Lasers

Section 4.
1 Stimulated Emission and Photon Amplification An electron in an atom can be excited from an energy level E1 to a higher energy level E2 by the absorption of a photon of energy h = E2 E1.
E2 h
Chapter 4
Stimulated Emission Devices LASERs

E1
This absorption process requires that the electron change energy such that it moves from an allowed state to another allowed state. If the change is from the valence band to the conduction band, then the minimum energy is equal to the bandgap energy.
Stimulated Emission and Photon Amplification When an electron at a higher energy elevel transits down in energy to an unoccupied energy level, it may emit a photon.
h E2
Stimulated Emission and Photon Amplification Another possibility is the electron can be induced to undergo the downward transition.
E2 h In h Out h
E1
E1
The electron can undergo the downward transition by itself quite spontaneously.
This last transition is called stimulated emission.
Spontaneous Emission
E2
Stimulated Emission
E2 h In h Out h
E1
E1
In spontaneous emission, the electron falls down in energy from level E2 to E1 and emits a photon of energy h = E2 E1 in a random direction. Thus a random photon is emitted. This spontaneous process occurs at some rate which will be constant under thermal equilibrium. It is important to note that these spontaneous energy level changes will always occur and the spontaneous emission of a photon may result.
In stimulated emission, an incoming photon of energy hv = E2 E1 stimulates the whole emission process by inducing the electron at E2 to transit down to E1. Due to the coupling of the electric fields of the photon and the transitioning electron, the emitted photon is in phase with the incoming photon, it is the same direction, it has the same polarization and it has the same energy hv = E2 E1. Stimulated emission is the basis for obtaining photon amplification since one incoming photon results in two outgoing photons which are in phase. In an avalanche device, the two outgoing photons interact again so that two more photons are emitted and so on.
Population Inversion
To obtain stimulated emission, the incoming photon should not be absorbed by another atom at E1. When we are considering a collection of atoms to amplify light, we must have the majority of the atoms at the energy level E2. Otherwise the photon is more likely to be absorbed rather than create a stimulated emission. When there are more atoms at E2 than at E1 we have what is called a population inversion. It should be apparent that with only two energy levels we can never achieve a population E2 > E1 because, in the steady state, the incoming photon flux will cause as many upward excitations as downward stimulated emissions.
Optical Pumping
Suppose that an external excitation causes the atoms in a system to become excited to some higher energy level. This process is called pumping. Optical pumping is when the external source of excitation if incident photons. The original ruby laser used a xenon flashlamp to provide this pumping (external excitation) energy. See the picture on page 162. The energy to which the system is pumped is usually a short lived energy state and the atom rapidly decays to a nearby energy level that is long lived (there is a lot of work involved in finding a system where this is the case!). Here the atoms accumulate at this meta-stable state. Either a spontaneous emission or other random photon can then initiate the stimulated process with other atoms at this energy level.
Optical Pumping
h32 E3 h13 E2
Metastable state
Lasing Emission
E3 E2 E3 E3 E2
E2
IN
OUT
h21 E1 (b)
h 2 E1
Coherent photons
The emission from E2 to E1 in the previous figure is called the lasing emission. By trapping nearly all of these photons in an optical cavity, the intensity builds up in much the same way as we build up voltage oscillations in an electrical oscillator circuit. What leaks out of the optical cavity is a highly coherent radiation at high intensity. It is this coherency of a well defined wavelength (same energy), same polarization, and same direction that makes laser light distinctly different from random sources such as a filament light or LEDs.
E1 (a)
E1 (c)
(d)
The principle of the LASER. (a) Atoms in the ground state are pumped up to the energy level E3 by incoming photons of energy h13 = E3 E1 . (b) Atoms at E3 rapidly decay to the metastable state at energy level E2 by emitting photons or emitting lattice vibrations; h32 = E3 E2 . (c) As the states at E2 are long-lived, they quickly become populated and there is a population inversion between E2 and E 1 . (d) A random photon (from a spontaneous decay) of energy h21 = E2 E1 can initiate stimulated emission. Photons from this stimulated emission can themselves further stimulate emissions leading to an avalanche of stimulated emissions and coherent photons being emitted. 1999 S.O. Kasap, Optoelectronics (Prentice Hall)
Section 4.2 Stimulated Emission rate and Einstein Coefficients A useful LASER medium must have a higher efficiency of stimulated emission as compared to the efficiencies of the system in allowing spontaneous emission and absorption. Consider a medium that has N1 atoms per unit volume with energy E1 and N2 atoms per unit volume with energy E2. The rate of upward transitions from E1 to E2 by photon absorption will be proportional to the number of atoms N1 and also to the number of photons per unit volume with energy h = E2 E1. This number of photons per unit volume with energy h is called the energy density in the radiation.
Stimulated Emission Rate Thus the upward transition rate is,
R12 = B12 N1 (h )
where B12 is a proportionality constant called the upward Einstein B12 coefficient. The (h) term is the photon energy density per unit frequency which represents the number of photons per unit volume with an energy of h.
Stimulated Emission Rate The rate of downward transitions from E2 to E1 involves both the spontaneous and stimulated emissions. The spontaneous rate of downward transitions from E2 to E1 depends on the concentration of N2 atoms which are at the energy E2. The stimulated rate of downward transitions from E2 to E1 depends on both the concentration of N2 atoms which are at the energy E2 and the photon concentration (h) with energy h = E2 - E1. Thus the total downward transition rate is
Stimulated Emission Rate To find the coefficients A21 (spontaneous emission), B12 (absorption) and B21 (stimulated emission), we will consider the medium under thermal equilibrium. In equilibrium, there is no net change with time in the populations at E1 and E2, thus R12 = R21. Upward transitions (absorption) = Downward transitions (emission). In thermal equilibrium at room temperature and above (> 3 kbT), Boltzmann statistics state:
N2 = e N1 E2 E1 kBT
R21 = spontaneous rate + stimulated rate = A21 N 2 + B21 N 2 (h )

Where A21, B12 and B21 are called the Einstein coefficients.
Plancks Black Body Radiation

In thermal equilibrium, radiation from the atoms must give rise to an equilibrium photon energy density eq(h) that is described by Plancks black body radiation distribution law.
Stimulated Emission
Under thermal equilibrium, R12 = R21 which means
B12 N1 (h ) = A21 N 2 + B21 N 2 ( h )

Under these conditions, it can be readily shown that B12 = B21 and also
eq (h ) =
8 h 3
(c ) e
3
h k BT
A 21 8 h 3 = B21 c3
Now consider the ratio of stimulated to spontaneous emission.
The above equation is for thermal equilibrium which is what a laser is NOT in during operation. We will use the law to help us determine the coefficients only.
R21 ( stim) B21 N 2 (h ) B21 (h ) B21 c3 = = = (h ) = (h ) R21 ( spon) A21 N 2 A21 A21 8 h 3
Stimulated Emission
Stimulated Emission
R21 ( stim) c3 = (h ) R21 ( spon) 8 h 3
R21 ( stim) N = 2 R12 (absorp ) N1
There are two important conclusions from the last two slides: The ratio of stimulated emission to absorption is
R21 ( stim) B N (h ) N 2 = 21 2 = R12 ( absorp ) B12 N1 ( h ) N1
For stimulated photon emission to exceed photon absorption , we need to achieve population inversion that is N2 > N1. For stimulated emission to far exceed spontaneous emission, we must have a large photon concentration (h) which is achieved by building an optical cavity to contain the photons. Lastly, recall that the laser is based on non-thermal equilibrium so the above results are qualitatively only.
Section 4.3 Optical Fiber Amplifiers

A light signal that is traveling along an optical fiber over a long distance suffers marked attenuation. It becomes necessary to regenerate the light signal at certain intervals for long haul communications. Classic amplification: regenerating the optical signal by photodetection, conversion to an electrical signal, clean up and amplification and then conversion back from an electrical to light energy by a laser diode. Is has become practical to amplify the signal directly by using an optical amplifier.
Optical Fiber Amplifiers

