Testing - Concrete Mix Design, Quality Control and Specification - page 170

requirements rather than the specific ingredients seems the best way
forward. The problems with specifying minimum cementitious contents are
discussed in Chapter 6 and therefore, in our opinion, the quantity of cementi-
tious material should be the concern of the premix company. The relative
proportion of different cementitious materials can be more complicated as
it may affect long-term performance. For example, in the Middle East, sup-
plementary cementing materials are imported and generally more expensive
than Portland cement. They are often specified at minimum replacement
levels to reduce temperature rise or improve durability. Variation in the
proportion of these materials can result in significant detrimental effects on
the concrete, and any procedure that can detect such variation as soon as
possible would be very helpful.
The mutual suspicion that often exists between specifiers and suppliers is
counterproductive to achieving optimum performance and reducing vari-
ability to the benefit of all parties. Certainly, restrictive specifications that
do not allow the premix supplier to adjust the mix in response to variations
in materials or mix registers that fix mix proportions do not encourage
transparency when it comes to information of actual batch weights.
Because of the large volumes involved in producing concrete and the
errors associated with the analysis, the authors are not convinced that
attempts to confirm the correct mix proportions by wet analysis techniques
should be the focus of quality control except for critical parameters such as
free water and air content described earlier.
7.6 TEST PROCEDURES FOR ASSESSING DURABILITY
We will focus on chloride-induced corrosion, which is the predominant
reason for deterioration of reinforced concrete. The subject of test proce-
dures to assess durability in severe environments is a veritable minefield
where only the brave (or the foolish) dare to tread. However, a fundamental
change in the way specifications are written seems necessary. The authors
believe effective and user-friendly performance specification is the answer.
The object of appropriate performance tests should be to provide an accept-
able probability of achieving the specified service life. However, there are a
number of problems with establishing an agreed-upon test procedure and
performance criterion upon which to base such a specification. First, there
are distinct transport mechanisms: sorption of water containing chlorides,
permeation of the chloride solution, and the diffusion of free chloride ions.
These may act singly, simultaneously or in series depending on the expo-
sure condition and the moisture content of the concrete. For example, con-
crete in the lower splash zone when impacted by a wave will initially be
exposed to permeation of seawater under some pressure. Simultaneously
and then continuing for some time after permeation, sorption will occur