The hydration kinetics of slag is generally divided into three stages: (1) a nucleation period during which product growth is accelerating, (2) a phase boundary controlled stage, and (3) a diffusion controlled stage.
The reaction degree of slag in blended cement is influenced by many factors. The main factors affecting the reaction of slag in blended system include the reactivity of slag that could be defined as (C+A+M)/S, the fineness of grinding (specific surface area), the vitreous fraction of slag, the replacement level of slag in blended system, the hydration temperature and the water/solids ratio.
Precisely, the rate of reaction of slag decreases with the decreasing water/solids ratio and with increasing proportion of slag in the blend. Higher hydration temperatures increase the reactivity of slag. However, the composition of cement and the incorporation of additional gypsum in slag blended cement have no influence on the extent of reaction of slag at ages of 28 days to 1 year, though variations in Portland cement can affect early strength. The following figure is a representation of the effect of hydration conditions and slag characteristics on the reactivity of slag.
Representation of the effect of hydration conditions and slag characteristics on the reactivity of slag.
As it is well-known, the activation of latent slag in blended cement is implemented by calcium hydroxide (CH, Ca(OH)2) which is the main cause maintaining the high pH value of pore solution.
The activation happens as the break of glass layer of slag particles. However, how does the break take place in detail? Researchers have already proposed a theory to explain the process, which will be discussed in the post. The main composition of glass is SiO2, thus the break of glass layer is actually the break of chemical bond between Si and O (Si-O). In the first step, the bond Si-O is broken by hydroxyl,
Then the other part of broken bond Si-O– further react with water (H2O), as the following reaction equation,
By the second step, hydroxyl is returned. The break of Si-O bond is finished. Therefore, the two steps can be combined into one equation,
From the overall process, it can be seen that hydroxyl acts as activator and is not consumed.
This is the theory which explains the activation of slag by CH in slag blended cement.
The morphology of inner product region derived from alite or belite is identical to those present in neat OPC paste. In mature pastes, fully hydrated small grains of both alite and GGBFS often display a coarse morphology.
Generally, inner product in larger grains has a dense homogeneous morphology with very fine porosity, and large slag grains often have a rim of fine textured C-S-H which can exist for many years and merge into outer product C-S-H gel. Mg, Al-rich laths or possibly platelets present within the inner product C-S-H gel, which are oriented either towards the outer boundary of the inner product or randomly even at the outer boundary.
The development of the plates occurs to different extents in different slag grains, possibly depending on composition and also appearing to depend on particle size considering platelets or laths typically appear very early in small grains. When the plates are very well developed they predominate over the C-S-H gel; however, in many inner product regions, even in 3 year old pastes, the plates are less well developed and a high proportion of C-S-H gel is still present. It is not obvious from observations in the early stages that the morphology in the inner product regions is plate-like, thus it is better to be described as laths or needles.
Another microstructural feature of the inner products is the presence of small, round and poorly crystalline particles, which are rich in iron and aluminum and also often contain significant amounts titanium.