One practical optical amplifier is based on the erbium (Er3+ ion) doped fiber amplifier (EDFA). A short stretch of rare earth dopants is fused to a single mode long distance optical fiber. The erbium (Er3+ ion) implanted fiber has the following energy levels.
Energy of the Er3 + ion in the glass fiber
1.54 eV 1.27 eV
E3 E3
Non-radiative decay
980 nm
Pump 0.80 eV
1550 nm
E2
1550 nm
Out In
E1
EDFA
Energy of the 3+ ion Er in the glass fiber
Section 4.4 Gas Lasers: The He-Ne LASER

E3 E3
Non-radiative decay
1.54 eV 1.27 eV
After optical pumping, there fiber has a long lived population inversion at E2 = 0.80 eV.
980 nm
Pump
0.80 eV 1550 nm
E2
1550 nm
Out In
With the HeNe Laser one has to confess that the actual explanation is by no means simple since we have to know such things as the energy states of the whole atom. The actual stimulated emission occurs from the neon atoms. The helium are used to excite the Ne atoms by atomic collisions. states of the whole atom. The text discusses the effect of exciting the outer shells of both the neon and helium atoms. The important fact is that both can be excited to higher energies by the excitation of only one electron in the outermost shell.
E1
Incident photons at the 1550 nm wavelength will achieve stimulated emission as long as the pumping continues. If the optical pump fails, this systems will see attenuation of the signal at the EDFA.
Er3+-doped fiber (10 - 20 m) Signal in
Optical isolator Wavelength-selective coupler Splice Splice
Optical isolator
Signal out
= 1550 nm
= 1550 nm
Pump laser diode = 980 nm
Termination
The He-Ne LASER

The excitation of the He atom by an electron collision puts the excited He atom into a metastable (long lasting) state. There are momentum arguments in the text, but the result is that a large number of excited He atoms build up during the electrical discharge because they are not allowed by the quantum mechanics to simply decay back to the ground state. When an excited He atom collides with a Ne atom, it transfers its energy to the Ne atom by electronic resonance energy exchange since it has a state equal to the exited He electron state. The collision excites the Ne and de-excites He down to the ground level energy.
The He-Ne LASER

With a large number of He-Ne collisons in the gaseous discharge, we end up with a large number of excited Ne atoms and a population inversion between the two outermost electron states of the Ne atom.
He
20.61 eV
Ne
Collisions
(1s 1 2s1 )
(2p5 5s1 )
20.66 eV 632.8 nm Lasing emission
(2p5 3p1 )
Fast spontaneous decay ~600 nm Electron impact
(2p5 3s1 )
Collisions with the walls
(1s 2 )
Ground states
(2p6 )
The principle of operation of the He-Ne laser. He-Ne laser energy levels (for 632.8 nm emission).
1999 S.O. Kasap, Optoelectronics (Prentice Hall)
Example 4.4.1 Efficiency of the HeNe Laser A typical low power 5 mW HeNe laser tube operates at a dc voltage of 2,000 V and carries a current of 7 mA. What is the efficiency of the laser? Solution:
Efficiency = Output Light Power 5 103W = = 0.036% Input Electrical Power (7 103 A)(2, 000V )
Example 4.4.2 Laser Bean divergence The laser beam emerging fram a laser tube has a certain amount of divergence. A typical HeNe Laser has an output beam with a diameter of 1 mm and a divergence of 1 mrad. What is the diameter of the beam at a distance of 10 m?
Laser radiation
Laser tube L
Typically HeNe efficiencies are less than 0.1%. What is important is the high concentration of coherent photons. The 5 mW over a beam diameter of 1 mm is 6.4 kW m-2.
Example 4.4.2 Laser Bean divergence

Laser tube
Laser radiation
Section 4.5 The Output Spectrum of a Gas Laser

The output radiation from a gas laser is not one single well-defined wavelength corresponding to the lasing transition, but covers a spectrum of wavelengths with a central peak. This broadening of the emitted spectrum is a direct result of the Doppler effect. The gas atoms/molecules are in random motion with an average kinetic energy of 3/2 kBT. Thus when the a gas atom is moving away, the observer detects a lower frequency. Let 1 = the shifted frequency, 0 = the frequency of the energy transition, and velx = the velocity of the gas atom presumed to be away from the observer.
Solution:
We can assume that the laser beam emanates like a light-cone as shown above with an apex angle of 2. The angle 2 is then the total divergence of the beam. Thus 2 = 1 mrad. If r is the increase in the radius of the beam over a distance L then
tan =
r 1 r = L tan = (10m) tan( 103 rad ) = 5mm 2 L
The spot size is then 2w0 + 2 r = 1mm + 2(5mm) = 11mm.
1 = 0 (1
velx ) c
The Output Spectrum of a Gas Laser

When the atom is moving towards the observer along the laser tube (xaxis) the detected frequency is higher.

The velocities of a gas obey the Maxwell distribution. Thus the stimulated emission wavelengths in the lasing medium exhibit a distribution about a central wavelength 0 = c/0. For the Doppler broadened emission, the lineshape is a Gaussian function. The linewidth in terms of Full Width at Half Maximum (FHWM) is
2 = 0 (1 +
velx ) c
Since the atoms are in random motion the observer will detect a range of frequencies due to this Doppler effect. The frequency or wavelength of the output radiation from a gas laser will have a linewidth = 2 - 1.
1,2 = 2 0
2k BT ln(2) Mc 2
Where M is the mass of the lasing atom or molecule.

Optical Gain
(a)
Doppler broadening

Consider an optical cavity of length L with parallel end mirrors.
L

Only standing waves with certain wavelengths however can be maintained within the optical cavity. Any standing wave in the cavity must have an integer number of halfwavelengths /2 that fit into the cavity length L.
Mirror
Stationary EM oscillations Mirror
This style of optical cavity is called a Fabry-Perot optical resonator or etalon. The reflections from the end mirrors of a laser give rise to traveling waves in opposite directions within the cavity. These oppositely traveling waves interfere constructively to set up a standing wave.
m = L 2
Where m is an integer called the mode number of the standing wave.

For most gases the index of refract is very nearly n = 1. When this is not the case, the wavelength must be modified to reflect the actual wavelength within the cavity. Or in other words, do not use the free space wavelength inside a optical cavity unless you know that it applies. Each successful (constructive interference) standing wave within the laser tube is called a cavity mode.
Allowed Oscillations (Cavity Modes) m(/2) = L

Modes that exist along the cavity axis are called axial (or longitudinal) modes. This is the only type of mode that can exist when the end mirrors are flat. Any off-axis mode will walk off the mirror very quickly and not see constructive interference. When the mirrors are not flat, other modes may exist. An example is an optical cavity formed by confocal spherical mirrors.
L Spherical mirror Optical cavity F A
Spherical mirror
(b)
2 R
R Wave front

The gas laser output thus has a relatively broad spectrum with peaks at certain wavelengths corresponding to various cavity modes existing within the Doppler broadened optical gain curve.
Relative intensity
Example 4.5.1 Doppler Broadened Linewidth Calculate the Doppler broadened linewidths in terms of both frequency and wavelength for HeNe Laser with = 632.8 nm The gas discharge temperature is 127C = 400 K Ne atomic mass = 20.3 g mol-1 Laser tube length = 40 cm 1. 2. 3. 4. What is the linewidth in the output wavelength spectrum? What is the mode number m of the central wavelength? What is the separation between two consecutive modes? How many modes do you expect within the linewidth 1,2 of the optical gain curve?
(c)
The net envelope of the output radiation is a Gaussian distribution. The frequency width of an individual spike has a finite width due to the losses at the mirrors (the Finesse of the cavity is finite), and other nonidealities of the cavity.
Example 4.5.1 part 1. 1. What is the linewidth in the output wavelength spectrum? The central frequency 0 is
0 =
c = 3 108 = 4.74 1014 Hz = 474THz 632.8nm
Example 4.5.1 part 1. 1. What is the linewidth in the output wavelength spectrum? The FWHM wavelength 1,2 is
1,2 1,2
The observed FWHM width of the frequencies 1,2 is given by

