Part I: Do anti-icing agents ever increase road's slickness?
Lab studies were done on asphalt testing a number of anti-icing chemicals with a focus on the relationship between a road's slipperiness and type of chemical applied, temperature, humidity, velocity, and other factors.
Editor's note: The following is Part I of a report—"Liquid Anti-Icing Chemicals on Asphalt: Friction Trends"—derived from a study performed to show the relationship, if any, of anti-icing chemical application and a road's slipperiness. The first part of the report describes a summary of the study as well as the chemicals tested and the procedure used. Part II will focus on the results with each chemical tested. Part III will be an analysis and discussion of the results and Part IV will offer conclusions and recommendations. The study is of significance to the public works community due to the rise in litigation related to vehicle accidents on treated roads as well as the increased use of anti-icing agents.
By Timothy S. Leggett, P.Eng.
and Gerald D. Sdoutz, E.I.T.
Forensic Dynamics Inc.
Contents
Slickness issues often related to driver perception
Prudent use of chemicals can reduce likelihood of slickness
Anti-icing chemicals tested are listed
Anti-icing chemicals depend on humidity for friction capability
Computer modeling places likelihood of slickness very low
Testing conducted in a climate-controlled test chamber
Earlier research showed friction was also velocity dependent
The purpose of this research was to determine whether there existed a chemical "slipperiness", on an asphalt surface, as a result of a transition from liquid to a solid, and vice versa, of typical anti-icing chemicals presently in use. The number of reported slickness issues as a result of chemical treatments is infinitesimally small, (presently estimated at less than 1/1000th of 1% of all liquid anti-icing treatments).
Slickness issues often related to driver perception
Prior research has shown that the slickness issues are often related to driver perception, contaminants on the roadway, other than the chemicals themselves, and chemical dilution resulting in re-freeze. There have been some incidents reported where the chemicals themselves, prior to re-freezing, created a slipperiness for some unexplained reason. This research has shown that, indeed, when most chemicals transition from liquid to solid, and solid to liquid, a "slurry" phase is formed. This produces a relatively short-lived reduction in co-efficient of friction for most chemicals. This reduction is anywhere from non-existent (CMA and CMA-25) to a substantial 22%, (Liquidow).
The research has shown that all chemicals tend to be unstable in the "slurry" phase during the state transition, meaning that they pass through, this possibly slippery, phase quickly, and that it is unlikely that this phase can exist for long periods of time. Relative humidity values above those required to cause the state transition, appear not to affect the friction dramatically. However, at humidity levels in the high 20s to low 30s, most chemicals will begin to dry out, (after application as a liquid) potentially resulting in somewhat lower friction values during the transition phase.
All chemicals, upon continued dehydration, reached a solid state. On the asphalt surface, the solid state co-efficient of friction, of most chemicals, is essentially equivalent to that of a clean and dry asphalt roadway. Some even increased the coefficient of friction above 1.0 (CMAK, Ice Ban, and MCP). Due to an unexpected deterioration of the test tire during the research on asphalt, liquid and transition state friction results showed a steady decline as testing continued. These results should, therefore, not be used to compare chemicals. The results, however, are felt to represent conservatively low estimates of the friction which can be expected if one of the tested chemicals was applied to a contaminant free asphalt roadway. For continued testing a new tire will be used for each test. (Back to top)
Prudent use of chemicals can reduce likelihood of slickness
It appears that prudent use of the chemicals, (particularly with regards to application rate, frequency, and other contaminants) bearing in mind expected humidity levels, can further reduce the likelihood of slickness developing, particularly in the fall season when most incidents are reported to have occurred. It is felt that most anti-icing agent related incidents are most likely a result of the chemical being applied following a dry period, which likely causes a slippery emulsion to be formed by the chemical and the oil, grease, glycol, etc. contaminants which have build up on the roadway. Flushing the roadway with water, prior to chemical application, would prevent this.
The present research deals with tests which were performed on an asphalt surface in an effort to better understand what effect reliance of anti-icing chemicals on temperature and humidity, would have on the road friction co-efficient (ie. on asphalt). (Back to top)
Anti-icing chemicals tested are listed
Note: Other chemicals were tested as a result of our involvement in litigation, and chemical manufacturer research and development. However, these results cannot be released at this time. (Back to top)
Anti-icing chemicals depend on humidity for friction capability
Presently, we are continuing to perform such tests for end-users, distributors and manufacturers. The involvement of an end-user generally coincides with a reported slickness and a concern by the user that the chemical he/she is in possession of may not meet expectations.
Manufacturer involvement usually indicates a desire, on the part of a progressive company, to design a product with better friction capability, as compared to a pure chemical or a chemical with a rust inhibitor mixed in solution. The Dow Chemical company, for example, is experimenting with a variety of chemicals designed to increase the friction performance of their existing product.
