Practice has tended to differ from that of the UK, with pfa generally finding use as a component of cement, or considered k times as effective as cement (with k ranging from 0 to 0.5) when added at the concrete mixer.
Material with a wider range of fineness has also, in the main, been permitted.
The new European standard BS EN 450 for pfa has, as a result, included limits on material quality which are likely to have implications for UK practice.
The fineness has been set at 40% retained on a 45 µm sieve, compared to the BS 3892, Part 1 value of 12.0%, while the loss-on-ignition (LOI) has reduced from 6.0 to 5.0% (with 7.0% permitted nationally).
The other main change is in the control of material quality, with the requirement that pfa fineness should be within 10% of the supplier’s declared mean.
For most other properties, only relatively minor differences exist between the two standards.
The full impact of BS EN 450 will be felt by the concrete construction industry in 1999, when the European Standard BS EN 206 for concrete, which calls up pfa to BS EN 450, is introduced, making it possible for engineers to more readily apply the new pfa standard. This paper describes the work of study carried out at the University of Dundee, specifically aimed at addressing the technical and practical issues associated with the use of pfa to BS EN 450 in structural concrete.
Influences on concrete strength
A wide range of materials were considered to fully examine the influences of BS EN 450 pfa on concrete performance. The work included pfa covering, or just at the limit of, the BS EN 450 fineness and LOI ranges, from various UK sources. The Portland cement (PC) used was of class 42.5N to BS 12 and the aggregates of 20mm maximum size and sand to Zone M of BS 882.
The initial series of mixes was used to assess the effect of pfa fineness and LOI on concrete strength and covered the range of binder contents 250 to 550 kg/m3 , with 30% pfa replacement. The free water and coarse aggregate contents of these mixes were fixed at 175 and content reduced with increasing binder content to maintain the yield.
Compressive strength tests on 100 mm concrete cubes were carried out at ages up to 28 days after curing in water at 20˚C.
Fineness
A comparison of the 28-day strengths for concretes containing pfa of varying fineness (LOI 3.5 – 5.5 %) over the range of binder contents is given in Figure 1. This shows that there was a gradual reduction in strength with decreasing pfa fineness (increasing sieve retention), the effect increasing both with binder content and concrete age. Similar results were obtained from other tests on concrete containing pfa from a single source, but varying fineness over the BS EN 450 range. Therefore, the use of BS EN 450 pfa at the coarser end of the permitted range for equal binder content concrete will lead to strength reductions, compared to BS 3892, Part 1 pfa.
Loss – on – ignition
The results from test on concrete containing pfa of the same fineness (27.0%), but varying LOI (3.5 and 8.0%) for binder contents of 250 – 450 kg/m3 indicated differences generally less than 2.0 N/mm2 between these. Hence, an increase in LOI to 7.0% allowed nationally in BS EN 450 and that in the recently revised edition of BS 3892, Part 1, is of little practical significance, provided the LOI is relatively constant.
Equivalent strengths
Given the pfa fineness effect noted, a means of achieving equivalent strength with BS EN 450 pfa was required. A number of methods exist for proportioning pfa concrete, including the k-factor and optimization methods. However, to minimize the changes required, it was decided to approach this by simple adjustment of the water/cement ratio. With the small strength changes observed with pfa fineness, modifying the w/c ratio was likely to be minor and the use of this method easy to apply in practice.
The result shown in Figure 1 were used to establish the adjustments needed over the range of strengths to take account of fineness, with examples for 5, 20 and 40% retained on a 45 µm sieve given in Figure 2. This method was used initially at a single design strength with fine and coarse pfa, and appropriate changes made to the w/c ratio via either (1) the water content or (2) binder content or (3) both. It was found for all methods of adjustment that similar strength development profiles were obtained for concretes containing coarser pfa, compared to those with finer material.
Following this approach, the mix proportions given in Table 1 were developed. These were then tested (by adjustment the water content) for strength development (100 cubes, water – cured at 20 ˚C); the results are shown in Figure 3. As indicated, this method enabled equivalent strength to be achieved between fine and coarse pfa across the range of design strengths to 28 days and the results fell within a narrow band at all other ages. Tests of concrete containing pfa of variable LOI, but similar fineness, were also carried out with these mix proportions and this was again found to have little influence on strength.
Strength test results for these concretes, cured in air, showed similar effects. Indeed, the influence of moisture loss on hydration was dominant and any effect of pfa properties was minor.
Performance of BS EN 450 pfa in concrete
Using the method described above, and adjustment the w/c ratio of the mix (through the water content) to control strength, other aspects of pfa concrete performance using material over the BS EN 450 pfa fineness range were evaluated.
For the engineering properties, work from a related study has been considered, covering a range of design strengths. For durability, mixes developed in Table 1, straddling the requirements of BS 5328 (minimum binder content, maximum w/c ratio) for the particular property under consideration, were tested. All concretes in this work were water – cured, unless otherwise indicated. The test methods used (either standard methods or techniques developed at the University of Dundee – see references) and the results obtained are given in Tables 2 and 3.
Engineering properties
The result from tests for modulus of elasticity, creep coefficient and drying shrinkage are given in Table 2. As indicated, these followed expected behavior in terms of the effect of design strength on each property. However, no influence of pfa fineness on any of these was noted.
Durability
The result from tests for the main concrete or reinforcement deterioration processes, including chloride diffusion, carbonation depth, sulfate resistance, freeze/thaw and abrasion are given in Table 2. Again, expected behavior in terms of the effects of design strength on each property was obtained, but no influence of pfa fineness was found. Tests on variable LOI pfa concrete indicated that LOI also had little or no effect on durability up to a /Loi of 8.0%
Practical issues for use
The fineness of pfa over the BS EN 450 range will influence concrete strength, with reductions for each 5.0% increase in sieve retention of between approximately 0.5 N/mm2 and 1.5 N/mm2 over the practical binder range at 28 days. The permitted ±10.0% in the standard on the supplier’s declared mean can contribute to increased variability by up to 3.0 N/mm2 and may therefore add to the design margin of concrete producers. Pfa suppliers may, therefore, decide or be required to reduce the permitted variability. In contrast, the LOI limit of 7.0% is satisfactory, with little influence on compressive strength.
Tests using the range of Portland cements produced or available in the UK, combined with pfa to BS EN 450 indicate that the trends for pfa fineness, obtained with the reference cement, are maintained with the other sources of Portland cement. In addition, the variability in compressive strength of concrete associated with a range of pc/pfa combinations, using material from different sources, is no greater than that obtained for pc concrete.
A method of mix proportioning that takes account of the effect of pfa fineness on concrete strength has been devised. It is simple and can be integrated into existing concrete mix design procedures and production practice. Other aspects of concrete performance including engineering and durability properties are equivalent for pfa concretes over the BS EN 450 fineness range, providing the concretes are of the same design strength, achieved in the w/c ratio, via the mix water content.
the authors are currently undertaking a technology transfer programme for BS EN 450 pfa funded by DETR/industry. A technical guidance document will be produced and a number of seminars held during 1998 and 1999 in preparation for the new standard.
Acknowledgements
The authors would like to acknowledge the DETR and Industrial Partners: BRMCA, Castle Cement Ltd, Electricity Supply Board (Eire), National Power plc, NIGEN and Scottish Power – Ash Sales for their funding and guidance. In addition, thanks are given to Dr B J Mage for his help during the experimental work.