This version reflects the comments of the core participants as reviewed and incorporated in accordance with CORD's FIPSE-supported Curriculum Morphing Project.

MODULE 1-6
LASING ACTION

(1) When light travels through a material, part of the light energy is absorbed by the atoms in the material. The amount of light absorbed is dependent upon the characteristics of the material and its thickness. Optical components, such as lenses and windows, are made of materials that absorb very little of the light energy in the wavelength region within which they are designed to function. Optical filters are designed to transmit only a particular portion of the light that strikes them. They may attenuate some wavelengths or eliminate some selected wavelengths, and transmit the remaining ones with no change. The absorption of light is a critical process in the optical pumping of solid and liquid lasers.

(2) When a laser beam passes through the active medium of a laser, energy is added to the laser beam through a process called "optical gain."

(3) This module discusses the absorption of light by materials and the gain of a laser medium. The similarities of these two processes are examined, and both are measured experimentally in the laboratory.

Upon completion of this module, the student should be able to:

1. Explain the term absorption coefficient and state its units.
2. Given any two of the following quantities, calculate the third:

   a. Irradiance incident upon an optical material.
   b. Irradiance transmitted through the material.
   c. Transmission of the material.

3. Given any two of the following quantities, calculate the third:

   a. Absorption coefficient of a material.
   b. Thickness of the material.
   c. Transmission of the material.

4. Explain with a diagram and in a short paragraph the exponential law of absorption.
5. Describe in two or three sentences each how the transmission of the following filters varies with changes in the wavelength of light:

   a. Neutral density.
   b. Cutoff.
   c. Band-pass.

6. Given the optical density of a filter, calculate its transmission.
7. Explain the following terms with the use of diagrams:

   a. Normal population distribution.
   b. Population inversion.

8. Given two of the following quantities, calculate the third:

   a. Gain coefficient of a laser.
   b. Length of active medium.
   c. Amplifier gain.

9. Draw a diagram that displays gain as a function of wavelength for a typical laser emission line.
10. Draw and label the energy-level diagram of a four-level laser.
11. Explain the energy-transfer processes that increase the population of the upper losing level in gas lasers and in solid lasers.
12. Explain the energy-transfer processes that depopulate the lower losing level in gas lasers and solid-state lasers.
13. Given the appropriate equipment, measure the transmission of three colored filters at the HeNe laser wavelength; and calculate the absorption coefficient of each.
14. Given the appropriate equipment, measure the amplifier gain of a HeNe laser tube; and calculate the gain coefficient.

ATTENUATION OF LIGHT

(4) Figure 1 depicts a beam of light traveling through a piece of optical material. Some of the light energy is absorbed by the material, and some is transmitted.

Fig. 1 Attenuation of a light beam

(5) The transmission of the optical material is given by Equation 1.

Equation 1

(6) Example A illustrates the use of this equation in solving a problem.

EXAMPLE A: CALCULATION OF THE TRANSMISSION
Given:	The light incident upon the material in Figure 1 has an irradiance of 2.5 mW/cm2. The irradiance of the transmitted light is 0.50 mW/cm2.
Find:	The transmission.
Solution:	T = 0.20 20% of the light is transmitted.

(7) In some cases, almost no light is absorbed, and the transmission is almost 1.0. In others, there is no transmission at all (T = 0). Reflection and scattering of light together with absorption account for losses in all optical systems, but reflection and scattering are not considered in this module.

-----

THE EXPONENTIAL LAW OF ABSORPTION

(8) Obviously, any increase in the thickness of the absorbing material will decrease the irradiance of the transmitted light. Figure 2 depicts light traveling through four identical pieces of filter material, each 1 mm thick and each filter absorbing one-half the light incident upon it. Figure 3 is a plot of the transmission of this filter material as a function of total filter thickness. Transmission in this case is based upon incident and transmitted power, rather than upon irradiance. The curve in Figure 3 is called an "exponential curve." It begins at an initial value of 1.0 and approaches zero asymptotically as thickness increases. The degree of transmission for any thickness of a material is given by the exponential law of absorption, as stated in Equation 2.

