You're not going to be able to make anything equivalent to a commercial product, but you can make a heat pipe that does work as an experiment. I've done just this out of some refridgerator tubing and a brass valve.
WARNING: This is dangerous. If you're going to do it at all, do it outside and wear face/eye protection.
Basically, I pinched and soldered one end of the tube and put the valve with a second, bent piece of copper tubing on the other. To make the heat pipe, I filled the tube completely with acetone, shut the valve and put the soldered end into a cup of nearly boiling water. At this point, I could feel the tube vibrating with what I imagine was the fury of boiling acetone so I opened the valve briefly and closed it again. The assembly spit acetone out of the bent copper tube after the valve (It's bent so you can direct the acetone away or into a receptacle of some sort) and I closed the valve as soon as I saw vaporous acetone come out, even if there was still liquid coming out.
Now, I had an evacuated tube with approximately the ideal amount of liquid acetone to act as a heat pipe. I let the pipe cool and then put the soldered end into a cup of nearly boiling water. Lo and behold, the top of the heat pipe became hot quite rapidly whereas a simple piece of copper pipe equal in length to the heat pipe did not even warm up.
Naturally, the heat pipe only worked vertically because I did not include a wicking structure or anything like that.
Well, looking around, I found some heatpipes. zalman had some older vga coolers that used heatpipes that weren't permanently soldered to the coolers. I could probably by one of those kits that comes with 2 heatpipes.
That is also called a thermosyphon by some people. The same thing should work with just using r134a freon. The pressures involved will make it such that the amount of air in the system may be minor enough to not affect overall performance much even without vacuuming out the system first.
What I am personally planning on doing in the near future is making essentially the same thing Inept did, but with water as the substance inside. With water you could easily get the bottom end of a tube up to boil to evacuate the air and then bend the other end over and solder shut. I might just go do a test run with a single pipe before I hook up a cpu block and huge radiator complex.
Last night I was playing around making a single tube version with just water. It is possible to solder the end shut, but reasonably tricky to keep air from getting back in. Further, since water has such a low vapor pressure at room temp it seems like almost any air getting in seriously harms performance. Basically, after I first made got it sealed the tube (4~5 foot piece of 1/2" soft copper) worked great on the bottom 3/4. Today it has about 1/4 working most likely due to some pinhole leak in the solder.
I believe if you dont mind using something more flamable, and have plenty of fireproof clothing, then using acetone or something similar while shoving some cotton down the length of the device might work for your purposes.
After seeing how much water I shot out of my tube before it all went to steam going out, I am not sure if I want to try that yet.