1,2 = 2 0 2k BT ln(2) 2(1.38 1023 )(400) ln(2) = 2(474THz ) = 1.51GHz Mc 2 (3.35 1026 )(3 108 ) 2
= (1.51109 Hz )
632.8nm = 2.02 pm 4.74 1014 Hz
This last value is different by about 18% of that determined from the rms velocities of the gas which confirms that other processes are broadening the linewidth as well.
In problem 4.2, you are asked to find the minimum and maximum wavelengths corresponding to the extremes of the spectrum at the half-power points.
min = 0 1,2 = 632.8nm .00101nm = 632.79899nm
1 2 1 = 0 + 1,2 = 632.8nm + .00101nm = 632.80101nm 2
max
Example 4.5.1 part 2. 2. What is the mode number m of the central wavelength?
Example 4.5.1 part 3. 3. What is the separation between two consecutive modes? The separation m is the separation between two consecutive modes m and m + 1.
m = m m+1 =
2L 2L 2L since m is a very large number. m m + 1 m2
2(40cm) m0 = = = 1, 264, 222.5 0 632.8 1012 cm
2L
Since partial modes do not constructively interfere in the cavity the nearest full mode to the wavelength given is m0 = 1,264, 222.
Also since m = 2L/ we can find the separation of modes just in terms of the cavity length and the central wavelength.
m = 2 (632.8 109 m)2 2L 2L = = 0 = = 0.501 pm m 2 2 L 2 2 L 2(40 102 m) 0
When the index of refraction is other than 1, we must include that in 2 the equation. = 0
m
2n 2 L
Example 4.5.1 part 4. 4. How many modes do you expect within the linewidth 1,2 of the optical gain curve? The number of modes between the half-intensity points will depend on have the cavity modes and the optical gain curve overlap.
Optical gain between FWHM points
Example 4.5.1 part 4. 4. How many modes do you expect within the linewidth 1,2 of the optical gain curve? Without going into the specifics of the optical gain curve, we can estimate the number of modes in the linewidth by
Modes =
5 modes
Number of laser modes depends on how the cavity modes intersect the optical gain curve. In this case we are looking at modes within the linewidth 1/2.
Cavity modes
Linewidth of spectrum 2.02pm 1,2 = = 4.03 Separation of two modes m 0.501pm
(a)
Thus we can expect 4 (or maybe 5) modes within the linewidth of the output.
(b)
4 modes
Example 4.5.1 part 4. 4. How many modes do you expect within the linewidth 1,2 of the optical gain curve? (exact answer)
max = 632.80101nm
Mode = m in nm min < m < max? 1,264,220 632.80125 No 1,264,221 632.80075 Yes 1,264,222 632.80025 Yes 1,264,223 632.79975 Yes
Section 4.6 Laser Oscillation Conditions Optical Gain Coefficient g Consider a general laser medium which has an optical gain for coherent radiation along some direction x.
h E2 E1
min = 632.79899nm
1,264,224 632.79925 Yes 1,264,225 632.79875 No
Optical Gain g( o) x g( )
Laser medium
This table is similar to that in homework problem 4.2. By comparing of individual modes, we can see the exact number of modes in the linewidth. In this case the number of modes = 4.
P+ P
x
(a)
(b)
(a) A laser medium with an optical gain (b) The optical gain curve of the medium. The dashed line is the approximate derivation in the text.
Optical Gain Coefficient g Consider an EM wave propagating in the medium along the x direction. As it propagates its power (energy flow per unit time) increases due to the greater stimulated emissions (gain) over (losses) spontaneous emissions and absorption. The power increases as exp(gx) where g is the optical gain per unit length and is called the optical gain coefficient of the medium. The gain coefficient g is defined as the fractional change in the light power (or intensity) per unit distance.
Optical Gain Coefficient g Optical power P along x at any point in the cavity is proportional to the concentration of coherent photons Nph and their energy hv. These coherent photons travel with a velocity c/n where n is the refractive index of the cavity. This in time t they travel a distance x = (c/n) t. Thus the gain coefficient becomes
g=
P 1 N ph 1 N ph 1 n N ph = = = P x N ph x N ph c cN ph t t
n
Optical Gain Coefficient g The difference between stimulated emission and absorption rates gives the net rate of change in the coherent photon concentration.
dN ph dt = Net rate of stimulated photon emission
Optical Gain Coefficient g The emission and absorption processes are distributed in photon energy by Doppler broadening, energy band spreading and other processes. This means that the optical gain will reflect this distribution.
dN ph dt
= N 2 B21 (h ) N1 B12 (h ) = ( N 2 N1 ) B21 (h )

h E2 E1
Where B12 = B21 under thermal equilibrium.

Laser medium
Optical Gain g( o) x g( )
We can now use this result for dNph/dt.
n g= ( N 2 N1 ) B21 (h ) cN ph
P+ P
Optical Gain Coefficient g We can express (h) in terms of Nph by noting that (h) is the radiation energy density per unit frequency.
Threshold Gain gth Consider an optical cavity with mirrors at the ends.
Reflecting surface 2 R2 Pf Ef Pi Ei Steady state EM oscillations 1 Reflecting surface Cavity axis R1 x
(h 0 )
N ph h 0
Now we can write the optical gain coefficient in terms of photon concentrations and energies. N B h B nh n n g ( 0 ) = ( N 2 N1 ) B21 (h ) ( N 2 N1 ) ph 21 0 = ( N 2 N1 ) 21 0 cN ph c N ph 0 c 0 The last equation gives the optical gain of the medium at the center frequency 0. It takes a more rigorous derivation to find the lineshape over the of a real laser.
The cavity contains a laser medium so that lasing emissions build up to a steady state and we have continuous operation. The optical cavity acts as an optical resonator. Consider an EM wave with an initial optical power Pi starting at some point in the cavity and traveling in a round trip and arrive back at the starting point with a final optical power Pf.
Threshold Gain gth
Reflecting surface 2 R2
Pf Ef
Pi Ei Steady state EM oscillations 1
Reflecting surface Cavity axis R1 x
Threshold Gain gth
Reflecting surface 2 R2
Pf Ef
Pi Ei Steady state EM oscillations 1
Reflecting surface Cavity axis R1 x
Under steady state conditions, oscillations do not build up and do not die out. Thus Pf must be the same as Pi. Then there should be no optical power loss in the round trip and that the net round trip gain Gop =1.
Gop = Pf Pi = 1 for steady state.
There are other losses such as absorption and scattering during propagation in the medium. The decrease in power is proportional to
Power Loss e x
Where is the attenuation (or loss) coefficient of the medium. Represents all losses in the cavity and its walls except for light transmission losses through the end mirrors and absorption losses already accounted for by the gain coefficient g.
Reflections at the faces 1 and 2 reduce the optical power since the reflectance R1 and R2 are less than one.
Threshold Gain gth

All the losses have to be made up by stimulated emissions in the optical cavity (which is the optical gain in the medium). As the wave propagates, its power increases as
Threshold Gain gth

For steady state oscillations Gop = Pf/Pi = 1 must be satisfied. The value of the gain coefficient g that makes Pf/Pi = 1 is called the threshold gain gth.
Power Gain e gx
The power Pf of the EM radiation after one round trip of path length 2L is given by
gth = +
1 1 ln 2 L R1 R2
Pf = PR1 R2 e g (2 L ) e (2 L ) i
The equation gives the optical gain needed in the medium to achieve a continuous wave lasing emission. The necessary gth has to be obtained by suitably pumping the medium so the N2 is sufficiently greater than N1.
Threshold Gain gth

The threshold population inversion N2 N1 = (N2 N1)th.
Threshold Gain gth

It should be apparent that the laser device emitting coherent emissions is actually a laser oscillator.
(N2 N1) and Po
( N1 N 2 )th gth
c from equation (4) on page 176 B21nh 0
Initially the medium must have a gain coefficient g greater than gth. This allows the oscillations to build-up in the cavity until a steady state is reached when g = gth. The reflectance of the mirrors R1 and R2 are important in determining the threshold population inversion required as the control gth in shown before: 1 1 gth = + ln 2 L R1 R2
The figure to the right shows the steady state output power P0 and the population difference N2 N1.
N2 N1
Po = Lasing output power Threshold population inversion
(N2 N1)th
Pump rate
Threshold pump rate
Until the pump rate can bring (N2 N1) to the threshold value (N2 N1)th, there would be no coherent radiation output.
Threshold Gain gth

