By Larry Roberts and Terry Harris
In this second part of a two-part series, we pick up the discussion from the July-August 2013 issue of Precast Inc. As you recall, we discussed what a mill certificate is, and how the data are gathered and reported. We also talked about the importance of tracking the results to control variability, and reported on compositional tests such as alkali content and Blaine fineness. In case you missed that discussion, you can find it on NPCA’s website at precast.org/cementmill.
Next in our discussion comes an alphabet soup of C3S, C3A and SO3 plus some of the performance tests. Refer to the example cement mill certificate on page 8 to follow the discussion(1).
C3S and C3A, cement phase composition
C3S (tricalcium silicate) is the major strength-producing phase in concrete curing and dominates through the first 28 days.
It would be tempting to say higher amounts of C3S should yield higher strength, especially at an early age, but it is more complicated than that. These amounts do not reflect differences in reactivity that are determined by minor element inclusions in the crystals and the heat history in the kiln. So while the C3S quantity is useful to the cement chemist who is tracking a specific cement and for determining specification compliance, a comparison should not be made between different cements on this basis alone.
It is useful to track the C3S, however, and note any step change in a given cement. Sometimes the C3S level may go down and the cement fineness may go up, or vice versa. These variations occur when the cement producer balances one factor with another to keep the performance as consistent as possible, and is to be expected. Large variations can have other effects, though, as discussed under fineness.
C3A (tricalcium aluminate) is the most highly reactive element in cement, and its hydration components are changed in sulfate attack. This is why C3A is limited in Types II and V cements. Calcium sulfate (CaSO4) is added to cement to control this high-reactivity component (SO3 as indicated on the mill certificate). C3A is not limited in Type I cement, so when a producer labels a cement mill certificate as “Type I/II” it really means it is a Type II cement that also meets the Type I requirements. The only restrictive difference between Type I and Type II cements is that Type I has slightly higher three-day and seven-day strength requirements, but as they are typically well below current market-determined strength levels, all Type II cements are essentially Type I as well.
What does C3A tell us that we should track?
- C3A content can be an issue when early strength is important, as high-C3A cements tend to be somewhat more reactive at an early age. A substitution of a Type I/II for a Type I must be approached with caution, because it is likely, although not inevitable, that early strength could be lower (even though they are nominally both Type I). This is where the mill certificate becomes useful. A change of around 1 to 3% in C3A is unlikely to have a strong impact caused by this difference alone, but a switch from, say, an 11% C3A Type I to a 4% C3A Type I/II would be in the direction of lower early reactivity and should be carefully tested to assure that the performance for the application is acceptable. One example to consider in this case is sufficient strength gain for form removal.
- Normal C3A variation in a cement plant should not have a significant impact, but sometimes cement from the same plant can change. For example, purchasing clinker from another plant to make the cement may change the composition. In this case, the charting of the mill certificate data can prove to be a useful alert that early strength performance should be checked.
- The C3A level is also important in determining admixture dosage. Through some fairly complex chemistry, rapid-reacting C3A hydration products can absorb many admixtures during early hydration. This is the source of the well-known “delayed addition effect” where admixtures, especially superplasticizers, can be made more effective by adding them later in the mixing sequence. This happens because it takes a few minutes for the sulfate in the cement to dissolve and control the C3A reaction. Changes in the C3A level of a cement can influence this effect, so it’s helpful to understand the normal behavior of the cement being used. If a significant C3A change is noted, as in several percentage points, rechecking the admixture dosage requirement can be useful.
In some mill certificates, the other two major compounds – C2S (dicalcium silicate) and C4AF (tetracalcium alumino ferrite) – are listed as well, but since there are no specification limits on these, it is not required. These are often referred to as “potential” compounds, calculated by analyzing the elemental composition and assuming idealized chemical reactions in the kiln. Although this approach has been used for more than 70 years, more extensive testing has shown that these numbers are approximately, but not exactly, correct.
SO3, inorganic processing data
SO3 (sulfur trioxide) is an indirect measure of the amount of gypsum or calcium sulfate (CaSO4) in the cement. As discussed above, SO3 helps control C3A reaction. The amount analyzed consists of both the sulfate (SO3) in the cement clinker and the amount added during grinding, typically in the form of the mineral gypsum. A few points are in order:
- ASTM C150/C150M limits the amount of sulfate that can be added to cement of different types. In some cases, these default sulfate limits are too low for some concrete applications – for example, when the cement is used with certain admixtures or combinations of admixtures, with admixtures at high dosages, with Supplementary Cementitious Materials (SCMs) containing reactive alumina, and/or at higher temperatures. However, a cement producer cannot know in advance all of the mix designs and materials that a cement will be used with, and durability issues can arise with over-sulfated cements. Volumes have been written about this phenomenon. If the concrete system is under-sulfated, either a runaway setting or a severely retarded setting can result. In either case, significant depression of early strength results can occur. With recent changes in the ASTM specifications to allow higher levels of sulfate, so long as durability is maintained, this situation will be even rarer. But monitoring SO3 levels on mill certificates can help identify potential sources of problems, particularly if problems arise in concrete production when new materials or combinations are being used (cement, SCMs or admixtures).
