By Larry Roberts and Terry Harris
Everything that goes into our concrete mix includes some sort of documentation. It could be a certification, a mill test report, test record or test data, statement of conformance – and the list goes on. Some records simply state adherence to an applicable standard, but some contain data that can be overwhelming.
When we buy cement, we are provided with a cement mill certificate by the cement producer. Cement mill certificates, also called mill test reports, contain a lot of information that can prove useful for tracking changes in your concrete and controlling variability in your mix. Usually this certificate must be included in submittal documents to show that the cement conforms to the requirements of applicable specifications, such as compositional or performance limits. Unfortunately, the other information in the document is often not used and the certificate is simply filed away, yet reading and tracking that additional information may be helpful in achieving smooth operations in the precast industry. We’ll take a look at some of the more useful information for precasters.
What’s in it for me?
First, let’s discuss what a mill certificate is and what it is not. ASTM C150/C150M, “Standard Specification for Portland Cement,” provides guidelines for such certificates along with an example. However, each cement producer has its own format, often supplying more information than the example shows. We will focus on portland cements for this article, and you may want to refer to the example on page 8 to follow the discussion.
At the top of the example certificate, you will find the cement type. Due to overlaps in requirements and strength levels typically well above specification limits, some cements meet several types. This may show in some mill certificates as “Type I/II” or in extreme cases “Type I/II/V.” Sometimes the same cement will have more than one certificate when different designations are used, depending on which type the customer ordered. It is useful to consider the cement type that is more restrictive.
Both compositional and performance information is provided in the CHEMICAL table in the left column and the PHYSICAL table in the right under the Standard Requirements section. Some of the performance information for long-term tests may apply to cement produced earlier than the time period covered, and this will be noted as shown in the example for the results of C1038 testing at the bottom of the PHYSICAL table.
Further, although ASTM explicitly details the test methods and analyses used to develop the data, there can be significant differences between producers in how often they are collected and the time periods represented. It is important to understand that the certificate is not an analysis of a particular shipment of cement (unless required by the purchaser). This is generally the case except in very large projects, as it would require extensive storage capacity and delay shipment until after the test results are obtained.
The usual certificate reports the average composition and performance over some specified range of time. In the ASTM C150/C150M example, the range is given as a seven-day period. In practice, mill certificates representing a monthly average are typical, sometimes a calendar month, sometimes merely stated as “monthly basis” without further clarification. Sometimes they are reported on a “lot” basis with no clear time period. It is best to understand exactly what type of average is represented by talking with the producer if it is not clear on the certificate.
In any case, the results reported are for the average of a large number of different samples analyzed and performance-tested over the time period chosen. Thus, normal, day-to-day production variation is averaged out, and a certificate gives no indication of such variation. Also, the samples tested on each day are usually physical composites of several samples taken over a production period, such as a shift or the full day. In some cases, these may be as frequent as hourly, meaning that the physical tests could have been done on a blend of 24 different samples, while the chemical results would be a mathematical averaging of each hourly sample. The mill certificate number then comes from an averaging of all the days in the time period. Logically, then, no single sample – be it a small sample for laboratory testing or a 30-ton bulk delivery – will have exactly the composition or performance reflected by the average values in the accompanying certificate. To make this even clearer, at the height of the production period, generally in the summer, the cement actually shipped may be only one or two days old, yet the certificate shows seven- or even 28-day strength, based on historical production data. So a fair definition of the certificate is:
“A summary of the averages of tests (physical and compositional) of cement of the type indicated, tested over some defined period.”
If the certificate represents average cement testing, what good is it? Actually a lot, but it takes effort to gain the value of the information. First, let’s state the obvious: There are no concrete tests reflected in the certificate, no tests with admixtures, no tests with supplementary cementitious materials (SCMs), and no tests at other-than-laboratory temperatures. It can’t tell us everything we would like to know, but it can give some strong hints if we track the information regularly.
It is highly recommended to use simple control charts on which the variables discussed below are plotted each time you receive a new certificate. If concrete testing results and mix design information are plotted on similar charts, relationships may surface that are too complex to discuss here.
