Polymer-modified cement, mortar and concrete Durability

Water absorption
When comparing PMC of given type produced at constant workability, it is found that the water absorption and permeability are reduced as the p/c is increased and this may be attributed to reduced porosity. However, there is some loss in strength, especially flexural, although this is not usually considered to be of practical significance as strengths are still greater than for unmodified mortar.
In keeping with these observations, the frost resistance of PMC is much improved. Thus, whereas an unmodified mortar may suffer damage in fewer than 100 cycles of freezing and thawing, e.g. ÿ18 to ‡4ëC, there is no change with the PMC even after 300 cycles (Ohama, 1998).
Carbonation
With regard to carbonation, a trial showed that, for an SBR modified mortar, carbonation depths were 2±3mm compared to 10±20mm, depending on exposure conditions, for the conventional material, (Ohama, 1998).
Acid attack
Most PMCs are attacked by inorganic and organic acids and sulfates since these agencies attack the cement phase. They are also attacked by organic solvents  which dissolve the polymer films. However since porosity is lower than is the case for unmodified mortars then rates will also be lower (Ohama, 1998).

Chloride diffusion

Values of the effective diffusion coefficients (Deff ) of chloride ions in a variety of polymer modified cements, using a well established steady-state diffusion technique (Page, 1981), have been determined (Zeng, 1996a) and are given in Table 10.4. Samples were compared at constant w/c and were all well cured. A number of interesting points were noted from the results obtained which may be summarised as follows:

· In the case of the unmodified cement reducing w/c led to reduced values of Deff , as might be expected and results are comparable with those reported in other work.
· At a given w/c, the addition of 10% SBR 1 reduced Deff by about a half.
Increasing the addition to 20% had little further effect. However, comparison at constant w/c underestimates the real benefit of adding the latex. With the ability to reduce w/c whilst maintaining workability (see Section 10.2.2), a comparison should be made between say 0.4 w/c for the unmodified and 0.3   w/c for the modified pastes. In this case the reduction in Deff is then nearly an order of magnitude.

· Most of the other types of latexes were also effective in reducing Deff but none were quite as good as the SBR. Note, however, that Ac 2 is an exception to this general observation. This polymer had a considerable retarding effect on cement hydration and consequently produced a very open pore structure which in turn resulted in a value of Deff much higher than that for the unmodified paste. Thus it should not be assumed that all latexes will be beneficial.
· Whether the polymer was added as a redispersible powder or latex made little difference to the values of Deff recorded.

Claims that PMCs of the types studied may have improved resistance to chloride ion penetration due to the large pores being filled with polymer or sealed with a continuous polymer film are unlikely to be correct since, if anything, the polymers tended to increase porosity and the polymer films, when tested separately, were not effective chloride diffusion barriers in themselves.
The values of Deff through polymer films produced by drying in a Petri dish were found to be 8.65 and 11:35  10ÿ8 cm2 sÿ1 for SBR 1 and AC 1 respectively (Zeng, 1996a).

As discussed earlier, the wet-dry curing regime is not one that would normally be considered appropriate for good durability and the existence of a CAZ has been shown. Slices taken from such a sample do indeed show a larger value of Deff in the CAZ, Table 10.5 (Salbin, 1996). It should be noted, however, that this value of Deff decreases over time as further hydration of the cement occurs due to wetting in the diffusion cell. This finding may account for anecdotal observations of higher chloride concentrations found at the surface of some PMC structures but with little penetration of chloride in depth due to the rapidly reducing porosity away from the evaporation surface.

From a practical point of view, these results show that when comparing different systems it is very important to clearly state the basis on which the comparison is made, i.e. constant w/c or constant workability. In addition the fabrication and curing procedures used in preparing samples for the diffusion

test should be clearly stated and compared with the recommended procedures when the material is used in practice.
Thermal/fire
The thermal resistance and incombustibility of PMC are not as good as those of unmodified material. Mechanical properties deteriorate rapidly with increased temperature, their resistance being governed by the nature of the polymer used especially glass transition temperature and the p/c. Incombustibility depends on the nature of the products formed during thermal decomposition. These factors tend to restrict the use of PMC to below 100±150ëC (Ohama, 1998).