Chapter 6: ‐Emitting Diodes

Photoluminescence and Basic transitions efficiency Light-emitting diodes Internal quantum efficiency External quantum efficiency Device structures LED materials Heterojunction high-intensity LEDs LED characteristics

PHYS5320 Chapter Six 1 and Electroluminescence absorption can create ‒hole pairs. Excess and holes can recombine and result in the emission of in direct bandgap materials. The general property of light emission is referred to as luminescence. When excess electrons and holes are created by photon absorption, then photon emission from the recombination process is called photoluminescence. Electroluminescence is the process of generating photon emission when excess carriers result from an caused by an applied electric field. We will mainly be concerned with injection of carriers across a pn junction. The pn junction light-emitting diode (LED) and diode (LD) are examples of this phenomenon. In these devices, electric energy, in the form of a current, is converted directly into photon energy. PHYS5320 Chapter Six 2 Basic Transitions Basic interband transitions: (i) Emission energy is very close to the bandgap energy of the material. (ii) Emission involves energetic electrons. (iii) Emission involves energetic holes. A combination of these emission processes will produce an with  a certain bandwidth. 

2 1/ 2  h  Eg  I h  Eg exp   kT  GaAs

PHYS5320 Chapter Six 3 Basic Transitions

Recombination processes involving impurity or defect states: (i) Conduction band to acceptor transition. (ii) Donor to valence band transition. (iii) Donor to acceptor transition. (iv) Recombination due to a deep trap. This process is usually non-radiative.

PHYS5320 Chapter Six 4 Basic Transitions

Non-radiative Auger process: (i) Recombination between an electron and hole, accompanied by energy transfer to a free hole. (ii) Recombination between an electron and hole, accompanied by energy transfer to a free electron. The involved third particle will eventually lose its energy to the lattice in the form of heat. The process involving two holes and one electron occurs predominantly in heavily doped p-type materials, and the process involving two electrons and one hole occurs primarily in heavily doped n-type materials.

PHYS5320 Chapter Six 5 Luminescence Efficiency Since not all recombination processes are radiative, an efficient luminescent material is one in which radiative transitions predominate. The quantum efficiency is defined as the ratio of the radiative recombination rate to the total recombination rate for all processes. Because the recombination rate is inversely proportional to the lifetime, the quantum efficiency can also be expressed in terms of lifetimes. Rr nr q    R nr r The interband recombination rate will be proportional to the number of electrons and holes available. Rr  Bnp The values of B for direct bandgap materials are on the order of 106 larger than for indirect bandgap materials. One problem encountered with the emission of photons from a direct bandgap material is the re-absorption of emitted photons. This re-absorption problem must be solved to achieve highly efficient devices. PHYS5320 Chapter Six 6 Light‐Emitting Diodes When a voltage is applied across a pn junction, electrons and holes are injected across the space charge region where they become excess minority carriers. These excess minority carriers diffuse into the neutral regions where they recombine with majority carriers. If this recombination process is a direct band-to-band process, photons are emitted. Such devices are then called light-emitting diodes (LEDs). The recombination rate is proportional to the diode diffusion current, so the output photon intensity will also be proportional to the ideal diode diffusion current.

The maximum emission wavelength from a direct bandgap semiconductor material is related to (but not exactly equal) the bandgap energy.

hc 1.24μm     Eg Eg eV

PHYS5320 Chapter Six 7 Internal Quantum Efficiency The internal quantum efficiency of a LED is the fraction of diode current that will produce luminescence. It is a function of the injection efficiency and a function of the percentage of radiative recombination events compared with the total number of recombination events. internal  q The three current components in a forward-biased diode are the minority carrier electron diffusion current, the minority carrier hole diffusion current, and the space charge recombination current.

eD n   eV   J  n p0 exp  1 n     eWn   eV   L   kT   i n JR  exp  1 20   2kT   eDp pn0   eV   Jp  exp  1 Lp   kT  