Fig. 2 Transmission of light through a series of filters

Fig. 3 Transmission as a function of thickness

Equation 2

T = e^–kx

(9) The unit cm^–1 in this equation should not be confused with the reciprocal centimeter used to indicate wave number in the previous module. In this case, the absorption coefficient is measured in terms of absorption per centimeter. The absorption coefficient is numerically equal to the reciprocal of the thickness of a specific material that results in a transmission of 1/e (0.368) of the incident light. The units of thickness and absorption coefficient must be reciprocals of one another in order that their product, the exponent of e, will remain a dimensionless quantity.

(10) Example B demonstrates the use of Equation 2 in solving problems.You should be able to evaluate terms like e^–kx easily with the help of a scientific calculator, using the e^x key.

(11) Equation 2 may be solved for the absorption coefficient through the following steps:

Equation 2

		(by taking the reciprocal of both sides of the equation)
		(by taking the natural logarithm of both sides of the equation)

Rearrangement of terms yields Equation 3 for the absorption coefficient.

Equation 3

(12) This equation is utilized to solve a problem in Example C.

(12a) Equations like T=e^–kx (Eq 2) and k= l n (Eq 3) involve exponential functions and logarithmic functions. If you are not comfortable with handling these, the following quick review may help you.

When a function is expressed as the base e raised to a certain power, either negative or positive, it can be evaluated quickly with your scientific calculator. For example, evaluate e^–5.7.

1. enter 5.7 on the keyboard so that 5.7 shows in the window
2. press the "change sign" key, often labeled +/–, and make sure –5.7 shows in the window.
3. with –5.7 in the window, simply press the e^x key on your calculator
4. the value of e^–5.7 shows up in the window to be 3.3459...´10^–3, or about 0.0033.

When a function is expressed as a natural logarithm, l n x, for example, the scientific calculator again makes the evaluation easily. Note in the derivation from Eq 2 to Eq 3, taking the l n of ekx simply results in the exponent to which e had been raised. Thus:

l n(e²) = 2
l n(e^–2.1) = –2.1
l n(e^kx) = kx

Now evaluate a term involving the "l n" function: l n 5.7

1. enter –5.7 on the keyboard and observe 5.7 in the window
2. since the number is positive, there is no need to change the sign
3. press the "l n" key (not the "log" key) and read the result
4. l n 5.7 = 1.74

EXAMPLE C: CALCULATION OF ABSORPTION COEFFICIENT
Given:	A 10-mW laser beam strikes a piece of filter material 1.2 cm thick. A beam of 0.42 mW is transmitted.
Find:	Absorption coefficient of the material.
Solution:

-----

ABSORPTION AS A FUNCTION OF WAVELENGTH

(12b) Optical filters come in many varieties. Generally, they are used to (1) select one wavelength region over another, (2) select a very narrow wavelength region and exclude all other wavelengths above and below this region, (3) select a broad range of wavelenths which, as a group, are all transmitted with the same intensity.

Optical filters are generally made of colored glass, metals, and thin dielectric films, often arranged in complicated "sandwich" geometries to achieve a special type and degree of filtering of the light incident on them. You will find such names as colored glass filters, thin metallic film filters, hot and cold mirrors, neutral density filters, bandpass filters, narrow band interference filters, and broadband interference filters, if you peruse the section on "Filters," covered, for example, in catalogs of optical fabrication companies such as Melles Griot. You should "walk through" such a section to develop an appreciation for the many types of filters available. Here we shall concentrate on the important types–cutoff filters, narrow bandpass filters, and neutral density filters.

(13)The absorption coefficient (and thus the transmission) of any material is a function of the wavelength of the light striking that material. Until now, discussion in this module has dealt with thickness variations of absorbing materials at a single wavelength. This section assumes a fixed absorber thickness and presents transmission variations as a function of wavelength changes. Three types of optical filters are used as examples.

(14) Figure 4 gives the transmission of two optical filters as a function of wavelength. The blue filter in Figure 4a allows most of the blue light to pass, but absorbs other colors of the visible spectrum, while the red filter (Figure 4b) allows only red light to pass through. This type of filter has a sharp division between high- and low-transmission regions and thus is called a "cutoff filter." Optical components of this type are used to eliminate unwanted wavelengths in many optical systems. Laser safety goggles are an important example of cutoff filters.

Fig. 4 Transmission-versus-wavelength for cutoff filters

(15) Figure 5 depicts the transmission curve of a band­pass filter. This type of filter passes a narrow band of wavelengths but blocks all light outside this band. Darkroom safelights contain band­pass filters that pass only those light wavelengths to which the film is not sensitive.

