Morgan-Keenan (MK) Spectral Classification

Morgan-Keenan Spectral Classification

(O,B,A,F,G,K,M, hi to lo T)


Based on the appearance of the spectra of stars, a spectral classification scheme was devised in the late 1800s and the early 1900s at Harvard in a group led by Annie Jump Cannon. The criteria used to define the sequence were based primarily on the strengths of the hydrogen Balmer lines but other features were also considered. Today, other criteria are used and so the ordering is rather more obscure. The ordering is O, B, A, F, G, K, M . This is a temperature sequence starting from the hottest stars at O going to the coolest stars at M.

Examples of Stellar Spectra

Note that different spectral lines are seen and that the strengths of the lines are quite different. Below is a rough description of the spectral classes:

Spectral Class

Temperature

Comments

O

28,000-50,000 K

ionized atoms, especially helium

B

10,000-28,000 K

neutral helium, some hydrogen

A

7,500-10,000 K

strong hydrogen, some ionized metals

F

6,000-7,500 K

hydrogen and ionized metals, such as calcium and iron

G

5,000-6,000 K

ionized calcium and both neutral and ionized metals

K

3,500-5,000 K

neutral metals

M

2,500-3,500 K

strong molecular lines, e.g., titanium oxide, and some neutral calcium

We know that the chemical compositions of most stars are roughly the same, why do stars show such different spectra? To clear up this point, we discuss the hydrogen lines.


Formation of the Hydrogen Lines

The hydrogen lines are weak in cool stars, increase in strength as the temperature increases, reach a peak around the A stars, and then weaken at higher temperatures.

Question: Why don't all stars show strong hydrogen lines if hydrogen is the most abundant element in the Universe?

The reason has to do with the energy level structure of hydrogen and the temperatures of the stars.


The competition between excitation and ionization leads to the maximum hydrogen line strength in stars with T roughly 8,000-10,000 K, A stars.


Similar arguments can be made for the other lines seen in stellar spectra.