When the pumping rate exceeds the threshold value, then N2 N1 remains clamped at (N2 N1)th, because the value of g must equal gth. This above fact may be difficult to accept. But remember that we are considering a system that has specific energy configurations with a large but limited number of available quantum states.
(N2 N1) and Po (N2 N1)th N2 N1
Po = Lasing output power Threshold population inversion
Example 4.6.1 Threshold population
N th gth
2 8 n 2 0 sp
c2
Pump rate
Threshold pump rate
Now consider a HeNe laser with wavelength 632.8 nm. Tube length L = 50 cm Mirror Reflectances R1 = 100% and R2 = 90% Linewidth = 1.5 GHz Loss coefficient = 0.05 m-1 Spontaneous decay time constant sp = 300 ns Index of refraction n = 1 Calculate the threshold population inversion. Solution: The emission frequency is (book is in error).
0 =
c 3 108 = 4.74 1014 Hz 632.8 109
Additional pumping (above that required for threshold) increases the rate of stimulated transitions (see B21 in equation (8) on pg 177) and hence increase the optical output power P0.
N th gth
2 8 n 2 0 sp

Now consider a HeNe laser with wavelength 632.8 nm. Spontaneous decay time constant sp = 300 ns Index of refraction n = 1 The required population inversion is
2
Now consider a HeNe laser with wavelength 632.8 nm. Tube length L = 50 cm Mirror Reflectances R1 = 100% and R2 = 90% Linewidth = 1.5 GHz Loss coefficient = 0.05 m-1 Spontaneous decay time constant sp = 300 ns Index of refraction n = 1 The required gain is
gth = +
c2
c2 8 (1) (4.74 10 ) (300 10 )(1.5 109 Hz ) Nth = (0.155m 1 ) (3 108 ) 2

14 2 9
Nth gth
2 8 n 2 0 sp
1 1 1 1 1 1 ln ln = 0.05m + = 0.05 + 0.01053 = 0.1554m 2L R1R2 2(50102 m) 1 0.9
= 4.376 1015 m 3
Phase Condition and Laser Modes

The discussion on threshold gain considered only the intensity inside the cavity. But the same round trip gain dialogue shows that there is a phase condition as well.
Reflecting surface 2 R2 Pf Ef Pi Ei Steady state EM oscillations 1 Reflecting surface Cavity axis R1 x

Thus
kmedium = 2 nkmedium (2 L) = 2 n2 L
round trip = m(2 )
Since any multiple of 2 also satisfies the phase condition we can rewrite the last equation with m(2) where m is an integer 1, 2,
nkmedium (2 L) = m(2 )
Now with a little more algebra and recalling = 2/k

nkmedium (2 L) = m(2 ) n(2 L) = m(
The total phase change after one round trip from Ei to Ef must be multiple of 2.
round trip = m(2 )
2 ) L = m( medium ) 2n kmedium
m(
medium
2n
) = L where m is the approximate number of cavity modes.

These modes are controlled by length L of the optical cavity along its axis and are called longitudinal axial modes. The analysis so far assumed the ideal case and assumed the wave inside the cavity was a perfect plane wave. All practical laser cavities have a finite transverse size and often use spherical mirrors to aid alignment of the cavity.
Optical cavity

The electric field description of these modes are called transverse modes or transverse electric and magnetic modes (TEM).
TEM00 Optical cavity TEM10 TEM00 TEM10
(a) (a)
Spherical mirror Spherical mirror
(b) (b)
Wave fronts Wave fronts TEM01 TEM11 TEM01 TEM11
(c)
(d)
With spherical mirrors there can be off-axis self-replicating rays such as (a) in the figure above. The modes represent a particular electric field pattern in the cavity that can replicate after one round trip.
Each allowed mode corresponds to a distinct spatial field distribution at a reflector.
Mode Indexing
These modal field patterns can be described by three integers p, q, m And designated as TEMpqm.
TEM00 TEM10 TEM00 TEM10
Mode Indexing
The TEM00 has an intensity distribution that is radially symmetric about the cavity axis and has a Gaussian intensity distribution across the beam cross section. It has the lowest divergence angle and is thus a highly desirable field distribution in most applications.
The integer m is the number of modes along the cavity axis x and is the usual longitudinal mode number. m is usually large and not written. Thus we write the TEMpq descriptions where p and q describe the spatial distribution at the face of the exit reflector.
(c)
(d)
In this figure the x direction is coming straight at the observer. The y direction is horizontal (left to right).The z direction is vertical (down to up).
(c)
(d)
p is the number of nodes in the transverse y direction. q is the number of nodes in the transverse z direction. m is the number of modes along the x direction.
Thus TEM10 has one intensity reversal in the y direction (two separate intensities). Thus TEM01 has one intensity reversal in the z direction (two separate intensities).
Section 4.7 Principle of the Laser Diode

Consider a degenerately doped direct bandgap semiconductor pn junction
p+
Junction
Principle of the Laser Diode

The depletion region in such a pn junction is very narrow. There is a built-in voltage V0 that gives rise to a potential energy barrier eV0 that prevents electrons in the CB of the n+-side diffusing in the CB of the p+-side.
p+
Ec Eg Ev EF p
Ho les in V B El ectro ns
n+
Junction
n+
The Fermi level EFp in the p-side is in the valence band. EFn in the n-side is in the conduction band. All energy levels up to the Fermi level are assumed occupied by electrons
Ec Eg Ev EF p
Ho les in V B El ectro ns
eV o EF n Ec
eV o EF n Ec
Electro ns i n C B
Electro ns i n C B
(a)
Ev
There is similar potential energy barrier stopping hole diffusion from the p+-side to the n+-side.
(a)
Ev
In the absence of an applied voltage, the Fermi level is continuous across the diode. EFp = EFn.

The diode is now forward biased with an applied voltage V greater than the bandgap voltage: eV > Eg. This applied voltage separates EFn and EFp by the applied voltage eV. Since the diode is forward biased, the potential barrier is reduced to almost zero. The electron majority carriers are injected through the depletion region and become minority carriers in the p+-side of the diode.
p+
Ec Eg
In v ers io n reg i o n

In the depletion region, there are more electrons in the conduction band at energies near Ec than electrons in the valence band near Ev. This is the required population inversion.
Energy
CB
EF n Ec eV Ev EF p
Holes in VB = Empty states Electrons in CB
n+
EF n Ec eV EF p
VB
Density of states
p+ n+
EF n Ec eV EF p Ec
(b)
The holes have a similar result. Thus we have a diode current due to the movement of both charge carriers.
This population inversion region is a layer along the metallurgical junction and is called the inversion layer or (more commonly) the active region.
Eg
(b)

With some few (relatively) electrons in the valence band in the active region, an incoming photon with an energy Ec Ev suffers little absorption. So the incoming photon can stimulate an electron to fall down from Ec to Ev.
Ec
Energy

Electrons in CB
Optical gain
CB
EF n Ec eV Ev EF p
Holes in VB = Empty states
EF n EF p
Photons with energy between Egap and EFn EFp cause stimulated emissions. Photons outside this energy range are absorbed since there are no available states in the system in the energy range. It is apparent that the population inversion between energies near Ec and those near Ev is achieved by the injection of carriers across the junction under a sufficiently large forward bias.
0
h Eg
At T > 0 At T = 0
VB
Density of states
p+
Optical absorption
n+
EF n Ec eV EF p
p+
Ec Eg
n+
EF n Ec eV EF p
The region where there is a population inversion and hence more stimulated emission than absorption has an optical gain. The optical gain depends on the photon energy.
Eg
(b)
The pumping mechanism is the forward diode current and the pumping energy is supplied by the external voltage. This is called injection pumping.
(b)
Current

In addition to population inversion we also need to have an optical cavity to implement laser oscillation.
Current Cleaved surface mirror

The pn junction uses the same direct bandgap semiconductor material throughout and hence has the name homojunction.
Cleaved surface mirror
L p+
L
Electrode
GaAs
n+
GaAs
Electrode Active region (stimulated emission region)
L p+
L
Electrode
The ends of the crystal are cleaved to be flat and optically polished to provide reflection and hence form an optical cavity. For example, the index of refraction for GaAs is about 3.6 so the reflectance is
GaAs
n+
GaAs
n n 3.6 1 2.6 R= 1 2 = = = 0.31947 n1 + n2 3.6 + 1 4.6

2 2
A schematic illustration of a GaAs homojunction laser diode. The cleaved surfaces act as reflecting mirrors.
Current
Current

L p+
L Electrode
GaAs
n+
GaAs
The required threshold gain is

gth = + 1 1 1 1 1 ln = + ln = + (2.282) 2L R1R2 2L 0.1021 2L
The dependence of the optical gain of the medium on the wavelength of the radiation can be seen from the energy distribution of the electrons in the CB and the holes in the VB around the metallurgical junction.
L p+
L
Electrode
GaAs
n+
GaAs
Lasing radiation is obtained when the optical gain in the medium can overcome the photon losses from the cavity. Below the threshold current Ith, the light from the device is due to spontaneous emission (and not stimulated emission). So below the threshold current Ith, the light from the device is random and like an LED. This is the basis for the superluminescent LEDs.
For the gas laser with mirror reflectances of 100% and 90% we had
gth = + 1 1 1 1 1 ln = + ln = + (0.1054) 2L R1R2 2L 0.9 2L
Thus the semiconductor diode requires a considerable gain through pumping to achieve threshold.
Current

There are two critical diode currents. First is the diode current that provides just sufficient injection to lead to stimulated emission just balancing absorption.