Fig. 5 Transmission-versus-wavelength for a "spike" or band-pass filter

(16) Transmission of a neutral density filter is illustrated in Figure 6. This filter is designed to have the same transmission for all wavelengths over a broad range of the spectrum. The gray lenses of most sunglasses are neutral density filters for visible light.

Fig. 6 Transmission-versus-wavelength for a neutral density filter

(17) A term often used to describe the transmission of neutral density filters, laser safety goggles, and other low-transmission filters is optical density. If the optical density of a filter is known, its transmission can be calculated from Equation 4.

Equation 4

T = 10^–OD

Thus, a filter with an optical density of 2.0 has a transmission of 10^–2, or one percent. (Table 1 lists some OD values.) Example D illustrates the use of this equation in solving another problem.

TABLE 1. Optical Density Values.

GAIN IN A LASER

POPULATION INVERSION

18) Figure 7 illustrates the normal population distribution of atoms in their various energy levels in a given material at some given temperature.

Fig. 7 Normal population distribution

(19) The lengths of the horizontal lines in this figure represent the number of atoms in each energy state. The majority of the atoms are always in the ground state, and the population of each successively-higher energy state is less than that of any other lower energy state.

(20) The absorption coefficient of this material at a wavelength of light corresponding to the transition from E₃ to E₂ is proportional to the quantity N₂ – N₃. A greater difference in population of the two energy states results in a Greater absorption coefficient. If atoms were to be removed from E₂ and added to E₃ to reduce the population difference, the absorption coefficient would be reduced.

(21) If the populations of the two states are the same, the absorption coefficient is, in effect, zero. Although absorption still occurs, stimulated emission occurs at a similar rate, replacing photons as they are removed by absorption. The two processes balance each other, and the net difference is zero, hence an effective zero absorption coefficient.

(22) In Figure 8, the populations of energy states E₂ and E₃ have been altered until the quantity N₂ – N₃ is negative.

Fig. 8 Population inversion between states E₂ and E₃

(23) The population inversion between E₂ and E₃ now requires that the value of the absorption coefficient also must become negative. This condition is called a "population inversion" because the populations of the two atomic states have been inverted with respect to each other. When a population inversion exists, the absorption coefficient in the equation T = e^–kx is negative; and the transmission is then necessarily greater than one. Under these conditions, stimulated emission occurs at a rate greater than the rate of absorption, and the light passing through the material is amplified.

(24) The gain of such an optical amplifier is dependent upon the quantity N₃ – N₂; therefore, greater population inversion produces greater gain. To increase the gain of a laser, some method must be utilized to increase of the upper energy state of the lasing transition. Since the lasing process itself will transfer atoms to the terminal state of the losing transition, some process must be available to reduce the population of this lower energy state. If atoms are allowed to remain in this terminal state, the population of the state will increase to the point at which a population inversion no longer exists, and the losing activity of the medium will cease.

-----

AMPLIFIER GAIN OF A LASER

(25) The amplifier gain of a laser medium is given by Equation 5.

Equation 5

G_a = e ^ax

The gain coefficient usually is expressed in cm^-l, with the active medium length in cm; but sometimes the gain is in m^-1, and the active medium length is in m. In solving problems, one must ensure that the unit of gain coefficient is the reciprocal of the unit of length.

(26) Example E illustrates the use of Equation 5 in solving a problem.

(27) Equation 5 is of the same form as Equation 2 (the exponential law of absorption) with a replacing –k and G_a replacing T. If these quantities are substituted into Equation 3 for the absorption coefficient, the result is Equation 6 for the gain coefficient.

Equation 6

(28) Example F illustrates a problem solved by use of Equation 6.

EXAMPLE F: CALCULATING GAIN COEFFICIENT
Given:	A HeNe laser tube 50 cm long has a gain of 1.08.
Find:	The gain coefficient.
Solution:

-----

LASER GAIN AS A FUNCTION OF WAVELENGTH

(29) Module 1­5, "Absorption and Emission of Light," discussed briefly the width of spectral lines and the factors leading to broadening of a line. The same factors apply to the wavelength range over which the active medium provides gain in a laser. Figure 9 exhibits the general shape of the gain­versus­wavelength curve for most lasers. In all cases, the active medium provides gain, not at a single wavelength, but over a narrow range of wavelengths. The effect of the gain profile upon laser output will be discussed further in Module 1­7, "Optical Cavities and Modes of Oscillation."

