In my particular design, aquarium water and CO₂ get premixed in the bubble counter. That mixture flows into the top of the sealed reactor chamber. The effluent is drawn from the bottom of the reactor. Most of the CO₂ gas dissolves in the bubble counter, before it ever enters the reactor, but a little bit of the CO₂ enters the reactor as a gas. The water control valve is located after the bubble counter and before the reactor. This valve placement means that the valve can never clog with aragonite, because aragonite is highly soluble in the low-pH (about 5.8) water that passes through the valve. Although most of the CO₂ dissolves in the bubble counter, any undissolved CO₂ gas is trapped at the top of the reactor, where it eventually dissolves and can do useful work, rather than being released to the atmosphere.

This single-pass (plug-flow) arrangement creates a concentration gradient between the top and bottom of the reactor. At the top, the water is rich in dissolved CO₂, but the dissolved aragonite level is the same as that of the aquarium water. Therefore, aragonite is highly soluble at the top of the reactor. As the water moves slowly to the bottom of the reactor, aragonite dissolves and consumes the dissolved CO₂, until all available CO₂ has reacted. The water at the bottom of the reactor should have no unreacted CO₂. The pH of my reactor's effluent is about 7.2, which is considerably less acid than the effluent of a recirculating reactor.

I believe that a reactor of the plug-flow design uses CO₂ more efficiently than reactors that have recirculation pumps. The usual recirculating reactors are mixed so vigorously that we may assume that the water is homogeneous throughout the whole reactor. I mean that the water has the same dissolved CO₂ and dissolved aragonite concentrations everywhere in the reactor. Thus, if aragonite is soluble anywhere in the reactor, it must be soluble everywhere in the reactor. In particular, it must be soluble at the point where the effluent is drawn. In other words, the effluent has wasted (unused) CO₂ in it. The effluent will be more acid than it needs to be to carry the amount of calcium carbonate that it carries. The reef aquarium has to do something with that excess CO₂.

I have DIY instructions for this reactor at http://www.cs.cmu.edu/~cap/raid/careactors/waterfilter/. I put that page on my web site about a year ago, but I made the switch to non-recirculating reactors earlier than that.

Add or View Comments