Yahya Ghasemi Print III pdf

4.1.5. Surface area measurements
Measurement of surface area is of great importance in calculations based on layer theories and 
has a significant effect on water requirement of the concrete mixtures. However, no 
commonly accepted methods of measurement exist. The surface area is usually measured 
either by Blaine test or BET test. 
Blaine test derives surface area from the resistance to flow of air through a porous bed of a 
powder and should be conducted based on instructions in ASTM C204-16 standard. The 
method was originally introduced for measurement of surface area of cement; however, the 
test has been used for determination of the measures of fineness of various materials. It should 

---

32
Methods
be understood that the obtained values are relative to fineness of cement and are not absolute 
measures of surface area.
BET test (Brunauer et. al., 1938) works with adsorption of gas molecules (usually Nitrogen) 
on a solid surface which can be used for measurement of SSA. The results from BET depend 
on the adsorbate molecule utilized and its adsorption cross section. BET can lead to results 
that are several time larger than results from Blaine test for a given material since the values 
from BET test include surface area of inner pores of the material as well as external surfaces. 
Both Blaine and BET tests are fairly expensive and complex to perform and will eventually 
lead to not so accurate results. Surface area can also be mathematically calculated based on 
some basic assumptions and by using the particle size distribution. Calculation of specific 
surface area is one of the main scopes of this thesis and is discussed more in detail in Papers 
III and IV in Appendix 1 and also in Section 3.3.2. 
4.2. Theoretical methods
4.2.1. Studied particle packing models
Among the several existing particle packing models, three were chosen for comparison of 
calculated packing density values to conducted measurements in laboratory and the results 
were reported in Paper II in Appendix 1. The chosen models included Modified Toufar, CPM 
and 4C. Modified Toufar was chosen for its simplicity, CPM for its supposedly accuracy and 
4C for its approach to utilizing both particle packing theory and ideal curve fitting.
The models are different in terms of calculation, required input data and application. Modified 
Toufar was originally designed for binary mixtures and mentions no requirements about the 
packing methods. CPM is capable of handling multi component mixtures including 
aggregates and cement and requires packing density of each fraction of materials. Moreover, 
the method of packing is considered in the model by utilizing index 
K
. DTI 4C model can 
calculate the packing density of three component mixture and requires hard packing density of 
constitutes.
It should be mentioned that in order to treat the models in a similar way, loose packing data 
was used as an input to all of the discussed models. Furthermore, the packing densities were 
measured on two wide range fractions defined as fines (0-8 mm) and coarse (8-16 mm) and 
not every sieve fraction as it is suggested by CPM.