Fig. 9 Gain as a function of wavelength

ENERGY TRANSFER IN LASERS

ENERGY TRANSFER IN GAS LASERS

(30) The excitation mechanism in most gas lasers is a direct-current discharge through the gas (the active medium). Gas atoms gain energy through collisions with energetic free electrons produced in the gas ionization process. In ion lasers, the atoms are pumped upward in energy through several intermediate energy states by multiple collisions. In lasers, such as the HeNe and CO₂ types, that contain gas mixtures, the electron collisions excite one type of atom or molecule to an upper energy state. This excited atom or molecule then collides with an atom or molecule of the losing gas, and energy is transferred through this collision.

(31) The depopulation of the lower losing level E₂ in gas lasers may occur through spontaneous emission of photons or through additional collisions, either with atoms or with the walls of the container.

-----

ENERGY TRANSFER IN SOLID LASERS

(32) Most solid lasers (and liquid lasers) employ optical pumping. Light from a flashlamp is absorbed by atoms in the active medium, raising them to "level" E₄ (Figure 10). In solid and liquid lasers the level E₄ is not a single level; rather it consists of a series of energy levels, or pumping bands as depicted in Figure 10. The fast decay transitions from E₄ to E₃ often are radiationless transitions that contribute to heating the laser active medium.

Fig. 10 Energy-level diagram for Nd:YAG

(33) In order for the optical energy from the flashlamp to be absorbed efficiently, its wavelength must correspond to a transition from the ground state to an available pumping band of the active medium. For Nd:YAG crystal, the strong pumping band has a wavelength range of 790­820 nm, and the weak pumping band has a range of 730­760 nm. The lamps used with these lasers, therefore, must produce their maximum output at these wavelengths for efficient laser operation. This matching process is called "spectral matching" of the pump source and active medium.

(34) The transition from E₂ to E₁ usually is a radiationless transition. Both the population of energy level E₂ and the rate of decay from that level depend upon the temperature of the active medium. Water cooling normally is employed in solid lasers to remove waste heat and to depopulate the lower lasing level by reduction of the ambient heat level.

-----

THE THREE­LEVEL LASER

(35) In a three­level laser (Figure 11), the lower losing level is the ground state. This type of laser can have a population inversion only if more than half the atoms in the ground state are pumped to higher energy states. In the four­level laser, such intense pumping is unnecessary because a high ground­state population does not affect the population inversion between the upper and lower laser levels. The ruby laser is the only important three-level laser.

Fig. 11 Three-level laser

-----

THE FOUR­LEVEL LASER

(36) Figure 12 is the energy­level diagram of a four­level laser model. The lasing transition occurs between E₃ and E₂. Energy level E₃ is called the upper lasing level and is a metastable state, with a lifetime of 10^–6 seconds or longer.

Fig. 12 Four-level laser

(37) Atoms remain in this metastable state for a relatively long time, increasing both the population inversion and the probability of stimulated emission. Level E₂ has a short atomic lifetime, which means that atoms leave this energy state quickly and return to the ground state.

(38) An upward transition to the metastable state is just as unlikely as a downward transition from it; therefore, the pumping of atoms directly from the ground state to the upper lasing level would be impractical. An alternative energy level, E₄, is a short-lived atomic state above E₃, to which atoms may be pumped more easily. Since short energy transitions are more likely to occur than long energy transitions, most of the atoms pumped to atomic state E₄ can be counted on to decay rapidly to atomic state E₃, the metastable state.

1. A 10­mW HeNe laser beam strikes an optical filter 1 mm thick. The transmitted power is 250 mW. Calculate the following values:

   a. Transmission of the filter.
   b. Absorption coefficient.
   c. Transmission of a piece of the same material 2.5 mm thick for HeNe laser light.

General Comment from Glenn Oliver

2. A filter material has an absorption coefficient of 5.0 cm^-l. What thickness of this material is required to reduce an incident irradiance of 15 mW/cm² to a transmitted irradiance of 1 mW/cm²?
3. A filter 0.015 mm thick exhibits a transmission of 87%. Calculate its absorption coefficient.
4. One piece of filter material placed in a HeNe laser beam reduces the power from 3.0 mW to 2.25 mW. What is the transmitted power if three more identical filters are used?
5. Explain with a diagram and in a short paragraph the exponential law of absorption.
6. Draw and explain transmission­versus­wavelength curves for the following filter types:

   a. Neutral density.
   b. Band­pass.
   c. Cutoff.

