Theory Behind the He-Ne Laser

 Stimulated Emission

 Population Inversion

 Resonance

 Output

 Equipment

Stimulated Emission
Stimulated emission occurs when an photon passes in close proximity to an atom
excited to the energy of that photon (E = hf). When the photon passes, the atom
emits another photon that is coherent with the incident photon, meaning that the
two photons have same direction and are in phase. To see a red beam (6328 Å)
from the He-Ne laser, we desire a transition from the neon's 3S state to its 2P state
emission.  When and if this emitted photon encounters another excited neon atom,
stimulated emission occurs. 

This yields 2 coherent 6328 Å photons--the first bundles of laser light--to stimulate
further Ne atoms.

Population Inversion
Obviously, each time an excited atom is stimulated to emit a photon, the
population of excited atoms decreases by one. To get a laser beam, there must be a
large number of these transitions taking place. Therefore, the key to the laser's
action is population inversion: there must be more atoms in the desired excited
state than in the state to which the electons transition. In the He-Ne laser, as with
many gas lasers, this is accomplished by utilizing metastable states which
transition to relatively short intermediate states. The process is as follows:

Across the ends of the laser tube containing the He-Ne gas, is a potential difference
of about 2000 volts, which strips electrons off a conductor. Some of these electrons
interact with the gas which may yield several possible outcomes.  In many cases,
the electron will excite a ground-state helium atom to its 2 1S state.  Due to
selection rules for electron transitions, this state is metastable, which is not only
handy, but necessary for a population inversion to occur (click here to see
an energy-level diagram of what's going on).  Now, some of these excited He
atoms will come into contact via a collision with a ground-state neon atom.
Because neon's 3S state is at an energy very close to that of the 2 1S state in helium,
through a resonance effect, the collision will excite Ne to its 3S state.  It is from
this state that the desired 6328 Å emission can occur.  Because helium's 21S state is
metastable, there is an increased probability that it will encounter and excite a neon
atom. 

A Neon atom excited to this 3S state may transition to ground and emit a photon.
At sufficiently high pressures in the laser cavity, however, the neon atoms may
undergo radiative trapping, whereby emitted photons are quickly reabsorbed by
other neon atoms. So even if this escited state DOES decay, the photon will
probably be absorbed again soon and the net result will be no change in the excited
population. The Neon atoms 2p state will decay very rapidly to the 1s state. So,
while electrons spend a "long time" in the 3s state due to radiative trapping,
electrons spend a very short time in the 2p state because the jump down to the 1s
state very quickly. This sets up a population inversion: higher population of atoms
in the 3s than in the 2p. So, when stimulated emission occurs, those photons will
likely find another excited atom to stimulate. Losses due to stimulation and random
transitions will be counterbalanced by this population inversion.

Resonance
Once we achieve a population inversion and experience stimulated emissions, there
must be some mechanism to increase the gain of the laser output.  Normally, this is
accomplished by putting mirrors on the ends of the laser cavity. Photons emitted
along the longitudinal axis of the cavity will then reflect back and forth in the
cavity stimulating more emissions along this same axis.   Eventually, the gain will
reach a threshold, and the gas will 'lase.'  Mirrors usually reflect about 95-99% of
incident light (thus the light in the cavity is about 100 times more intense than the
output).  The mirrors may be coated so that they reflect only desirable radiation. 

Output
       Though energy is quantized in electron transitions, Laser light is not truly
monochromatic as one might expect.  Even if we suppress undesirable transitions
(such as the 3.339 um and 1.15 um IR transitions in the He-Ne), the laser output
will still not be truly monochromatic.  The primary reason for this is the
phenomenon of Doppler spreading.  When we drive quickly down the street while
obnoxiously blaring our car horn, the horn sounds higher pitched to a person in
front of the car than it does to a person behind the car.  A similar thing happens
when the source and observer of light are moving relative to each other.  If an
observer moves away from a light source at v<<c, the frequency (f'') of the
incoming light appears to be shifted from the expected frequency (f) according to
the relation    f'' = f (1-v/c).  (This is why the spectral lines of stars moving relative
to earth are either "red shifted" or "blue shifted".)   As a result, atoms at different
velocities will absorb and emit photons of slightly different frequencies from the
same energy line.  In a gas, atoms have a broad range of velocities, so the output
peaks of a gas laser will be broadened.  We might expect the emision profile to be
a sharp peak, but due to Doppler broadening it is actually Gaussian.

    Although the gas emits radiation that has a Gaussian lineshape, the laser cavity
does not permit a continuous spread of frequencies to resonate.  Only those
frequencies that achieve constructive interference in the resonator will reach the
threshold gain level.  When constructive interference occurs, the length of the laser
cavity will be an integral number of half-wavelengths of the resonating frequency:
L = m*wavelenth/2, where m = 1,2,3,....  We can rearrange and substitute terms to
see that the permitted frequencies are v = m*c/2L.  The difference between
adjacent permitted frequencies is delta v =c/2L.  If the light in the cavity were pure
white light, the out put spectrum would have evenly spaced spikes separated by
delta v =c/2L.  It is said to look like a "picket fence."
    Since the gas emits light with a Gaussian line shape and the cavity permits light
with a "picket fence" lineshape, the output is like a combination of the two:     

    The laser cavity, however, is not just one dimension.  As a result, the output
includes both longitudinal modes and transverse modes (TEM modes).   The
transverse modes are determined primarily by the size and curvature of the
mirrors.  They are akin to TEM modes on a waveguide, yet their solutions include
Hermitian polynomials and a Gaussian function and should look something like
this:

For each transverse mode, there will be an associated longitudinal mode and visa
versa.   Nevertheless, the frequency separation between adjacent transverse modes
will not be uniform as with the longitudinal modes.

Equipment
    The primary instrument we used to study the He-Ne laser was the scanning
Fabry-Perot interferometer.  Like the laser cavity, the Fabry-Perot interferometer
consists of 2 confocal mirrors separated by distance L.  The mirrors have a
reflection coefficient of about 98%.  When an incident beam satisfies the
constructive interference condition (v = mc/2L), the light resonates in the cavity.  
As the intensity builds up inside the cavity the output beam builds up to nearly the
intensity of the input beam.  Thus, frequencies that resonate in the cavity are
passed through with little attenuation while non-resonant frequencies are
eliminated.