L p+
L Electrode
GaAs
n+
GaAs
The following figure shows the output light intensity as a function of diode current. LED like below Ith, then after threshold a Laser.
Optical P ower
Laser
This is called the transparency current Itrans since there is no net photon absorption and the medium appears to be transparent. Above Itrans there is optical gain. When the diode current reaches Ith the gain has reached g = gth. By definition then, the cavity continues to support stimulated emission since all losses are covered.
Optical Power
Optical P ower
LED
Stimulated emission
Optical P ower
Laser
Spontaneous emission
I
Ith
Typical output optical power vs. diode current (I) characteristics and the corresponding output spectrum of a laser diode.

Note that the output frequency spectrum may depend on the diode current. Later we will review mode hopping when the spectrum shifts suddenly.
Optical P ower

The main problem with the homojunction laser diode is that the required threshold current is very large. The current density Jth is too high for practical purposes (low life of the device, high heat, low efficiencies)
Optical P ower
Laser
Optical Power
Laser
Optical Power
Optical P ower
LED
Stimulated emission
Optical P ower
Optical P ower
LED
Stimulated emission
Optical P ower
Laser
Laser
I
Ith
I
Ith
Section 4.8 Heterostructure Laser Diodes

The reduction of the threshold current Ith to a practical value (i.e.. Those not needing cryogenic cooling) requires improving the rate of stimulated emission and improving the efficiency of the optical cavity. The reduction of the threshold current Ith can be achieved by confining the injected electrons and holes to a narrow region around the junction. Confining the carriers to a small region means that less current is needed to establish the necessary concentration of carriers for population inversion. Secondly, we can build a dielectric waveguide around the optical guide region to increase the photon concentration and hence the probability of stimulated emissions.
Heterostructure Laser Diodes

Thus we need both carrier confinement and photon confinement.
n (a) AlGaAs p GaAs (~0.1 m) Electrons in CB Ec
2 eV
p AlGaAs
(a) A double heterostructure diode has two junctions which are between two different bandgap semiconductors (GaAs and AlGaAs).
Ec
2 eV
Ec
1.4 eV
(b) Ev Holes in VB
Ev
(b) Simplified energy band diagram under a large forward bias. Lasing recombination takes place in the pGaAs layer, the active layer (c) Higher bandgap materials have a lower refractive index
Refractive index
(c)
Photon density
Active region
n ~ 5%
(d)
(d) AlGaAs layers provide lateral optical confinement.
) Ec
1.4 eV

n (a) AlGaAs p GaAs (~0.1 m) p AlGaAs
Electrons in CB Ec
2 eV
Ec
2 eV
(b) Ev Holes in VB
Ev
The p-GaAs region is a thin layer, typically 0.1 0.2 m. It is the active layer in which lasing recombination takes place. Both p-GaAs and p-AlGaAs regions are heavily p-type doped and are degenerate with EF in the valence band. When a sufficiently large forward bias is applied, EC on n-AlGaAs moves above EC of p-GaAs which leads to a large injection of electrons in the CB on n-AlGaAs in p-GaAs.
Electrons in CB Ec
2 eV
The injected electron are confined to the CB of the p-GaAs region by the barrier EC. The barrier EC. Is due to the change in the bandgap which arises from the change in doping. The small p-GaAs active region allows for the concentration of injected electrons to be increased quickly even with moderate increases in forward current. This is the carrier confinement requirement and thus the threshold current for population inversion is reduced.
Ec
1.4 eV
Ec
2 eV
(b) Ev Holes in VB
Ev
) Ec
1.4 eV

A wider bandgap semiconductor generally has a lower refractive index.
Electrons in CB Ec
2 eV
Ec
2 eV
(b) Ev Holes in VB
Ev
Refractive index
(c)
Ph
Active region
n ~ 5%
AlGaAs has a lower refractive index (higher bandgap) than GaAs. This change in the refractive index defines an optical dielectric waveguide that confines the photons to the active region of the optical cavity. We now have reduced photon losses and increase photon concentration. Recall that the photon confinement was to aid the optical resonance (oscillation). Light is still allowed to leak out through the Refractive p-AlGaAs. index
(c)
Thus both carrier and optical confinement lead to a reduction in the threshold current density.
n (a) AlGaAs p GaAs (~0.1 m) Electrons in CB Ec
2 eV
p AlGaAs
Ec
1.4 eV
Ec
2 eV
(b) Ev Holes in VB
Ev
Refractive index
(c)
Photon density
Active region
n ~ 5%
Photon density
Active region
n ~ 5%
(d)
(d)
Double-Heterostructure Laser Diodes

Without double-heterostructure devices we would not have practical solid state lasers that can be operated continuously at room Cleaved reflecting surface temperature. W
Stripe electrode Oxide insulator p-GaAs (Contacting layer) p-AlxGa1-xAs (Confining layer) p-GaAs (Active layer) n-AlxGa1-xAs (Confining layer) n-GaAs (Substrate)
2 1 Current paths Substrate 3
Double-Heterostructure Laser Diodes

Stripe electrode Oxide insulator p-GaAs (Contacting layer) p-Al xGa1-xAs (Confining layer) p-GaAs (Active layer) n-Al xGa1-xAs (Confining layer) n-GaAs (Substrate)
Cleaved reflecting surface W
The doped layers are grown epitaxially on a crystalline substrate.
2 1 Current paths Substrate
Substrate Electrode
Elliptical laser beam
Cleaved reflecting surface Active region where J > Jt h. (Emission region)
The term epitaxy (greek; "epi" "above" and "taxis" "in ordered manner") describes an ordered crystalline growth on a (single-) crystalline substrate.
Substrate Electrode
As seen from the figure above, many layers are used to form the slab waveguide and to allow the tailoring of current flow through the device. In particular, the stripe electrode down the top center of the device creates the current density down through the device. Such current flow channeling is called a gain guided device.
Elliptical laser beam
Cleaved reflecting surface Active region where J > Jth. (Emission region)
Schematic illustration of the the structure of a double heterojunction stripe contact laser diode
Buried Double-Heterostructure Laser Diodes

The previous double-heterostructure device did little to channel the photons laterally down the slab waveguide.
Oxide insulation p+-AlGaAs (Contacting layer) p-AlGaAs (Confining layer) n-AlGaAs p-GaAs (Active layer) n-AlGaAs (Confining layer)
Electrode
Example 4.8.1 Modes in a laser and the optical cavity length

Consider an AlGaAs based heterostructure laser diode. Optical cavity length = 200 m Peak free space radiation wavelength = 900 nm. Index of refraction n = 3.7 What is the mode integer m of the peak radiation?
2n 2(3.7)(200 106 m) m = L m = L= = 1644.4444 900 109 m 2n
n-GaAs (Substrate)
Schematic illustration of the cross sectional structure of a buried heterostructure laser diode.
A modified device is called the buried double-heterostructure. The sides of the active area are also doped to reduce the index of refraction so the the photons are now guided down a box rather than a slab. Since the optical power is confined to the waveguide defined by the refractive index variation, these diodes are called index guided.
Partial modes decay out so m = 1644.

