Physics/Plasma ball interacting with light bulbs
At the local STEM fest near us my daughter and I saw one of the students light up a fluorescent light bulb by holding it to a plasma ball. So for her science experiment this year she will be trying other kinds of light bulbs to see if any of them behave the same. Not knowing myself much about how light bulbs really work, I did a test run to make sure nothing was going to shatter when she tried it. I tried a CFL, Halogen, Regular incandescent, and an LED(40W). I thought maybe the Halogen would light up, but expected none to. Instead the LED produced a dim light if I held the metal part to the plasma ball. Excited by this I bought a smaller LED .5 watt night light figuring it would produce a brighter light, but it produced none at all(and I did check it in a night light and it does work). Not sure why a lower wattage bulb would produce no light, but the higher wattage one would produce a dim one. And no light was produced if I touched the glass bulb part to the plasma ball which is how the florescent bulb lights up. But there is definitely light when I touch the metal part. While we were looking up how a plasma ball works we happened across this site and I thought I might as well ask what might be going on so I don't have to tell my daughter 'I don't know' yet again.
Do be careful when performing such experiments with a plasma globe, the wrong kind of conductive part in contact with the globe and ground has been known to cause sparks and even fires. A plasma ball has high-frequency electrical fluctuations (in the 35-ish kHz range). Halogen bulbs generally require quite high current to heat their tungsten filament like a regular incandescent bulb does, so I'm really not surprised that it didn't light up. You would need to ground them for sure, and definitely capacitively coupling so much current would prove dangerous. A very low-power christmas tree bulb might actually light up this way, but I wouldn't recommend it. The difference in the LED bulbs is mysterious, but it all depends on what the LED bulb had to draw on, it could be that the low-wattage LED bulb didn't have enough capacitance (though you could connect it to a larger capacitor, but be careful with capacitors) to generate enough current to cause it to light up. The higher-wattage bulb might have had enough, but I don't know enough about your specific bulbs to know for sure. Current flowing back and forth in the typical fluorescent bulb is its usual mode of operation, much like what happens in the plasma globe itself (though usually at a lower frequency), so it's no surprise that this bulb would respond the best.