I guess from time to time when I'm asked, "When are you really going to completely retire?" (admittedly, I'm asked it a lot these days) I say, "But there is still so much more to be learned." And perhaps not surprisingly this includes, among other things, a subject that has been written about and discussed for many years plastic or drying shrinkage.
Drying Shrinkage

Carl Bimel, FACI
Consultant, Industrial Concrete Floors

Although there is little doubt but that all types of shrinkage, and certainly this type, have apparently always been a major area of concern for all types of concrete construction, I found out a long time ago that it is certainly true for concrete floors. And, as often is the case, it is not just shrinkage alone that must be considered, but how it relates to other inherent characteristics of all concrete.

Unfortunately there is little doubt but that when most people—probably everyone but finishers—hear about any problem that can in any way be related to shrinkage, they almost automatically think of slump and only slump. This in spite of the fact that concrete delivered to jobsites contains not less than three and typically more products and/or materials that were manufactured or processed to conform with completely different ASTM's, all of which invariably have a "wide window of acceptance." Sure slump or water content is important, but it should be considered only as one of many other items.

"Being over 80 years of age you might expect me to relate to older (some would consider ancient) history, but I personally believe, in this case, it is exactly the place to start."

So first let's look back and I mean way back to just what information is actually already available. As long as 50 to 60 years ago, men like Treval Powers, R.C. Meininger, and Tremper & Speliman had completed extensive research on the specific subject of shrinkage. As a matter of fact, I think it is obvious that Bob Ytterberg made great use of the work of the above-mentioned authors when he wrote his series of articles entitled Shrinkage and Curling of Slabs on Grade that was published by Concrete International in 1997.

In fact, for some time I had felt that Tremper Bailey & Speilman's Shrinkage of Concrete Comparison of Laboratory and Field Performance Highway Research Record No. 3, published by the Highway Research Board in 1963, was actually the best source for completely unbiased information available on shrinkage.

This well-documented study was not only specifically designed to learn what items cause or contribute to shrinkage, but at the same time determine the relative percent of input each contributed to the eventual total or combined potential for shrinkage. Although most of the causes are obvious, I do believe some people will be a little surprised at the relative impact of some specific items. For instance, I believe most people recognize the importance of having good, clean, sound, well-graded aggregate, but I doubt if they actually realize that this one single item can have the potential to contribute to more shrinkage than any other listed item. But perhaps the biggest surprise of all is the relatively low increase in shrinkage if a 7" to 8" slump concrete is used in lieu of a 3" to 4" slump concrete, and the fact it still has considerably less impact than a number of other items, including portland cement.

Perhaps one of the most interesting and informative articles on the subject of slump was published in the April 2000 issue of Concrete Construction. This article, entitled A New Look at Water, Slump & Shrinkage, described and indicated the results of a series of fully accredited laboratory tests made with three different quantities of portland cement, three different commonly used admixtures, and the required water necessary to obtain 3", 5", and 7" slumps 30 minutes after discharge from the mixer. The predetermined slumps were obtained by varying the actual water content and the type and dosage of the admixture. Cores were then taken for all 33 mixes and were tested in accordance with ASTM C-157 for shrinkage in 28 days, 90 days, 180 days, and 365 days. Admittedly all types of different assumptions could be made from the final report, but I believe it is certainly significant that the editorial in the issue was entitled It May Not Be The Water.

Now, let's look at cement, an often forgotten product in the equation.

First, it should be recognized, as indicated in the July 1996 issue of PCA's Concrete Technology Today, that some of the differences in cements manufactured in the 1950's and those manufactured in the 1990's could have a negative affect on shrinkage.

Although I have always felt it is important to know a lot more about many products used in concrete than the fact that it meets its appropriate ASTM; it is certainly true with portland cement, as well as all of its ingredients. In fact, although the aforementioned PCA report attempts to minimize the differences between cement manufacturers in the 1950's and 1990's, you'll note the changes that have taken place are the very items that not only could contribute to shrinkage, but in practice make it more difficult for today's finishers to produce really good floors. As difficult as it might seem, it is important to learn as much as you can about the set time, strength, and alkali content in the cement you will or probably have to use.

