When the weather gets colder, we know right away that wearing layers will keep us warmer. Simply put, if you want more protection, you add more thickness. The same principle applies to gas transmission testing. The rule of thumb is that if you double the thickness of the material, the barrier level will also double and the corresponding transmission will be reduced by half.
How much does thickness affect permeability?
What is the test time required for penetration of thicker samples to reach equilibrium.
A typical assumption is that doubling the thickness requires doubling the test time. This is not correct. Typically, each time the material thickness doubles, it takes four times as long for the permeability to reach equilibrium. Below is a comparison of 1mil and 5mil thickness PET films and their permeability levels. These films were chosen because they reach equilibrium in WVTR within a short time.
In this example, the water vapor transmission rate (WVTR) of 1mil PET film is 10.1 g/(m2 x day). Reaching 95% of this value takes less than 30 minutes. The WVTR of 5mil PET film is 2.17 g/(m2 x day) and takes nearly 450 minutes to reach 95% of the final value. We generally find that for thick samples, especially when measuring higher barrier materials, the final 5% to 10% permeability equilibrium can take a relatively long time.
Draw conclusions through testing
When testing barriers on thicker materials, the overall permeability decreases proportionally. When the material thickness increases by 5 times, the measured WVTR decreases from 10.1 g/(m2 x day) to 2.17 g/(m2 x day). As the material thickness increases, more time (more than 5 times) is required to test the sample to reach equilibrium.
As shown in the figure, both the permeability level and the time to reach equilibrium are affected by the thickness of the material. This is something to keep in mind when you optimize test conditions (such as flow rates for WVTR and CO2TR) and test duration to ensure equilibrium values. Standard test settings suitable for thin samples may produce inaccurate or premature results for thick samples.
From MOCON