Aggregates for concrete - Concrete Mix Design, Quality Control and Specification - page 56

In pure Portland cement mixes with no other material finer than 150 microns
it is clear that the water requirement will be higher with a very high cement
content, will reduce with reducing cement content to some optimum range
(perhaps 300-350 kg/m 3 ), and will then increase again with further cement
reduction. What is happening is that, in the optimum range, cement paste
fills the voids in the fine aggregate, excess cement requires additional water to
form a paste with that cement, and if there is an inadequate amount of paste,
additional water will be required to fill the fine aggregate voids.
Angular material in general has a higher void content than more rounded
material, but the introduction of finer aggregate material, whatever its
shape, may fill space that would otherwise be filled with cement paste or
water. So it can be seen that this will be beneficial when cement content is
below the optimum range and it should not be forgotten that cement par-
ticles are a crushed material of very poor particle shape.
Taking all this into account, there are no easy, universal answers to the
question of whether a particular fine material should be used.
An excellent tool for examining the properties of fine aggregates (natural
or crushed) is the New Zealand sand flow cone, as described by Harrison.
The sand flow cone is clearly suitable for examining the relative merits
of different fine aggregates and different blends of two or more of such
aggregates. However, it seems unlikely that it could be adapted to exami-
nation of the effects of varying cement content or, especially, the effects of
superfine materials such as silica fume or of chemical admixtures. The use
of rheology measurements on various proportions (after initial screening
with a sand flow cone) would appear the best method to account for all
components of the mix on fresh properties.
The proposed technique would not be as rapid as the dry sand flow test
and would require the use of a Hobart mixer or similar, so it would prob-
ably not replace the latter.
The objective should be to establish whether optimum gradings or
grading combinations established by the sand flow were still optimum
under a range of contents of cement, silica fume, and other fine materi-
als. A particularly important point would be to establish the optimum
content and fineness of material passing a 150 micron sieve in manu-
factured sand for various types of concrete (since this is an item that
could fairly readily be controlled). The test would also be useful to
ensure that unfavorable reactions did not occur between cement, admix-
ture, and superfine material (as reported in section 3.6 on mix design
competitions). Perhaps small test cubes could be cast to yield a strength
correction factor in mix design (i.e., to establish whether, and to what
extent, the materials combination under test gave a strength increase at
a given w/c ratio).
Using Day's MSF criterion, it is clear that the proportion of mortar in
a cubic metre of concrete will be approximately inversely proportional