For many years, Richard W. Burrows (2001 Wason Award winner for his article Visible and Invisible Cracking of Concrete has been trying to get the industry to consider manufacturing a coarse-ground portland cement with less C3A and less C3S, but unfortunately there apparently is just not enough demand in the marketplace for such a product. Although there is little doubt but that the availability of a portland cement, like Dick has in mind, would go a long way towards helping us to not only get better floors, but certainly floors that would have less potential for shrinkage. Unfortunately, the cement manufacturers don't, for all practical purposes, consider concrete for floors as a top priority.

There obviously is significantly more demand these days for cement for highways and multi-story construction where the emphasis is on high strength and early high strength.

Although there are a number of other items that can or certainly eventually will contribute to shrinkage, I like to refer to three things which I believe, one way or another, have contributed to this problem of shrinkage:

1. First, I hope the A/E's or their specifiers will recognize what is in the current ACI 301 (Specification for Structural Concrete) and the current ACI 318 (Building Codes), recommending how water/cement ratios should be selected. The current documents now recommend that instead of having an arbitrarily selected water/cement ratio dictate a required strength, the required strength for all practical purposes will determine the water/cement ratio. For additional background on this subject, please don't hesitate to refer to ACI 302 (Guide for Concrete Floor and Slab Construction) 6.2.3, as welt as ACI 360 (Design of Slabs on Grade) 9.5, regarding unnecessary and undesirable high strength. In fact, ACI 302 actually suggests that since there is no direct correlation between compressive strength and water/cement ratios, the two should not be combined in a specification for interior floors. There is little doubt but that unnecessary high cement content will increase the potential for shrinkage.

2. Second, is my concern when a question comes up regarding "durability" for an interior floor or slab. Although this is somewhat addressed in the previous paragraph, I believe it is very significant that in the current ACI 301 document, Part 3, Chapter 4, you'll actually see that durability is related to freeze/thaw conditions. I do recognize that the term "durable" is often referred to in conjunction with a good floor, but I believe it should be recognized how ACI 201 (Guide for Durable Concrete) actually addresses this situation. Although there is considerable copy in Chapter 1 (Freezing and Thawing) and Chapter 2 (Aggressive Chemical Exposure), the only reference regarding concrete floors, and even here a reference is made to ACI 302, is in Chapter 3 (Abrasion) regarding water/cement ratios which states "the surface can be improved by minimizing surface water/cement ratio, i.e. forbid any water addition to the surface to aid finishing." This caution is certainly understandable, but it should be also recognized that this comment, for all practical purposes, relates only to the finishing techniques employed and in no way to the base concrete.

3. My third concern is the fact that almost invariably when the first crack appears, shrinkage is first mentioned as a possible cause. It is true that shrinkage can be involved, but it should be recognized that ACI 224 (Cracking, Evaluation, and Repair of Cracks in Concrete Structures) actually says that most cracks are not the result of shrinkage, but the result of restraint to shrinkage. And it should be recognized that quite often it is very difficult, if not impossible, for the concrete contractor to eliminate all restraints and still comply with the engineer's designs. Obviously this is an item that should be completely resolved by all parties before construction begins.

For many years I've had a small plaque on my office wall with this inscription, "The Real World Concrete is a truly unique product with "built-in" inherent characteristics which unfortunately are adversely effected by countless conditions and variables, many of which the engineer or contractor have complete control." I'm frankly not sure where I came up with this message, but the complex problems related to shrinkage certainly fit in here somewhere.

I, of course, realize that it is literally impossible to read all articles, papers, studies, and yes, creditable test reports related to shrinkage, even just plastic or drying shrinkage, but 1 hope you will find time to review the completely unbiased research work done in the 1950-60's and then read the copy in the current issue of ACI 201, 302, and 360. Frankly, I believe many of you will be surprised.

About the Author:

Carl Bimel
ACI member, Carl Bimel, a consultant in Cincinnati, Ohio, has been involved in concrete specifications and construction for more than 35 years. He has specialized in design and construction of concrete flooring. Prior to establishing his own practice, he was the President of Maximent Corp., Cincinnati. He is a fellow, chair of ACI Committee 302, Construction of Concrete Floors, and 223, Expansive Cement Concretes; and 360, design of slabs on grade.


Back to ConcreteNews

© 2001 L&M Construction Chemicals, Inc. | ConcreteNews Fall 2001.

Subscribe to ConcreteNews