Consider an AlGaAs based heterostructure laser diode. Optical cavity length = 200 m Peak free space radiation wavelength = 900 nm. Index of refraction n = 3.7 What is the separation between modes in the cavity?
2n m = L = L m 2n

Consider an AlGaAs based heterostructure laser diode. Optical cavity length = 200 m Peak free space radiation wavelength = 900 nm. Index of refraction n = 3.7 Given optical gain vs 1,2 = 6 nm. How many modes are there within this bandwidth? In earlier examples, we showed that the distribution can be important in the exact number of modes. For this problem, we will assume the modes are centered at the peak wavelength and evenly distributed left and right of peak. 6nm = 10.96 number of modes = 1,2 = m 0.5473nm Number of modes = 10
m =
2nL 2nL 2nL(m + 1) 2nL(m) 2nL 2nL 2nL = = 2 = = m m +1 m(m + 1) m + m m 2 2nL 2 2nL 2 (900 109 m)2 = = .05473nm 2nL 2(3.7)(200 106 m)
2

Consider an AlGaAs based heterostructure laser diode. Optical cavity length = 200 m Peak free space radiation wavelength = 900 nm. Index of refraction n = 3.7 Given optical gain vs 1,2 = 6 nm. How many modes are there within this bandwidth if the cavity shrinks to 20 m?
m
Section 4.9 Laser Diode Characteristics

The output spectrum from a laser diode (LD) depends on two factors: the nature of the optical resonator used to build the laser oscillations and the optical gain curve (lineshape) of the active medium. The optical resonator is essentially a Fabry-Perot cavity which can be assigned a length (L), width (W), and height (H).
Dielectric mirror Fabry-Perot cavity
2
2nL
(900 109 m)2 = 5.473nm 2(3.7)(20 106 m) 1,2 m = 6nm = 1.096 5.473nm
Length, L Height, H
number of modes =
Width W
Diffraction limited laser beam
Number of modes = 1
Dielectric mirror
Dielectric mirror
Laser Diode Characteristics

Length, L
Fabry-Perot cavity

Length, L
Fabry-Perot cavity
The length L determines the longitudinal mode separation.
Height, H
Width W
The actual modes that exist in the output spectrum will depend on the optical gain these modes experience.
Height, H
Width W
The width W and height H determine the transverse modes, or lateral modes. If the transverse dimensions (W and H) are sufficiently small, only the lowest transverse mode TEM00 mode, will exist. This TEM00 mode however will have longitudinal modes whose separation depends on L. The emerging laser beam exhibits divergence. This is due to diffraction of the waves at the cavity ends. The smallest aperture (H in this figure) causes the greatest diffraction.
The optical gain depends on on the optical resonator structure and the pumping current level.
Relative optical power
P o = 5 mW
P o = 3 mW
Po = 1 mW (nm)
778 780 782
Output spectra of lasing emission from an index guided LD. At sufficiently high diode currents corresponding to high optical power, the operation becomes single mode. (Note: Relative power scale applies to each spectrum individually and not between spectra)

The laser diodes output characteristics also tend to be temperature sensitive. P (mW)
o 10 8 6 4 2 0 0 20 40 60 80 0 C 25 C 50 C

The output spectrum also changes with temperature. In the case of a single-mode laser diode, the peak emission wavelength 0 exhibits jumps at certain temperatures.
Single mode 788 786 Single mode Multimode
(a)
(b)
(c)
o
(nm)
784 782 780 Mode hopping
I (mA)
778 776 20 50 20 40 Case temperature ( C) 30 30 50 20 40 Case temperature ( C)
30
40
50
Output optical power vs. diode current as three different temperatures. The threshold current shifts to higher temperatures.
Case temperature ( C)
Peak wavelength vs. case temperature characteristics. (a) Mode hops in the output spectrum of a single mode LD. (b) Restricted mode hops and none over the temperature range of interest (20 - 40 C). (c) Output spectrum from a multimode LD.
As the temperature increases, the threshold current increases steeply, typically as the exponential of the absolute temperature.
A jump corresponds to a mode hop in the output. That is, at the new operating temperature, another mode fulfills the laser oscillation conditions and a discrete change in the wavelength.

Between mode hops, wavelength 0 increase slowly with the temperature due to the slight increase in the refractive index n and the cavity length with temperature.
Single mode 788 786 Single mode Multimode

A useful laser diode parameter is the slope efficiency which determines optical power P0 of the output coherent radiation in terms of the diode current above the threshold current Ith.
(a)
(b)
(c)
o
(nm)
784 782 780 778 776 20 30 50 20 40 Case temperature ( C) 30 50 20 40 Case temperature ( C) Mode hopping
slope =
P0 Watts where I is the diode current I I th Amp
30
40
50
Case temperature ( C)
Highly stabilized laser diodes are usually marketed with thermoelectric coolers integrated into the diode package to control temperature.
i i
( )
d h
Highly stabilized laser diodes are usually marketed with thermoelectric coolers integrated into the diode package to control temperature.
Example 4.9.1 Laser output wavelength variations

Consider an GaAs based laser diode. Given: dn/dt ~ 1.5 x 10-4 K-1 Emitted wavelength = 870nm Estimate the change in the emitted wavelength per degree change in the temperature between mode hops.
d d 2nL 2 L dn 2nL m = m m = m dT dT m m dT
Section 4.10 Steady State Rate Semiconductor Equations Consider a double heterostructure laser diode under forward bias. The current carries the electrons into the active layer where they recombine with holes radiatively.
Dielectric mirror Fabry-Perot cavity
Let the length = L, the width W, and the thickness d (height H in the figure)
Length, L Height, H Width W Diffraction limited laser beam
Putting the equation in terms of wavelength, use L/m = m/2n.

d m 2 L dn 2m dn m dn 870nm pm (1.5 104 K 1 ) = 35.27 = = = dT m dT 2n dT n dT 3.7 K
The rate of electron injection into the active layer by the current I is equal to heir rate of recombination by spontaneous and stimulated emission (under steady state conditions).
Steady State Rate Semiconductor Equations Rate of electron injection = Rate of spontaneous + Rate of stimulated.
n I = injected + Cninjected N ph edLW sp
Steady State Rate Semiconductor Equations Consider the coherent photon Nph in the cavity. Under steady state: Rate of coherent photon loss = Rate of stimulated emissions
N ph = Cninjected N ph
Where ninjected is the injected electron concentration. Nph is the coherent photon concentration in the active layer. sp is the average time for spontaneous recombination. C is a constant which depends on B21. The output light power P0 is proportional to Nph.
ph
Where ph is the average time for a photon to be lost from the cavity due to transmission through the end-faces, scattering and absorption in the semiconductor.
Steady State Rate Semiconductor Equations If t is the total attenuation coefficient representing all these loss mechanisms, the the power in a light wave, in the absence of amplification, decrease as exp(- tx). This is equivalent to a decay in time since the velocity * time = distance x
t
Steady State Rate Semiconductor Equations When the current exceeds Ith, the output optical power increases sharply with the current, so we will let Nth = 0 at I = Ith. This last assumption just sets an arbitrary level as zero for the convenience of counting.
I th = nth edLW
sp
and J th =
n ed I = th sp Area
e t x = e
ph
where ph =
n and n = refractive index c t
Threshold is reached when the stimulated emission just overcomes the spontaneous emission and the total loss mechanism in time ph. Thus
The electron concentration at threshold nth = 1 C ph
When the current exceeds the threshold current, the excess carriers above nth recombine by stimulated emission. The steady state electron concentration remains constant at nth though the rates of carrier injection and stimulated recombination have increased.
Steady State Rate Semiconductor Equations Above threshold, with the electron concentration clamped at nth,
ph I I th = Cnth N ph and N ph = ( J J th ) edLW ed
Section 4.11 Light Emitters for optical fiber communications The type of light source suitable for optical communications depends not only on the communication distance but also on the bandwidth requirement. For short haul applications (ex local area networks) LEDs are preferred as they are simpler to drive, more economic, have a longer lifetime and provide the necessary output power. However the output spectrum for LEDs is much wider than that of a laser diode. LEDs are typically used with multimode and graded index fibers because the dispersion arising from the finite linewidth of the output spectrum is not a major concern with these fibers (over the short haul links envisioned here).
To find the optical output power P0 consider the following. It takes t = nL/c seconds for photons to cross the laser cavity. Only Nph is moving towards one mirror at any instant (steady state!) The light that escapes is (1 R) of the radiation. Thus P0 is
P0 = ( 1 N ph )(cavity volume)(Photon energy) 2 t (1 R )
P0 =
hc 2 phW (1 R) ( dLW )( h ) 1 ph ( J J th ) (1 R) = ( J J th ) nL 2 ed 2en c
Laser Diode Equation
Light Emitters for optical fiber communications For long haul and/or wide bandwidth communications, laser diodes are invariably used because of their narrow linewidth and high output power. The type of light source suitable for optical communications depends not only on the communication distance but also on the bandwidth requirement.
Light power Laser diode
Light Emitters for optical fiber communications The laser diode is the clear winner when linewidth must be narrow. It is also the winner for another important parameter, rise time. The speed response of an emitter is generally described by a rise time. If the driving current is applied suddenly as a step input to the diode, the rise time is the time it takes for the light output to rise from 10% to 90% of the final value. Laser diodes have shorter rise times and are used whenever wide bandwidths are required.
10 mW
LED
5 mW
0 50 mA 100 mA
Current
Typical optical power output vs. forward current for a LED and a laser diode.
Note that the laser diode has a restricted current range where the light output is linear.
Section 4.12 Single Frequency Solid State Lasers Ideally the output spectrum from a laser device should be as narrow as possible. The generally means that we have to allow only a signle mode to exist. One method of ensuring only a single mode of radiation is in the laser cavity is use frequency selective dielectric mirrors a the cleaved surfaces of the semiconductor. The distributed Bragg reflector is a mirror that has been designed like a reflection type diffraction grating. It has a periodic corrugated Distributed Bragg structure. reflector
A B
Active layer Corrugated dielectric structure
Single Frequency Solid State Lasers Intuitively, partial reflections of waves from the corrugations interfere constructively to give a reflected wave only when the wavelength corresponds to twice the corrugation periodicity.
A B
q(B /2n) =
(b)
Two partially reflected waves such as A and B have an optical path difference of 2 where is the corrugation period. This 2 is called a Bragg wavelength B and is given by the condition for in-phase interference.
q(B /2n) =
(a)
(b)
(a) Distributed Bragg reflection (DBR) laser principle. (b) Partially reflected waves at the corrugations can only constitute a reflected wave when the wavelength satisfies the Bragg condition. Reflected waves A and B interfere constructive when q(B/2n) = .
B
n
= 2
Single Frequency Solid State Lasers

