After spending over 10 years in the compressed gas industry, this is a nostalgic subject for me. It takes me back to 1962 when I started using Teflon tape in my plants as well as for compressed gas cylinder valves. I literally had to teach my maintenance and operations guys how to apply or use the tape. My plants were overseas.
The Compressed Gas Association (CGA) had, and I believe it still has, standards for each of the unique and different cylinder valves according to the contained gas product. I remember each brass cylinder valve for Oxygen, Nitrogen, Air, CO2, Acetylene, Nitrous Oxide, Argon, etc. had nominal 3/4" or 1" TAPERED male threads which matched the corresponding cylinder female threads. Litharge (a concoction known only to folks in the industry) was used on Oxygen cylinders and met the need for non-combustibles. I've seen string and paint used in the past on CO2, N2, and Acetylene - but I didn't allow that in my plants.
I don't know why Shell (Oil, as compared with Chemicals) doesn't want to use Teflon tape - I would suggest you ask them; I'm sure they would be glad to share their reason(s) with you.
However, I do have some suspicions as to why they would take that position:
1. Teflon is notorious for having a "bad" memory - sort of like old, retired engineers. In other words, it doesn't retain it's original tensile strength. After a while, it relaxes and starts to "run" or stretch. This has been proven many times in past bearing and gasket applications. The resultant stress failures have forced manufacturers to use REINFORCED TEFLON - usually with glass or other fibers to assist in keeping its memory. This flaw found in pure teflon may be one reason for not relying on its sealing properties on critical, HP gas cylinders.
2. Hacksaw is 100% correct in challenging the proper application of the tape by unknown or untrained workers or pipers. Hacksaw probably knows how to properly apply the stuff; I certainly know how (after doing it a jillion times, it seems); but can we trust others to methodically do the same correct procedure time after time? Shell may have opted not to take the risk.
3. As Hacksaw infers, if the tape is not applied correctly, it can do harm by contaminating the internal volume it is sealing. No sealant will make up for corroded or bad threads. Excess sealant only makes the situation worse. For example: An experienced pipe fitter knows that you must relinquish the first two or three tapered threads of the male connection to any sealing effect. For the lay person, this means that you must start the tape application 2-3 threads INSIDE the connection in order to assure that no excess trailing "tails", or starting pieces, are allowed to freely fall off the threads inside the connection and mix with the internal fluid. These stretched and frail pieces of teflon can play havoc with a HP gas application. I know first hand, because it happened to me: the stuff is inherently inert and doesn't degrade or break up nor dissolve. Therefore, it stays around inside the cylinder waiting for the opportunity to be caught up in gaseous stream rushing out of the cylinder valve and directly into the usual gas regulator.
For those folks not familiar with this industry, almost all these HP gas cylinders have to be applied with an immediate down-stream regulator - and these regulators are usually 2-stage jobs and are notorious for their rather small orifices and gas passages. This presents an immediate invitation to a flow plug up!
Now, image yourself as a patient in a hospital emergency room with an immediate need for Oxygen.....and the regulator plugs up!!! I don't think you (or your lawyers) would take this very lightly in the inevitable law suit that would follow. This is probably one of the strongest reasons for staying away from mis-applied,frayed teflon tape.
Acetylene cylinders are a special case. They don't fall under the category of High Pressure (max. fill pressure = approx. 300 psig), but they are packed internally with a filler material that is doused with Acetone liquid. It is the Acetone that makes the inherently unstable Acetylene very stable by absorbing the gas and keeping it in solution until the pressure is released by opening the cylinder valve. It is this prescense of filler material inside the cylinder that makes the possibility of frayed teflon pieces staying directly beneath the brass valves inlet and possiblity plugging the valve seat. Should this happen, you have a more sticky and hazardous situation: you now cannot empty the cylinder contents - and you have to resolve the situation. The only way to do this is to slowly unthread the safety disc on the valve and release the contents down to zero pressure so you can subsequently replace the plugged valve with a new one. This is a touchy kind of situation for which you won't find too many volunteers. I prefer to avoid these kind of situations by simply eliminating the possible contributants that cause them: in this case the Teflon. Shell may have arrived at this same conclusion.
I hope this trip down memory lane helped to logically explain some safety reasons as why Teflon may not be acceptable as a thread sealant in compress gas cylinder valve applications. There may be others.
Art Montemayor