The set of experiments included herein were the direct result of a meeting sponsored by the Snow & Ice Co-operative Pooled Fund Program (SICOP) of the American Association of State Highway and Transportation Officials, which was held in March of 1999, in Minneapolis. At that time, it was theorized that some chemicals may contain a special "slippery" state, and that this special state was responsible for a small number of reported slickness incidents, which had occurred around the continent. (Back to top)
Computer modeling places likelihood of slickness very low
The earlier research on a glass substrate indicated that when a liquid anti-icing chemical transitions into a solid, it may pass through what is referred to as a "slurry" phase. This has been shown to produce a relatively short-lived reduction in the co-efficient of friction for most chemicals. On a sandblasted glass surface, the reduction ranged from a low of 0.4% (Freezgard with IceBan) to 29% for Liquid Dow.
The purpose of this latest research was to determine whether such a friction reduction could be expected on asphalt. (Back to top)
Testing conducted in a climate-controlled test chamber
The friction was measured using a drag sled, equipped with a BF Goodrich tire, weighing precisely 10.9 lbs. The pull force was measured using a Mettler Toledo 100 lb load cell, with a sensitivity of 0.001 lbs. The drag sled was pulled across the test surface, using a constant velocity motor, at a rate of approximately 30cm per second. This allowed for data collection of approximately 30 dynamic force measurements, as the sled was pulled over an approximate one meter distance, at a sampling rate of about 10 measurements per second. Each set of force measurements was averaged to determine the pull force for each test run. The available friction for each test run was calculated from this average pull force and the weight of the drag sled. Importantly, the data was collected at a drag sled velocity of about 1.0 kph (0.6 mph). (Back to top)
Earlier research showed friction was also velocity dependent
Between tests, the drag sled was removed and triple washed and triple rinsed, as was the asphaltic surface. At the start of each test, prior to the introduction of the particular liquid anti-icing agent which was to be tested, a set of 'dry runs' was performed, and the dry friction value of the drag sled on the asphalt surface was verified to monitor the condition of the test apparatus. It was thought this would assure that the test set-up was identical for all tests. As in prior research, a baseline pure water test was performed at the beginning of the experiments, and at the end.
For each test, the anti-icing chemical was applied at a rate of 60 Liters per lane kilometer (25 gallons per lane mile), using a pump spray mister. This method of chemical application was implemented to model the chemical's distribution on an actual roadway by traffic.
Part II will focus on the testing of each chemical. (Back to top)
About the authors: Tim Leggett, P.Eng., senior engineer and partner with Forensic Dynamics Inc., a firm which specializes in motor vehicle accident reconstruction, product liability, slip and fall analysis, fire investigations, engineering design and animations. He has more than 15 years of accident reconstruction experience.
Gerald Sdoutz, E.I.T., is an engineer-in-training with a bachelor of mechanical engineering, as well as a bachelor of biological sciences. He will be qualified as a professional engineer this summer, and is co-author of the latest report on anti-icing chemical friction issues on asphalt.
Forensic Dynamics Inc. has been involved in "slickness issues" relating to winter road liquid anti-icing campaigns since 1991. As of 1998, the firm has been involved with a number of skid tests, in order to fully understand these friction issues. In 1999, the firm constructed what is thought to be the world's first indoor, climate controlled, friction test facility to accurately measure friction characteristics under various temperature, humidity and application rates. Tests are being performed for chemical manufacturers, distributors, state and county agencies, and the U.S. Federal Highway Administration.
Edited by Joyce Everhart
Seventeen chemicals were tested in this data set. These included:
It has been established that there is a dependency on humidity for anti-icing chemicals, with regards to their friction capability in a previous test performed in the lab using an of an etched glass surface. Since asphalt is the preferred choice for most driving conditions (as opposed to a sandblasted glass surface) it was requested that similar experiments be performed on this surface, in order that users might have a better understanding of what friction they could expect to find.
Based on some preliminary computer modelling, Wilf Nixon, PhD, of the University of Iowa, concluded that the likelihood of a slickness incident occurring was less than 1/1000 of 1% of all liquid chemical applications. There have been other reported reasons for slickness occurring, or conditions upon which slickness was thought to have occurred. These will be discussed later on.
All testing was performed in a climate controlled test chamber. The test surface was a 1.5 meter long, 0.3 meter wide, section of asphalt removed from an arterial road in Kamloops, BC, after approximately 15 years of service. The climate controlled test facility was used to set the test temperature at a constant 5°C and alter the humidity values, to permit the applied anti-icing chemicals to dehydrate, and subsequently re-hydrate on the asphalt surface. Such movement, between liquid and solid states, and back to liquid form, cannot be controlled in the real world, hence, the environmental chamber was necessary to fully modulate these transitions.
Prior research has shown that friction is velocity dependent. The velocity dependance of friction will be discussed later in this report.
Managing Editor, Public Works Online
jeverhart@vertical.net