A porous cup is attached to a simple pressure gauge. When the space around the cup is filled with helium, the pressure measured by the gauge begins to rise as the rate of inward diffusion of helium through the walls of the cup exceeds the outward diffusion of air. When the demonstration is repeated with carbon dioxide, the pressure falls instead of rising.

Gas diffusion apparatus (see Figure)
One 400-mL beaker containing a few pieces of dry ice (covered with watch glass)
1 watch glass that fits the beaker
Helium tank and regulator
Hydrogen gas tank and regulator (if requested)
8 ft of tygon tubing to connect to natural gas outlet and helium tank

This demonstration is an excellent closing for lectures on gas properties. It would be shown to illustrate the kinetic molecular picture of gas behavior, especially the idea that, at a fixed temperature, lighter molecules have higher average speeds (and hence diffusion rates) than heavier molecules. Use the apparatus in the configuration shown in the first figure below.

Make a schematic diagram of the porous cup part of the apparatus on the blackboard and bring out the idea that molecules can diffuse through the channels in the cup walls. The lighter molecules will diffuse faster than the heavier ones. When the same gas is on both sides of the cup walls, nothing appears to happen. Ask the class what happens to the pressure inside the cup when the surrounding air is replaced by a lighter gas like methane. Work toward the idea that the pressure should increase (at least temporarily).

Having prepared a foundation for what will be seen, place an inverted beaker over the porous cup and direct a stream of methane gas (natural gas from the tap) into the beaker. (Be sure that you have leveled the water in the funnel and the manometer and that the stopcock connected to the manometer is open.) The water in the manometer will rise 7-8 inches. Before doing the next part, open the cup and the apparatus to allow the methane to escape and be replaced by air. Do this after every part of the demo.

Ask the class what would happen with an even lighter gas like helium. Repeat the demo using a stream of helium gas from the tank. Normally, the water will rise 20-25 inches in the tube! If available, try the demo with hydrogen gas. I am told that water is forced out the top of the tube on this one.

A most interesting effect can be shown after the water column has reached its maximum height with either helium or hydrogen. Remove the inverted beaker so that the porous cup is surrounded by air again. The water column will immediately start to drop. While this is occurring, ask the class where the water column will stop dropping. What happens is that it keeps dropping well past the equal pressure point because, at least temporarily, helium is diffusing out faster than air can diffuse in.

The demo can be done with a gas heavier than air like carbon dioxide. Have a few pieces of dry ice placed in a covered beaker for a few minutes before presentation. This allows the gas to sublime and fill the beaker. Use the arrangement of the apparatus shown in the second figure. When the cup having air on the inside is immersed in carbon dioxide gas, the pressure in the system drops enough to cause a 4-5 inch movement in the manometer tube.

Removing the carbon dioxide containing beaker allows air to diffuse into the cup. The reverse of the air-helium effect can be seen in a small way here. Temporarily, air diffuses in faster than carbon dioxide diffuses out and the pressure inside the cup rises enough to produce a 1-2 inch rise above equal pressure in the water column.

Be aware that natural gas and hydrogen are inflammable. Other than dropping the helium tank on your foot, these are probably the only hazards.