A B
q(B /2n) =
B
n
= 2
Distributed Feedback Laser In the distributed feedback (DFB) laser there is a corrugated layer, called the guiding layer, next to the active layer.
Corrugated grating Guiding layer Active layer
(b) The Bragg wavelength equation has n = the index of refraction of the corrugated material. An integer q = 1, 2, 3, known as the diffraction order. is the corrugation period (spacing).
The distributed Bragg reflector (DBR) has a high reflectance around B but low reflectance away from B. The result is that only a particular Fabry-Perot cavity mode, within the optical gain curve, that is close to B can lase and exist in the output.
The cavity radiation spreads from the active layer to the guiding layer. These corrugations in the refractive index act as optical feedback over the length of the cavity by producing partial reflections. Thus optical feedback is distributed over the cavity length. The optical feedback analysis requires attention to both the distributed nature of the corrugations, the phase changes at each of these corrugations, and the left and right traveling waves inside the cavity.
Distributed Feedback Laser The allowed DFB modes are not exactly at the Bragg wavelengths but are symmetrically placed about B.
m = B
2 B
Cleaved-Coupled Cavity (C3) Laser In the cleaved-coupled cavity device, two different laser optical cavities L and D are coupled.
Active layer
2nL
(m + 1)
b)
0.1 nm
(c)
(nm)
In practice, either inevitable asymmetry introduced by the fabrication process, or intentional asymmetry, lead to only one of the modes appearing.
The two lasers are pumped by different currents. Only those waves that can exist in both cavities are now allowed because the system has been coupled. Cavity Modes
In L
e
In D In both L and D
Cleaved-Coupled Cavity (C3) Laser These two different set of modes coincide only at far spaced intervals.
Cavity Modes In L
e
Example 4.12.1 DFB Laser

Consider a DFB laser that has Corrugation period = 0.22 m Grating length L = 400 m Bragg effective index of refraction n = 3.5 Assume a first order grating, this is q = 1 Calculate the Bragg wavelength.
In D In both L and D
This restriction is possible modes in the combined cavity and the wide separation between modes results in a single mode operation.
Solution:
B
n
= 2 B =
2n 2(0.22 m)(3.5) = = 1.54 m q 1

Consider a DFB laser that has Corrugation period = 0.22 m Grating length L = 400 m Bragg effective index of refraction n = 3.5 Assume a first order grating, this is q = 1 Calculate the mode wavelengths. Solution:
m = B
2 B

Consider a DFB laser that has Corrugation period = 0.22 m Grating length L = 400 m Bragg effective index of refraction n = 3.5 Assume a first order grating, this is q = 1 Calculate the mode separation.
2
2nL
(m +1) = 1.54m
(1.54m)
Solution: The separation of the modes is the difference of the two.

(m +1) = 1.54m 0.847nm(m + 1)
2(3.5)(400m)
m = 1.54085m 1.5391m = 1.75nm
For m = 0, we have
m = 1.54m 0.847nm(0 +1) = 1.5391m and 1.54085m
In practice, only one of these modes will appear as the output.
Section 4.13 Quantum Well Devices A typical quantum well device has an ultra thin , typically less than 50 nm wide, narrow bandgap semiconductor such as GaAs, sandwiched between two wider bandgap semiconductors such as AlGaAs.
Quantum Well Devices At the interface between the semiconductors (at d in the figure), Ec and Ev are discontinuous. This forms a potential barrier and conduction band electrons in the thin GaAs layer are confined in the x direction of the figure.
E
d
c
D E3 E2 E1
The two semiconductors should be lattice matched in the sense that they have the same lattice parameter a. Lattice defects due to mismatch of crystal dimensions are thus minimal.
Dy
E E
n =2 n =1
d
D
z
d
D z AlGaAs AlGaAs z y x GaAs
E
v
g 2
E E
g 1
AlGaAs
AlGaAs
z y
x GaAs
This confinement is so small that we can treat the electron as in a onedimensional potential energy well in the x-direction which is free in the yz plane.
Quantum Well Devices

E
d
c
D E3 E2 E1
Quantum Well Devices

E
d
c
D E3 E2 E1
E E
n =2 n =1
E E
d
D
z
n =2 n =1
d
D
z
g2
E E
v
g1
AlGaAs
AlGaAs
z y
g2
E E
v
g1
AlGaAs
AlGaAs
z y
x GaAs
x GaAs
The energy of an electron in this three dimensional potential well of size d, Dy, and Dz is given by
E = EC +
2 h2 ny h2 n2 h2 n2 + * 2 + * z2 * 2 8me d 8me Dy 8me Dz
The holes in the valence band are also confined by the quantum well. Thus this system can be considered as a gas of electrons in a two dimensional space (constrained in the third dimension x in this case). Thus we have a system with a large number of states in a very small region. Under forward bias, electrons are injected in the thin GaAs region which serves as the active layer. These injected electrons readily populated the large number of available states (likewise there are a large number of available states in the valence band for the holes).
Where n, ny, and nz are quantum number with values 1, 2, 3, The potential barrier height is defined with respect to the arbitrary energy level EC. Since the dimensions Dy, and Dz are some much greater than d, the minimum energy is almost entirely found from term with n and d.
Quantum Well Devices Under current injection, the electron concentration at E1 increases rapidly and hence population inversion occurs quickly without the need for a large current to bring in a great number of electrons. Stimulated transitions of electrons lead to a lasing emission.
Ec E E1 h = E1 E1 E1 Ev
Quantum Well Devices There are two distinct advantages of this structure.
Ec E E1 h = E1 E1 E1 Ev
The threshold current for population inversion (and hence lasing emission) is markedly reduced in comparison to that for bulk semiconductors. Secondly, since the majority of the electrons are at or very near to E1, and the holes are at or near E1, the range of emitted photon energies is very close to E1 E1. Consequently the spread in the wavelength, the linewidth, in the output spectrum is substantially narrower than that in bulk semiconductor lasers.
Multiple Quantum Well Devices The advantages of the single quantum well structure can be extended to a larger volume of the crystal by using multiple quantum wells. In MQW lasers, the structure has alternating ultrathin layers of wide and narrow bandgap semiconductors.
Active layer E Ec Barrier layer
Multiple Quantum Well Devices The smaller bandgap layers are the active layers where electron confinement and lasing trasition take place. The wider bandgap layers are the barrier layers.
Active layer E Ec
Barrier layer
Ev
Though the optical gain curve is narrow, it is not necessarily single mode. The number of modes depends on the individual widths of the quantum wells.
Ev
A multiple quantum well (MQW) structure. Electrons are injected by the forward current into active layers which are quantum wells.
Example 4.13.1 A GaAs quantum well

