Re: Why do anions not produce light duing a flame test?

If you put a sample of a salt in a burner flame (adjusted to a very faint 
blue colour), it often produces a colour characteristic of the metal or 
cation part of the salt. Sodium salts produce an orange-yellow colour, 
potassium a pale lilac colour (sometimes swamped by the yellow of sodium 
impurities), lithium, strontium, and calcium different shades of red, and 
barium a lime green colour. If you look at these colours through a 
spectroscope, you find that they consist of just a few sharp lines of 
spectrum in each case.

The question is why, if you use sodium chloride, sodium bromide, and sodium 
sulfate, do you always see the same yellow sodium colour, and never see any 
lines that you might associate with the anion part -- bromide, chloride, or 
sulfate?

The light is emitted when an electron in the metal atom jumps from a higher 
energy level to a lower energy level, and gives off a photon with just the 
right energy to match the energy difference between the two levels.

In the solid state, a salt consists of ions. The chlorine atoms in sodium 
chloride have picked up an extra electron to become chloride ions, while the 
sodium atoms have lost an electron to become sodium ions. But in the gas 
phase, those sodium ions can pick up an extra electron to become neutral 
sodium atoms again. When they do, the electron usually starts in a high 
energy level, and then emits light as it drops to lower energy levels. An 
electron that starts off separate from a sodium ion has about 5 volts of 
energy to lose before it gets to the lowest energy state for a sodium atom. 
The prominent yellow sodium line corresponds to the loss of the last 2.1 
volt of this 5 in a single step. Visible light particles have energies 
between about 1.8 volts (red) and 3 volts (violet).

With the chloride ions, they have to lose their extra electron to become 
neutral chlorine atoms again. This means having to find extra energy rather 
than losing it, so no light would be emitted. Light could be emitted if some 
chlorine atoms were picking up extra electrons to become chloride ions. But 
in that case, the total energy that the extra electron would have to lose 
would be only about 3.5 volts, and the largest single step would be less 
than 1.8 volts, so any light that might be emitted would be in the near 
infrared rather than the visible.