‹ Back to articles

Can cement be flexible?

Can cement be flexible?
Martin Cummins

Published by Martin Cummins - UK Technical Support Manager, Bostik

Martin has been involved in the industry for over 20 years. He’s passionate about flooring in all its guises, with his particular expertise being in subfloor preparation products and adhesives. Martin is an active member of the CFA manufacturers committee and he sits on the CFA council. He aims to keep abreast of all the changes in the industry and feeds this knowledge back to installers to make sure when floors are laid they perform to their maximum potential.

As part of standardisation with our neighbours in Europe, there is a requirement on a lot of products to either get them tested or to test in-house and declare the various properties. The idea is that, by doing this, all manufacturers will be on a level playing field with their product claims, which will help the end user evaluate what they need.

This is particularly prevalent in the tiling market where there is legislation to get external testing carried out and performance declared. The adhesives used for bonding or fixing tiles are then classified according to various characteristics, such as whether they are cementitious (C), dispersion (D) or resinous (R) products and whether they have any special performance criteria such as fast setting (F), thixotropic (T) etc. They also have a classification of deformability, often referred to as elasticity or flexibility. All the classifications are given based on strict test regimes and conditions and enable performance differences to be appreciated between products. A tiler can then know what adhesive he requires for a particular project based on the needs of the tiles and the type of substrate.

When it comes to professional flooring there is no specific criteria for either smoothing compounds or adhesives. My belief is that this is because the scenarios and criteria for flooring is far more variable. When installing tiles, it is known that a strong, sound substrate is needed otherwise tiles will crack under movement. Furthermore, a bed of adhesive can be used if the substrate isn’t sufficiently flat, so the classification of adhesive is much more relevant.

In flooring we generally need to improve or prepare the subfloor to receive the floor covering by using a smoothing compound. Although there are test methods for assessing smoothing compound performance, there is no actual standard to test them to. As a consequence, the industry uses the testing of “screeds” to enable some sort of classification to be offered. You may have seen on datasheets or bags of smoothing compounds classifications such as:

CT-C18-F7 to EN 13813:2002.

Have you ever stopped to think what these are actually saying? Most people think, correctly, that the C refers to compressive strength (measured in N/mm2) which is a test that assesses how much of a load can be applied to the sample before it cracks. The sample is supported underneath during the tests so this is meant to imitate the ability to resist loads on the smoothing compound from above. Compressive strength can loosely be correlated to how hard wearing a subfloor is. A cementitious screed is typically 25 to 30N/mm2, whilst concrete can be anything up to 50N/mm2 depending on the construction needs.

The figure quoted is based on the achievement at 28 days under lab conditions and does not imply anything more than this. Equally important for flooring is the speed of strength build up as this will determine when the floor can be subjected to loads. A typical 18N/mm2 smoothing compound may only be 5N/mm2 the day after installation, but this is not a problem for walking on or laying flooring. It may take 7 days to achieve a strength of 12N/mm2, which may be required to provide support for heavy point loading such as hospital beds, forklift or pallet trucks, or heavy plant such as computer servers, whereas a high strength product may achieve this after 24 hours. Not all products are the same so always do your research.

The reference for the F aspect above is often misunderstood, with most people thinking it measures the flexibility of a product. The actual test is called a Flexural Test and involves an unsupported sample of product effectively trying to bend until it breaks. A high flexural strength basically means a product that needs a high load to break it. It doesn’t give any indication of whether a product is flexible in the true terminology, i.e. able to bend and distort. In fact, I can’t really come to terms with cement being flexible… it is a hard material that will crack if you try and bend it.

When we refer to flexible products we are looking at products that can stay bonded or may have micro cracking rather than visible cracks when under deflecting loads. Such situations may be on timber floors, metal panelled floors and even over underfloor heated systems.

Time and experience has shown us that lower compressive strength (and typically these are lower flexural strength also) products are the best option because they don’t “sheet off” the floor. In fact they will have lots of micro cracks but still retain the adhesion. Fibre reinforced products work by ensuring that any cracking will be somewhat held together by the fibres, resulting in micro cracking that prevents sheeting off the floor… they should not be considered as ‘flexible’ in the true sense as they will not bend and distort. In the tiling world these products are prevalent and very successful, but remember that subfloors for tiling need to be stable and sound regardless, hence 12mm upwards of Plywood when sheeting over floors.

I believe the old adage of not putting hard over soft is still a very valid rule of thumb for flooring. In the UK we limit preparation opportunities so are often overlaying previous installations, which are highly likely to consist of relatively low strength products or perhaps adhesive residues.

So the moral here is please don’t think of cement as having high flexural strength and therefore being flexible. Always assess what the product can do. Is it low strength offering excellent adhesion? Will it build strength quick enough for my application? Is priming necessary to give the desired bond? If in doubt phone technical support armed with all the information on the project and they can offer you a resolution.