7. The optical density of three filters is given below. Calculate the transmission of each.

   a. 0.6
   b. 2.5
   c. 4.0

8. Explain the following terms with the use of diagrams:

   a. Normal population distribution.
   b. Population inversion.

9. A HeNe laser tube has a gain coefficient of 0.0016 cm^­l and an active length of 46 cm. A 3.6­mW HeNe beam enters the tube. Determine the following values:

   a. Gain of the tube.
   b. Power of the beam leaving the tube.

10. A Nd:YAG laser rod 4 inches in length exhibits a gain of 1.5. Calculate the gain coefficient. (State units carefully.)
11. A laser active medium has a gain coefficient of 0.15 m^–1. What active length is necessary for a gain of 2.0?
12. Draw and label a diagram of gain as a function of wavelength for a typical laser emission line.
13. Draw and label the energy­level diagrams of the following lasers:

    a. Four­level.
    b. Three­level.

14. Why should the upper losing level be a metastable level?
15. Explain these energy­transfer mechanisms:

    a.Pumping in gas lasers.
    b.Pumping in solid lasers.
    c.Depopulation of the lower losing level in four­level gas lasers.
    d.Depopulation of the lower losing level in four­level solid lasers.

EXPONENTIAL LAW OF ABSORPTION

HeNe laser
Optical power meter
Set of gelatin or plastic color filters (6 each: blue, green, red)
1­meter optical bench
2 pin carriers
Micrometer
Filter holders
Beam expander
Lab jack

General Comment from Neil Miller

GAIN IN A LASER

HeNe laser
Optical power meter
Alignment table with 3­point adjustment
Meter stick
1­meter optical bench
Brewster window HeNe laser tube
Power supply for tube
Pin carrier
Tube holder

EXPONENTIAL LAW OF ABSORPTION

In this experiment, the student will measure the transmission of three filter materials as a function of filter thickness and calculate the absorption coefficient of each material.

1. Mount the laser on a lab jack near one end of the optical bench. Attach the beam expander to the output aperture of the laser. (See Figure 13.)
2. Using a pin carrier, mount the sensor head of the power meter about 15 cm to the right of the laser. Between the laser and the light detector, use another pin carrier to attach the filter holder to the bench, as illustrated in Figure 13.

   

   Fig. 13 Experimental setup for absorption measurements

3. With the micrometer, measure the thickness of several filters. Record the average thickness in Data Table 1.
4. Using proper safety precautions, plug in and turn on the laser. Adjust the range selector on the power meter to 10 mW full scale and turn on the power meter. Position the sensor head in such a manner that the active area of the detector is fully illuminated; however, do not overfill the detector.
5. Measure and record power output (P_O) of the laser.
6. Place blue filter sheet in the holder, and measure power transmitted (P_t) by the filter. Repeat with two, three, four, five and six sheets of blue filter. Record the results in Data Table 1.
7. Repeat Step 6 for green and red filters. Note: The range switch on the power meter will have to be adjusted to obtain some of the readings,since the power levels to be measured will vary over a range from about 2­3 mW to < 100 mW.
8. From the data obtained in Steps 7 and 8, plot a curve of transmitted power vs. thickness (no. of filters) for each of the three types of color filter. The resultant curves should provide verification of the exponential law of absorption. (Your first data point on each curve should be the output power of the laser, P_O; that is, the power when no filters are in the beam path.)
9. Use Equation 3 and the data in Data Table 1 to calculate the absorption coefficient K for each color filter. For these calculations, use the total thickness of the six filters and the corresponding power transmitted when all six filters are in the beam path. Record the calculated values in Data Table 2. The values of k obtained are dependent upon wavelength of light used. Different values of k would be obtained with l different form 632.8 nm.
10. On the basis of the data, which filter (a) absorbed the greatest percentage? and (b) transmitted the greater percentage of incident light of wavelength l = 632.8 nm? Explain how the graphs verify the exponential law of absorption.

GAIN IN A LASER

In this experiment, the student will measure the small signal gain of a HeNe laser tube and will calculate the gain coefficient.