- Since SO3 is present to control C3A, in a given cement the ratio of one to the other will normally be fairly constant. It is useful to plot this ratio, and if you note significant changes, ask the cement producer about the reasons for the change. If the ratio of SO3/C3A drops significantly, say two to three times more than the normal lot-to-lot variation, beware of the setting issues described above.
The cement mill certificate includes three performance tests that merit close attention: false set, Vicat setting time and strength.
False set. False set is the tendency for cement mixtures to stiffen early, typically due to the sulfate in the cement reacting with the water to form gypsum crystals over the first few minutes of mixing. Because these are weak and can be broken up or redissolved with additional mixing, the term “false” is applied. The number reflects how much residual workability is present as measured by the difference in paste penetration resistance before and after two mixing steps at a specified time. Thus a higher number is better (a 50% minimum is specified in ASTM C150/C150M).
This can be very important in the precast environment, because the short mixing cycles frequently used may allow the false set to occur after discharge, making placement without agitation (vibration) difficult or impossible.
False set may also be influenced by admixtures. Some admixtures can delay the onset of the gypsum precipitation, causing false set to occur where otherwise it would be prevented by normal mixing. Accordingly, tracking the false set number – and especially relating it to plant water demand records – can be very useful.
As we have seen with other parameters, it is very difficult to compare cements from separate sources on this basis, but tracking values for a cement over time can provide valuable information. The standard test method (ASTM C451) involves neat cement paste without SCMs or admixtures that may affect concrete performance. Nevertheless, if an increase in plant water demand corresponds to a decrease in the false set number, false setting may be occurring. Problems of this type usually can be overcome, sometimes by only a few seconds delay in admixture dosage or lengthening the mixing time. But early strength, and therefore plant efficiency, can be strongly impacted, so close monitoring is warranted.
The false set test is an optional requirement, but it is often reported anyway. If not, you should consider requesting it.
Vicat setting time. Vicat setting time is a measurement of how many minutes it takes for a concrete sample to set, as derived from a Vicat testing apparatus. Specification limits range from 45 to 420 minutes, and it is very unusual for a cement to be close to either extreme. The Vicat test (ASTM C191) is performed on a patty of cement paste at a very low water-to-cement ratio – typically in the range of 0.25, quite different from normal concrete. For that reason, while it is an indicator of concrete setting behavior, it should not be taken as a direct predictor. Cements with similar Vicat setting times can have different setting times in concrete, and conversely cements with different Vicat setting times can be similar in concrete. Moreover, since the Vicat test on plain cement paste does not capture any admixture or SCM effects, with today’s almost universal use of SCMs and admixtures, concrete results can vary significantly. Again the emphasis is on change. Plot the results on a control chart, and if a significant change is seen, something has changed in the cement and you should be aware that concrete setting characteristics may change.
Strength. A number of factors limit the utility of strength test results on a mill certificate.
- Types I, II and V cements require only three- and seven-day strength test results. Most mill certificates also report 28-day strength, but usually this is with the note that it is from an earlier time period or lot, for reasons discussed previously. Type III cements also require one-day strength to be reported. However, the specification requirements are generally significantly below actual strengths achieved, so they have little direct impact.
- ASTM C109/C109M strength testing for portland cements is done at a constant water-to-cement ratio for a specific mortar composition, so the results do not reflect concrete strength performance differences that may change due to the water demands of the concrete.
- With the wide use of admixtures and SCMs, concrete strengths may be determined in large part by the response of these materials to the cement characteristics and to each other. Clearly, the impacts of other concrete parameters (for example, mixing intensity and aggregate characteristics such as nominal sizes, strengths and gradations), cannot be captured by C109 mortar testing.
For these reasons, it is not recommended to compare cement strengths of different cements from mill certificate results for selection purposes. The proof is in concrete, in the mixtures used. But again, changes in the reported cement strength of a cement being used may be reflected in the concrete, and therefore should be put on a control chart and tracked.
ASTM also provides a better means for a cement company to communicate the actual variation of the cement strengths at seven and 28 days (ASTM C917, “Standard Test Method for Evaluation of Cement Strength Uniformity from a Single Source”). If a cement producer can supply this test report, it is very useful, because the samples used are grab samples, not composites, and many are taken each month. The test results give a good idea of how much a particular cement’s strengths vary on a short-term basis and thus may be more useful than the standard mill certificate. Several state DOTs require these strength reports, so they may well be available even if you have not requested them.
Obviously, the relation of cement composition and physical properties to concrete performance is a complex subject. We cannot expect average numbers like those on mill certificates to tell us everything, but it is useful to pay attention to them. Many field problems can be avoided by requiring the review and plotting of mill certificate data every time a cement shipment is received, and by being aware of the potential effects on concrete performance.
Larry Roberts is owner of Roberts Consulting Group LLC, Acton, Mass. Terry Harris, of Green Cove Springs, Fla., is manager of Technical Services, North America with Grace Construction Products, Cambridge, Mass. The authors invite comments and questions at any time.
(1) A copy of the example mill certificate, courtesy of ASTM International, is available on NPCA’s website at precast.org/certificate-example. ASTM C150/C150M is available for purchase from the NPCA Shop by calling (317) 571-9500 or toll-free (800) 366-7731.
ASTM International for permission to use the example from C150/C150M
Dr. Paul D. Tennis of the Portland Cement Association for his valuable comments