imposing more aggressive conditions in a controlled manner without changing the nature of the degradation mechanism being assessed. Often this is achieved by increasing temperature and/or increasing the concentration of chemical components in the test environment. However, if the test conditions are too extreme, failure modes other than those normally encountered in the field could influence or dominate the results, thus rendering the test meaningless. Acceptance of results from testing that initiates unrealistic failure modes could lead producers down the wrong path, wasting resources to rectify phantom or nonexistent product performance issues. If the test results are inappropriately used for quality control (QC), they could lead to the rejection of entire lots of perfectly acceptable products. Knowing the product’s field failure modes and understanding the physical mechanisms leading to those failures can provide guidance for the development of appropriate accelerated test design and methodology.

Many accelerated testing techniques have gained wide technical acceptance and have been standardized by technical organizations like ASTM International. Such standards are extremely useful in comparing data from different studies and different materials. They are often written as broadly as possible, but they cannot capture all of the nuances of every engineering situation or application. As such, it is always necessary to scrutinize analytical results to make sure that they provide an accurate description of real-world behavior. Performing accelerated tests should not be viewed like baking a cake from a recipe—applying a particular standard test for any and all materials or test conditions does not guarantee accurate insight about future behavior in all cases, particularly if observed field failures cannot be reproduced. Making life cycle predictions without appropriate interpretation of accelerated testing results can be a recipe for disaster.

Concrete railroad ties are one example of the challenge of correctly applying accelerated tests, both for QC and as a prediction of durability in the field. First used in the late 1800s and improved through focused research in the 1960s and 70s, concrete railroad ties make up about 20% of the ties in service on major railroads. Although not economically viable at their inception, concrete ties have made in-roads because they have a longer life span, require fewer ties per mile, need less maintenance, and can carry heavier loads than traditional wooden ties. With the goal of predicting performance in the field, concrete ties are subjected to accelerated freeze-thaw testing via the American Society for Testing and Materials (ASTM) standard C-666 Rapid Freezing and Thawing in Water [1], a widely used and controversial method in the concrete industry. This method is used to test a range of materials, from concrete building slabs to high performance concrete. The debate over the test arises when laboratory results are inconsistent from lab to lab and when the test results do not correlate to field performance.

Although freeze-thaw is a well-known damage mechanism in the concrete industry, it is not one of the failure modes of importance in concrete railroad ties. In fact, the Federal Railroad Administration (FRA) of the U.S. Department of Transportation, which promulgates regulations for concrete railroad ties, does not address freeze-thaw. The FRA focuses on safety concerns associated with loss of gauge and risk of derailment. FRA publishes deterioration limits for ties in track [2], and based upon failure modes experienced in the field, namely rail seat deterioration (RSD), FRA now requires that railroads perform automated inspections of all track constructed with concrete ties [3]. RSD is the gradual wearing away of the concrete under the tie pad that holds the track to the tie and can be repaired in the field. Other historically observed failure modes include flexural cracking from the center binding and rail fastener failure.

Performance specifications for concrete railroad ties and their constituent parts are given by the American Railway Engineering and Maintenance-of-Way Association (AREMA) [4] and can be enhanced at the discretion of the manufacturer or buyer (railroad company). Some of these specifications focus on freeze-thaw testing using ASTM C-666. Several investigators from around the world have recognized a difference in laboratory freeze-thaw behavior for high performance concrete that is not reflected in the field performance of the material. This matches RJ Lee Group’s experience in this field, suggesting that the ASTM test does not simulate actual field performance. The ASTM allowable cooling rate, along with the water saturation required, overwhelms the normal kinetics of the water-air void system in the concrete. The water freezes and expands before reaching the air void and causes the concrete to crack. Whereas maximum cooling rates experienced outdoors can be about 3-4 oF per hour, the ASTM test allows rates 10 times that amount.

The ASTM standard itself, as it was designed for broad use, allows much leeway in the parameters of the test. Varying the test parameters within the limits of the standard should not affect the outcome, but with ASTM C-666 different results will be produced depending on the location within the allowable range. One important parameter, the cooling rate, can vary by a factor of three and still meet the standard. Economics of the test may entice a testing laboratory to increase the sample throughput by running the test at the maximum cooling rate for a quicker total cycle time, but that may lead to an increased failure rate much to the dismay of the manufacturer. Samples passing at one lab can easily fail at another. It is well known that the amount of concrete damage increases with faster cooling rates. ASTM does not provide data on lab-to-lab variability because of the “limited possibility that two or more laboratories will be performing freezing-and-thawing tests on the same concretes [1].”

But herein lies the necessity for critical interpretation of test results. A new class of high performance materials is available and currently in use, and the only accepted method of predicting long-term durability yields results that are inconsistent with real-world performance. Understanding the sensitive issues on an industry-wide scale is made difficult because of the inconsistencies on how the test can be applied within the industry. For example, AREMA deviates from the ASTM standard by recommending more stringent passing criteria and using smaller sample dimensions. Some labs even test specimens that contain prestressed steel reinforcement. Further inconsistencies in the industry originate from individual buyer specifications with different requirements for the frequency of freeze-thaw testing for QC, the number of samples to be tested for a given test result, the dimensions of the test samples, and the passing criteria.

Advances in cement materials and technology have led to the development of high performance concrete that is particularly well suited to placement in track as a railroad tie. The high compressive strengths that are possible today create advantages for the railroad industry in terms of tie performance, but present challenges in evaluating the long-term durability of these materials using ASTM C-666. The current qualification and acceptance criteria for high performance concrete railroad ties are based on an accelerated test with a low degree of precision and no correlation to observed field failures. The lack of industry recognition and apparent limit to the applicability of the test standard may result in the removal of safe, acceptable ties from service, costing many thousands of dollars in wasted time and material. A need exists to move away from a broadly written standard that attempts to characterize a wide range of materials, to one that is developed specifically for a given class of material— one that is a reliable predictor of field performance. The railroad industry, its suppliers and customers can benefit from the development of a more realistic accelerated test that accurately represents field performance of high performance concrete ties. This can be achieved by re-assessing the ranges of appropriate test parameters (temperatures, cooling rates, and heating rates) in ASTM C-666 to specifically address these new materials. This development process has precedent in other ASTM standards that modify severity of accelerated corrosion environments based on material and condition.


1. ASTM Standard C666/C666M, 2003, (2008), “Test Method for Resistance of Concrete to Rapid Freezing and Thawing, Procedure A, Rapid Freezing and Thawing in Water,” ASTM International, West Conshohocken, PA, 2008.

2. Federal Railroad Administration DOT, 49 CFR 213.335(d) “Crossties.”

3. Federal Register, Vol. 76, No. 175, Friday, September 9, 2011, Rules and Regulations, “Track Safety Standards; Concrete Crossties,” page 55819.

4. AREMA Manual for Railway Engineering, American Railway Engineering and Maintenance-of-Way Association (AREMA), Landover, Maryland, 2009, v 1, ch. 30, “Ties.”

To learn more about our concrete failure analysis visit this page.

Leave a comment