[dgasmd]

I personally wouldn't get a canister filter. They have too much flow through them and also have limited space for media as well. If it was me, I would get 2 flow though chambers (call them reactors if you'd like) and use each for its particular purpose. Put the necessary shut off valves and you can disconnect them from the line when you want to. Much neater and easier to work with too.

As far as carbon, I personally wouldn't be concerned too much with the stripping of trace element issue. God knows we have too much of all of them anyway, so a little stripping may be a good thing in my view. Even if you are using a carbon of "high quality", which ususally equals to expensive, I would put some of the carbon in a small container with some RO/DI water of 0 TDS and let it sit there for about an hour. Then, test the water for phosphate with a salifert kit. Most people I know that are to the max with their tanks have found that even the same brand of carbon that most times will test 0, from time to time will have some phosphates in it.

The last thing about both substances, carbon and phosphate removing media, is that both need very little flow to pass through them. Just enough to mantain the media moving, but not enough that it is tumbling. This allows the most amount of surface area to be exposed and you get the most for your money.