Material variation is the bane of concrete production. If nothing ever changed, production would be easy – but things do change. While variation within the time period will not be reflected in the certificate, change over a longer period will be. After all, the cement manufacturers are challenged to offer a consistent product from materials that are dug out of the ground and by their very nature must be variable. To do that, mixture proportions of the raw materials must be continually adjusted. With such multi-component raw materials, it is impossible to keep all composition factors the same. So the cement characteristics will vary, as will the characteristics of all concrete components.
Let’s list a couple of important variables and track how we can use the information to know when something significant has changed.
Alkali content is shown in the example as Na2O(%) and K2O(%) in the CHEMICAL table under the Standard Requirements section, and as equivalent alkalis (%) in the CHEMICAL table under the Optional Requirements section.
Equivalent alkali is the summation of the sodium oxide and potassium oxide content of the cement, expressed as a percentage of mass equivalent sodium oxide (often abbreviated Na2Oeq), by making a chemical calculation based on an equal number of molecules of sodium and potassium oxides. It is not required to be reported if the cement has no alkali limit, but most certificates report it. Even if the cement is not low-alkali cement (below 0.60%), the amount is important for three reasons:
- Air-entraining agents generally produce more air as the alkali content goes up. Where the alkali occurs in the cement can affect this, but to get a first approximation, we need to track the equivalent alkali content. If a change is noted in the mill certificate, air-entraining agent dosages may need adjustment. Above 0.60%, alkali changes of 0.1% can be significant. If the cement is quite low in alkali – say 0.30% – even a 0.05% alkali change can be important. It is impossible to say that such variation will definitely have an effect, but just by tracking it you can judge whether adjustment is necessary.
- Alkali can impact the reactivity of SCMs, especially Class F fly ash. Alkalis in the cement tend to raise the pH of the mix water. This attacks the reactive glasses in the ash and allows it to react chemically, improving the concrete strength and lowering the permeability, both of which can improve durability. A substantial decrease in alkali content may affect the strength performance or durability. Frequently, the optimum fly ash dosage will decrease as alkali content is reduced and there is less alkali available to attack the glass.
- When alkali-silica reactivity of aggregates is a concern, changes need to be tracked carefully. Concretes using low-alkali cements can still suffer from this if the concrete has relatively high cement contents, depending on how reactive the aggregate is. Cement content times alkali content provides the “alkali loading” of the concrete.
Along with chemical reactivity of the cement composition, cement fineness controls early strength development. In fact, most Type III cements today are merely higher fineness versions of the regular Type I, II or V cement produced at the same plant. In the search for strength, market forces have driven cement fineness higher for all types over the past several decades. The Blaine fineness (shown in the PHYSICAL table of the example) is determined by how fast air moves through a compacted pellet of cement, and thus provides an overall average measure of fineness.
Tracking the fineness can be useful in several ways:
- As fineness increases, concrete water demand generally goes up and may need adjustment. Water-reducing admixture demand may follow the same pattern.
- Air-entraining admixture demand will tend to go up with fineness, so adjustment may be necessary.
- If a substantial difference is seen in fineness but the strengths have remained the same, it is indicative of some change in the cement production. It may be related to clinker chemistry, kiln operation, clinker cooling, finish mill operation, use of a secondary clinker source, sulfate content, or a completely different cement source.
- Many things can change with fineness. For instance, the seven- to 28-day percentage strength gain experienced in the past may decrease as fineness increases, as more cement will have reacted within seven days. Stated another way, the early strength may be higher, but the ultimate strength may be reduced.
But it’s not our intent here to explain why and how all these things could relate, or all possible results; the important point is that a change in cement fineness signals a potential change in concrete performance and needs to be understood. The best place to start is with a discussion with the cement supplier.
In the next issue of Precast Inc., Part 2 of this article will cover cement compound analyses, such as C3S and C3A.
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.
ASTM International for permission to use the example from C150/C150M
Dr. Paul D. Tennis of the Portland Cement Association for his valuable comments