Consider a GaAs quantum well Effective mass of a conduction electron = 0.07me Calculate the first two electron energy levels wrt EC for quantum well thickness d = 10 nm. Solution:
h2 n2 (6.62611034 )2 En EC = * 2 = n 2 = 8.6062 1021 n 2 Joules 8me d 8(0.07)(9.111031 )(10 109 )2

Consider a GaAs quantum well Effective mass of a conduction electron = 0.07me Effective mass of a valance band hole = 0.50me What is the hole energy wrt EV? Solution:
' EV En =
h2 n2 (6.62611034 )2 = n 2 = 1.205 1021 n 2 Joules * 2 8mh d 8(0.50)(9.111031 )(10 109 )2
8.6062 1021 2 n eV = (0.0537eV )n 2 1.602 1019
8.6062 1021 2 n eV = (7.521 103 eV )n 2 1.602 1019
So E1 = 0.0537 eV and E2 = 0.2149 eV.
So E1 = 0.007521 eV

Consider a GaAs quantum well Effective mass of a conduction electron = 0.07me Effective mass of a valance band hole = 0.50me What is the wavelength of emission of the SQW versus bulk GaAs? Solution: For bulk GaAs with Eg = 1.42 eV
Section 4.14 Vertical Cavity Surface Emitting Lasers (VCSELs)
A vertical cavity surface emitting laser has the optical cavity axis along the direction of current flow rather than perpendicular to current flow as in conventional laser diodes.
Contact
Eg = h =
hc
hc (6.6261 1034 )(3 108 ) = = 873.83nm Eg 1.42(1.602 1019 )
The active region length is very short compared with the lateral dimensions so that the radiation emerges from the surface of the cavity rather than its edge. The reflectors at the ends of the cavity are dielectric mirrors made from alternating high and low refractive index quarterwave thick multilayers.
/4n2 /4n1
Dielectric mirror
Active layer Dielectric mirror
For the SQW also with Eg = 1.42 eV hc (6.6261 1034 )(3 108 ) QW = = = 837.7 nm Eg + E1 + E1' (1.42 + 0.0537 + 0.0075)(1.602 1019 )
Substrate Contact
Vertical Cavity Surface Emitting Lasers (VCSELs)
Vertical Cavity Surface Emitting Lasers (VCSELs)
Since the wave is reflected because of periodic variation in the refractive index as in a grating, the dielectric mirror is essentially a distributed Bragg reflector.
Contact
One of the principle advantages of this structure is that they can be arrayed to construct a matrix emitter useful in optical interconnects.
High reflectance end mirrors are needed because the short cavity length L reduces the optical gain. There may be 20 30 layers in the dielectric mirrors to obtain the required reflectance (~99%).
/4n2 /4n1
Dielectric mirror
Active layer Dielectric mirror
Substrate Contact
Section 4.15 Optical Laser Amplifiers A semiconductor laser structure can also be used as an optical amplifier that amplifies light waves passing through its active region.
Pump current
Optical Laser Amplifiers For the traveling wave amplifier, While the cavity is pumped, the mirrors have antireflection coatings so the optical cavity does not act as an efficient resonator. Pump current
Signal in Active region Signal out
Signal in Active region
Signal out
AR = Antireflection coating
AR
Partial mirror
Partial mirror
AR = Antireflection coating
AR
(a) Traveling wave amplifier
(a) Fabry-Perot amplifier
(a) Traveling wave amplifier
The wavelength of radiation to be amplified must fall within the optical gain bandwidth of the laser.
Light input is amplified by the stimulated emissions and leaves the optical cavity at higher intensity. Typically such amplifiers are buried heterostructure devices and have optical gains of around 20 dB.
Optical Laser Amplifiers The Fabry-Perot laser amplifier is operated below the threshold current for lasing oscillations.
Section 4.16 Holography
This section will not be covered. It is an interesting if very brief introduction into holography.
Partial mirror
Partial mirror
(a) Fabry-Perot amplifier
The active region has an optical gain but is not sufficient to sustain a self-lasing output. The wavelengths closest to optical gain bandwidth receive the most increase in intensity but other wavelengths are amplified as well.
End of Chapter 4

Lasers

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lasers

Uploaded by

Copyright:

Available Formats

Section 4.

Stimulated Emission Devices LASERs

This last transition is called stimulated emission.

Stimulated Emission Rate Thus the upward transition rate is,

R21 = spontaneous rate + stimulated rate = A21 N 2 + B21 N 2 (h )

Plancks Black Body Radiation

B12 N1 (h ) = A21 N 2 + B21 N 2 ( h )

R21 ( stim) c3 = (h ) R21 ( spon) 8 h 3

R21 ( stim) N = 2 R12 (absorp ) N1

R21 ( stim) B N (h ) N 2 = 21 2 = R12 ( absorp ) B12 N1 ( h ) N1

Section 4.3 Optical Fiber Amplifiers

Optical Fiber Amplifiers

Energy of the 3+ ion Er in the glass fiber

Section 4.4 Gas Lasers: The He-Ne LASER

Pump laser diode = 980 nm

The He-Ne LASER

The He-Ne LASER

Collisions with the walls

Example 4.4.2 Laser Bean divergence

Section 4.5 The Output Spectrum of a Gas Laser

r 1 r = L tan = (10m) tan( 103 rad ) = 5mm 2 L

The spot size is then 2w0 + 2 r = 1mm + 2(5mm) = 11mm.

The Output Spectrum of a Gas Laser

The Output Spectrum of a Gas Laser

Where M is the mass of the lasing atom or molecule.

The Output Spectrum of a Gas Laser

The Output Spectrum of a Gas Laser

Stationary EM oscillations Mirror

The Output Spectrum of a Gas Laser

The Output Spectrum of a Gas Laser

The Output Spectrum of a Gas Laser

The observed FWHM width of the frequencies 1,2 is given by

632.8nm = 2.02 pm 4.74 1014 Hz

2(40cm) m0 = = = 1, 264, 222.5 0 632.8 1012 cm

Linewidth of spectrum 2.02pm 1,2 = = 4.03 Separation of two modes m 0.501pm

1999 S.O. Kasap, Optoelectronics (Prentice Hall)

= N 2 B21 (h ) N1 B12 (h ) = ( N 2 N1 ) B21 (h )

Where B12 = B21 under thermal equilibrium.

We can now use this result for dNph/dt.

Threshold Gain gth

Pi Ei Steady state EM oscillations 1

Reflecting surface Cavity axis R1 x

Threshold Gain gth

Pi Ei Steady state EM oscillations 1

Reflecting surface Cavity axis R1 x

Threshold Gain gth

Threshold Gain gth

Threshold Gain gth

Threshold Gain gth

c from equation (4) on page 176 B21nh 0

Po = Lasing output power Threshold population inversion

Threshold Gain gth

Example 4.6.1 Threshold population

Example 4.6.1 Threshold population

Example 4.6.1 Threshold population

c2 8 (1) (4.74 10 ) (300 10 )(1.5 109 Hz ) Nth = (0.155m 1 ) (3 108 ) 2

1 1 1 1 1 1 ln ln = 0.05m + = 0.05 + 0.01053 = 0.1554m 2L R1R2 2(50102 m) 1 0.9

Phase Condition and Laser Modes

Phase Condition and Laser Modes

round trip = m(2 )

Now with a little more algebra and recalling = 2/k

round trip = m(2 )

) = L where m is the approximate number of cavity modes.