Friction is resistance. In this case, it shows how difficult it is for something to pass through the membrane. If you design a membrane with low water resistance, more Tolerance to salts and other things you want to remove can potentially give you a cleaner product with less work.
However, that model was shelved in 1965 when another group introduced a simpler model. This assumed that the membrane’s plastic polymer was dense and had no pores through which water could pass. Instead, we assumed that the water molecules in the salt solution dissolved in the plastic and diffused from the other side. This is why it is called the “solution diffusion” model.
Diffusion is the flow of chemicals from more concentrated locations to less concentrated locations. Think of a drop of dye spreading in a glass of water, or the smell of garlic wafting out of your kitchen. I don’t.
The model didn’t work, but Elimelech always suspected it was wrong. For him, accepting that water diffuses through membranes implied strange things. That is, it dispersed into individual molecules as water passed through it. “Why?” Elimelech asks. To break up clusters of water molecules, tons of energy. “To get water into the membrane, it has to evaporate.”
Still, Hook says, “Twenty years ago it was an abomination to suggest that it was wrong.” Hooke didn’t even dare to use the word “pore” when talking about reverse osmosis membranes. “For years I have called them ‘interconnected free-volume elements’.”
Over the past two decades, images taken using advanced microscopy have reinforced Hoek and Elimelech’s questions. Researchers have found that the plastic polymers used in desalination membranes are not as dense and non-porous. They actually contain interconnected tunnels, but they are very small, with a peak of about 5 angstroms in diameter, or 0.5 nanometers. Yet, since one water molecule is about 1.5 angstroms long, there is plenty of room for small clusters of water molecules to pass through these cavities instead of moving one at a time.
About two years ago, Elimelech felt it was time to do away with the solution diffusion model. he worked with the team. Li Wang, a postdoc in Elimelech’s lab, studied fluid flow through small membranes to make the actual measurements. His Jinlong He at the University of Wisconsin-Madison tinkered with his model, a computer that simulates what happens at the molecular scale when pressure forces salt water through a membrane.
Predictions based on the solution diffusion model should have the same hydraulic pressure on both sides of the membrane. However, in this experiment, the team found different pressures at the entrance and exit of the membrane. This suggests that pressure drives water flow through the membrane rather than simple diffusion.
