We modeled the Nalgene chamber setup and designed an ETC chamber that can be 3D printed. Two sets of vertical grooves were added to hold the electrodes in place. There is also a groove for fitting an Oring. The design can be downloaded from http://www.thingiverse.com/thing:106486
. We use a Makerbot Replicator 2 and it prints well. Test units are available for those who don't have a 3D printer (yet), but only the first three requests will be honored due to our limited capacity. Please leave your request in the comments area. Thanks.
Also, since you are using a Makerbot Replicator 2, I am assuming you are printing entirely in PLA. Has this material held up in the clearing process? We have used ABS, and noticed that the material deforms enought to cause the bubbles from FDM to start to come out of the material. I think that PLA actually has a lower heat deformation index than ABS, so I wonder if that has created problems for you.
For anyone else, once I have a really reliable and watertight design, I will be happy to share it as well. I will be getting and testing 2 new chambers this coming week. I have ordered them in an SLS printed nylon material, to see if that is a better option for heat and chemical resistance.
As for the PLA and temperature, it would be possible to run things in a cold room, but I am assuming that this would slow down the clearing process to some extent. It would be worth trying to run at ~37 degrees with the PLA... it's really the heat generated from the electrophoretic cell that creates the biggest problem. Externally, the structure holds... the ABS just gets warped on the inside, and eventually might crack. The walls of your chamber look pretty thick though. If you took the measurements from the protocol, 4.0mm should be a reasonable wall thickness to hold up.
The other thing that I have done for the ABS is melt it some on the sides with acetone vapor, to sort of seal it. If something similar can be done with PLA, it's probably a good idea too.
Here is the design for what I made if anyone is interested http://www.thingiverse.com/thing:108854. I tried printing without many shells (maybe 2), 10% infill and 0.25-mm layer thickness on a Replicator 2. Worked fine from what I can tell with the just-water test.