After the previous round of experiments, I assumed that the source of my problems was the water itself.  Several folks left comments suggesting that only distilled water will yield clear ice, and tap water or filtered water was simply too impure.  So, I grabbed a bottle of distilled Arrowhead water and tried freezing it.  Fail.  Then I tried boiling twice it and freezing it.  Fail.

Then, I came up with another idea… a radical idea.

The blocking issue in all of my trials was that the top layer of the water froze first and prevented the dissolved gasses in the water below to escape during freezing.  As a result, the middle of the cubes ended up cloudy.  However, under circumstances where water is constantly flowing (like in rivers and around glaciers) no such cloudiness occurs.

I filled a tall jar with regular tap water and placed an aquarium pump just below the water line.  The pump was oriented to suck water up from the bottom of the jar and shoot it out at a right angle.  The result: the water below the pump was 100% clear, but the water surrounding the pump’s outflow nozzle was completely cloudy.  Although I could have cut off the cloudy section of the block and cut the clear section into cubes, my criteria for success is that I can create clear ice with no manual intervention during or after the freezing process. 

Next, I thought, “What if I keep the water in motion as it freezes?” extrapolating the concept behind clear river ice.  I bought a vibrating “personal massager” (yeah, you know the subtext) from Bartell’s for $15.  I clamped the massager in place and balanced the ice cube tray on top.  When I filed the tray and turned on the massager, it created cool standing waves in the water (shown above) and made the texture of the water feel like gelatin.  I was hoping this vibration would be sufficient to allow gas bubbles to release during freezing.  After two hours of freezing, the water was still completely clear, but not yet fully frozen.  I was optimistic, and left the tray overnight to freeze.

The next morning, I was dismayed to see that this experiment had failed as well.  Sometime during the late stages of freezing, the ice had turned cloudy in the middle, just like all my other tests.

Next up, I’m going to try to force the ice to freeze from the bottom-up.  My idea is to place a heating pad over the ice cube tray and slowly reduce the heat over a period of hours.  Hopefully, this will ensure that the top layer of water is the last to freeze and all of the trapped gasses can escape.  In the meantime, I’ll just have to cool my drinks with the vibrator cubes.