The Mechanism of Air-entraining Admixture
The air-entraining admixtures are commonly used to improve properties of fresh and hardened concrete. It is recommended to improve the workability, resistance to freeze-thaw cycles, de-icing chemicals, sulfates and alkalis-silica reaction, while it is also noted that it would decrease strength of concrete.
So, what about the mechanism of air-entraining admixture? We focus more on its resistance to freeze-thaw cycles. The procedure is as follows:
1.Produce tiny, dispersed air bubbles.
First of all, air-entraining admixtures are kind of surface-active agents which acts at the air-water interface. The micro structure of the admixtures is much like a kind of detergents which contain Hydrophilic group（亲水基团） and Hydrophobic component（疏水部）. When we add it in cement, the Hydrophobic component will stick to the surface of the air bubble and leave the Hydrophilic group out. Then the combination of air bubble and detergents will float between the concrete particles.
So base on this cute structure, the air-entraining admixtures are able to cause water to foam during mixing. And the some fine foam could remain stable and eventually become air bubbles which are “locked” into cement paste during hardening. The air bubbles (which are formed from the foam caused by admixtures) are the key to our next study.
2.Provide space for water to expand upon freezing.
Actually it`s a little complicated but interesting process, I would like to introduce from The Osmotic Pressure Theory which is the basic of our discussion.
2.1The osmotic pressure theory(Powers)
Gibbs-Thomson equation gives the relationship between water freezing point and pore radius:
$$
In(\frac{T}{T_0}) = -2\frac{\Delta GV}{\Delta Hr}
$$
where:
$$
T = water\ temperature\ in\ pore
\\ T_0 = freezing\ point
\\ \Delta G = G(matrix/ice)-G(matrix/water)
\\ V = molar\ volume\ of\ water
\\ \Delta H = latent\ heat\ of\ fusion
\\ r = pore\ radius
$$
The curve can be plotted according to the equation:
The relationship is obvious from the curve: Freezing temperature of the water depends on the size of the pore. The bigger the pore is, the easier water in the pore to be frozen. So, the water in capillary pores(10nm-10μm) shall freeze first compared to the gel pores(0.5-10nm).
Let`s take it a step further. When the water in capillary pores freezes, the energy would get lower because supercooled water (water < 0°) has higher free energy than ice. And the concentration of the remaining unfrozen pore solution would increase because there is less water as it freezes.
Then according to the laws of thermodynamic and the theory of osmosis.
The laws of the thermodynamic: Diffusion from high to low free energy.
The theory of osmosis: Diffusion along concentration gradients.
With these theories, we understand the mechanism of freeze-thaw damage of concrete: If much water flow from gel to capillaries and freezes, the capillary will become full and pressure will develop. When the pressure reaches a certain level, cracks will appear in the concrete.
So, we can identify the direction of concrete improvement: Make water flow out of the capillary pores as much as possible to avoid excessive pressure caused by ice expansion.
For convenience, let’s sketch the microstructure of cement.
2.2The direction of water penetration in concrete
As shown in the schematic diagram, \( V_{air-bubble1}>V_{capillary pore}>V_{air-bubble2} \).
According to the conclusions given by 2.1, the freezing temperature can be concluded as: \( T_0^{air-bubble1}>T_0^{capillary pore}>T_0^{air-bubble2} \). It means that the icing order is as follows: air-bubble1→capillary pore→air-bubble2.
Now we discuss the effect of bubble size on the permeation direction:
①Air bubble 2 with smaller size than capillary pore
If the size of air bubble is smaller than capillary pore, the water will flow from the gel pore to capillary pore according to the laws of thermodynamic and the theory of osmosis.
To be specific, the water in capillary pore shall freeze first so that the salt ions in the bubble are higher and the energy is lower. In this case, the air bubble 2 cannot prevent the water osmosis direction towards capillary pore because the air bubble does not have chance to absorb water and provide room for its freezing expansion.
②Air bubble 1 with bigger size than capillary pore
On the contrast, if the air bubble is bigger than capillary pore, then the water shall flow from the capillary pore and gel pore to the air bubble 1. It means that the air bubble 1 can effectively prevent moisture from penetrating the pores by providing a lower energy and lower concentration.
In other words, any water in the air bubbles will begin to freeze at a temperature close to 0° that allows the processes of osmosis and desorption, reducing the level of saturation in the adjacent cement paste. The air voids act as “safety valves” drawing water from cement paste and serving as reservoir.
What`s more, this conclusion also indicate that the distance between air bubbles must not be so far apart that water cannot pass through the pores to desorb. That is why we need Critical Spacing Factor (≤0.2mm, 9% of air entrainment by volume)
Spacing factor: The distance of air voids must not be too great if osmotic pressure is to be relieved; hence the requirement of a critical spacing factor.
Spacing factor = average distance from any point in the paste to the edge of the nearest void, should not exceed 0.2mm(~9% of air entrainment by volume)
Finally, we understand why we hope to control the air voids with a size bigger than the capillaries.
3.Pull the water back into the capillary pores upon thawing.
Upon thawing, the water in the air bubbles(also known as air voids) returns to the cement paste because of the higher surface tension in the smaller capillaries and pores.
Last but not least, we shall recognize the disadvantages of using air-entraining admixtures: it usually causes loss in concrete strength (for each 1% of air causes 5% loss in strength). So, nothing in the world is perfect, even the air content should be controlled in the most suitable degree.