1. Mount the HeNe laser plasma tube with Brewster windows on an optical bench. (See Figure 14.) Attach electrical leads from the cathode and anode to the appropriate terminals on the power supply (unplugged at this point). Usually, the cathode is made of a larger surface area of metal; the anode is small, such as a stiff wire or small rod. Check the polarity to ensure that the leads have been attached properly.

   

   Fig. 14 Experimental setup for gain measurement

   KEEP YOUR HANDS AWAY FROM THE HIGH­VOLTAGE TERMINALS WHEN THE POWER SUPPLY IS ACTIVATED. Turn the power supply switch to the ON position and verify operation of the laser tube. Then turn off the power.

2. Place the HeNe laser on a 3­point adjustable table. Plug in and turn on the laser. Direct its beam down the bore of the plasma tube. Use the adjustment screws on the table to align the laser beam as closely as possible with tube bore. The beam should pass through the tube without striking the side walls. A piece of lens tissue is helpful in observing the position of the beam near the Brewster windows on the plasma tube. An index card placed in the beam about a meter beyond the plasma tube also is useful in achieving proper alignment.
3. Position the detector of the power meter to intercept the beam at a distance of 1 m from the plasma tube. If the detector is too near the end of the tube, it will receive large amounts of spontaneously­emitted light when the tube is on and an erroneous reading will result.
4. Measure the power of the HeNe laser beam transmitting through the plasma tube with the power supply to the plasma tube OFF. Record in Data Table 3. If the power is fluctuating, take power reading at the maximum needle position.
5. Turn on the power supply of the plasma tube, and record power of amplified beam. Note: For a 2­3 mW laser, the expected increase in power output produced by the laser amplifier will be approximately 100­200 mW.
6. Measure and record the effective discharge length of the plasma tube.
7. Calculate the gain of the plasma tube, and use Equation 6 to calculate the gain coefficient. Complete Data Table 3.

   ---

   

   DATA TABLE 1. EXPONENTIAL LAW OF ABSORPTION

   Equipment	Manufacturer/Model No.
   Laser Optical Power Meter
   Laser Power Output, PO _______________ mW Average Filter Thickness _______________ mm
   Number of Filters	Transmitted Power, Pt
   	Blue	Green	Red
   1
   2
   3
   4
   5
   6

   
   DATA TABLE 2.

   Color of Filter	Absorption Coefficient k (n), cm–1, at n = 4.74 × 1014 Hz; l = 632.8 nm
   Blue
   Green
   Red

   
   DATA TABLE 3. GAIN IN A LASER

   Active length: x = __________ cm Power with Brewster window tube off: P1 = _______________ mW Power with Brewster window tube on: P2 = _______________ mW Gain: Ga = P2/P1 = _______________ Gain coefficient: a = _______________REFERENCE

   ---

   

   Anderson, John D. Gasdypamic Lasers: An Introduction. From the Series Quantum Electronics ­ Principles and Applications, edited by Yoh­han Pao. New York: Academic Press, 1976.

   Barnes, Frank S. ed. Laser Theory. New York: The Institute of Electrical and Electronics Engineers, Inc. (IEEE Press), 1972.

   Driscoll, Walter G. and Vaughn, William, eds. Handbook of Optics. New York: McGraw­Hill Book Company, 1978.

   Lengyel, Bela A. Lasers. 2nd ed. New York: Wiley­Interscience, 1971.

   O'Shea, Donald C.; Called, W. Russell; and Rhodes, William T. Introduction to Lasers and Their Applications. Reading, MA: Addison­Wesley Publishing Co., 1978.

   Pressley, Robert J., ed. CRC Handbook of Lasers, with Selected Data on Optical Technology. Cleveland, OH: The Chemical Rubber Company, 1971.

   Shortley, George and Williams, Dudley. Elements of Physics. 5th ad. Englewood Cliffs, NJ: Prentice­Hall, Inc., 1971.

   Siegman, A.E. Introduction to Lasers and Masers. New York: McGraw­Hill Book Company, 1971.

   Verdeyen, Joseph T. Laser Electronics. Englewood Cliffs, NJ: PrenticeHall, Inc., 1981.

   Yariv, Ammon. Introduction to Optical Electronics. New York: Holt, Rinehart and Winston, Inc., 1971.

   ---