My advice: build a N2 laser first, just to get into it. Then build an even better N2 laser that can run on different gasses, and then start at the difficult gas lasers. A ceramic bore also is possible, but I have no idea if this can be homemade. Just start out with glass and run it only seconds at a time.
If you must use a metal for the bore, use stacked aluminum disks and and oxidize them. The AL oxide does not conduct, but you need a huge number of disks. The disks are then stacked in a water cooled alumina ceramic tube. That has been done, but they are hard to start.
A more modern tube is a set of copper disks mounted about 1/4" apart with a tungsten core brazed to the disk, the copper does the cooling, and the tungsten holds off the plasma. The copper disks have folds in them to spread the stress to the outer ceramic tube.
Your talking a excitation 350 to 1000 watts per centimeter of length for a modern CW argon. Pulsing makes use of duty cycle to allow the heat to escape.
The easiest home made argon that could run CW would be a quartz tube with graphite disks. The graphite heats to glowing hot and the cooling is by radiative heat into the cooling water around the tube. If you don't care about lifetime, the earliest argons were quartz bores, but the quartz dies very quickly and its hard to seal the cathode lead throughs direct to quartz.
A pulsed home made argon is possible. Pulsed Oxygen is a MUCH better candidate for a visible laser as is pulsed iodine vapor.
See:
:: Pulslaser ::
Sam's Laser FAQ - Home-Built Pulsed Multiple Gas (PMG) Laser
A Cold Cathode Pulsed Gas Laser" by R. K. Lomnes and J. C. W. Taylor in: Review of Scientific Instruments, vol 42, no. 6, June, 1971
The Scientific American Helium-Mercury design works, but your talking about 15 mW of average power.
I'd strongly suggest you read Lomnes and Taylor before starting any pulsed or cw gas laser construction.
This is lot of work. I've repumped and rebuilt existing tubes and its NOT easy, start with a N2 laser, then a N2 pumped dye laser, and work your way up. To really ARGON work CW you need a lathe, maybe a tig welder, and a turbo-pump roughing pump combination. I have a 270 litre per second ion pump, a turbo pump and two roughers and it can take me 24-48 hours to get to the level of vacuum needed for a cw ion laser, to process the cathodes. Pulsed lasers are far less stringent, you only need to get to .01 or .001 torr, I need to hit 10 to the minus 7 torr, minimum.
But a warning, a fingerprint or other crud in the wrong place inside the tube and your effort is wasted with argon.
Yes, welding grade argon works OK, sorta. . It works great for pulsed duty, but any oxygen or nitrogen or water vapor kills cw lasing. My rare gasses are about 700$ a bottle for CW grade gas. The difference is they bake out the tanks and purge them for scientific grade. Welding grade tanks are not flushed or baked, are often rusty on the inside, and who knows what was in them, before they repainted them for argon use. Helium, however, is reasonably pure from cheap tanks.
And yes, JDSU/uniphase makes their tube's main body from a copper alloy, with ceramic parts brazed in, but glass to metal and ceramic to metal seals are very tricky.
If you really want the details, a Air Force technical report by Bridges and Halstad gives the details of 4 early tube designs, but its 140$ for a hard copy and is NOT available as a free pdf. I have it. It would be hard for a non US citizen to get it.
One of these days I'll rewrite the PMG chapter and send it to Sam, but for now, its fairly correct.
Steve