PHYS5320 Chapter Six 8 Internal Quantum Efficiency The recombination of electrons and holes within the space charge region is in general through traps near the middle gap and is a non-radiative process. In , electroluminescence originates primarily on the p-side of the junction because the efficiency for electron injection is higher than that for hole injection. The injection efficiency is then defined as the ratio of electron current to total current. Jn   Jn  Jp  JR The injection efficiency can be made to approach unity by using an n+p diode so that Jp is a small fraction of the diode current and by forward biasing the diode sufficiently so that JR is a small fraction of the total diode current. The radiative efficiency is the fraction of radiative recombination. Rr 1/r nr q      Rr  Rnr 1/ r 1/nr nr r The non-radiative rate is proportional to the density of trapping sites within the forbidden bandgap. Clearly, the radiative efficiency increases as the density of trapping sites is reduced. PHYS5320 Chapter Six 9 External Quantum Efficiency The external quantum efficiency quantifies the conversion efficiency of electrical energy into an emitted external optical energy. It incorporates the internal quantum efficiency and the subsequent efficiency of photon extraction from the device. P optical   out external IV There are three loss mechanisms the generated photons may encounter: photon absorption within the semiconductor, Fresnel loss, and critical angle loss.

Since some of the emitted photons

have energies above Eg, the photons generated deep inside the semiconductor will be re-absorbed by the semiconductor.

PHYS5320 Chapter Six 10 External Quantum Efficiency Fresnel loss: generated photons must transmit across the dielectric interface between the semiconductor and air. The reflectance at normal incidence depends on the refractive indexes of the two dielectric materials. 2  n2  n1  R     n2  n1 

Example: the refractive index for GaAs is n2 = 3.66 and for air is n1 = 1.0. Calculate the reflectance coefficient at the interface at normal incidence.

2  3.66 1.0  R     0.326  3.66 1.0 

PHYS5320 Chapter Six 11 External Quantum Efficiency Critical angle loss: the refractive indexes of semiconductor materials are generally larger than that of air. There will be a critical incidence angle above which generated photons will experience total internal reflection.  n    sin1 1  c    n2 

Example: the refractive index for GaAs is n2 = 3.66 and for air is n1 = 1.0. The critical angle for total internal reflection at the GaAs-air interface is  n    sin1 1  15.9 c    n2  PHYS5320 Chapter Six 12 External Quantum Efficiency

PHYS5320 Chapter Six 13 LED Structures

p-layer grown p-layer is formed by dopant epitaxially on n+-layer diffusion into the epitaxial layer

n+p junctions p side (~m) closer to the surface to reduce re-absorption Epitaxial growth to reduce lattice strain-induced defects

PHYS5320 Chapter Six 14 LED Structures

How do we reduce the A common procedure is the TIR loss? encapsulation of the device within a transparent plastic It is possible to shape the surface of the medium (an epoxy) that has a semiconductor into a dome. The main refractive index matching with drawback is the difficult fabrication the semiconductor and has a process and associated increase in cost. domed surface.

PHYS5320 Chapter Six 15 LED Materials Gallium arsenide is an important direct bandgap semiconductor material for optical devices.

Compound semiconductor AlxGa1-xAs is of great interest. For 0 < x < 0.35, the bandgap energy is expressed as

Eg 1.424 1.247xeV

PHYS5320 Chapter Six 16 LED Materials

Another compound semiconductor used for optical devices is the GaAs1-xPx system.

PHYS5320 Chapter Six 17 LED Materials

PHYS5320 Chapter Six 18 Heterojunction High Intensity LEDs A junction between two differently doped that are of the same material (same bandgap energy) is called a homojunction. A junction between two different bandgap semiconductors is called a heterojunction. A semiconductor device structure that has a junction between two different bandgap materials is called a heterostructured device. The p-region must be narrow to allow photons to escape without much re- absorption. However, when the p-side is too narrow, some of injected electrons in the p-side will reach the surface by diffusion and recombine non-radiatively through crystal defects near the surface. We have to make a balanced choice to achieve a maximal external efficiency.

PHYS5320 Chapter Six 19 Heterojunction High‐Intensity LEDs One approach for increasing the intensity of LED output light makes use of the double heterostructure. The potential barrier, Ec, prevents electrons in the conduction band in p-GaAs passing to the conduction band of p-AlGaAs.

PHYS5320 Chapter Six 20 Heterojunction High Intensity LEDs

The bandgap energy of AlGaAs is greater than that of GaAs. The emitted photons do not get re-absorbed. Since light is not absorbed by p-AlGaAs either, it can be reflected to increase the light output. There are negligible strain-induced defects due to a small lattice mismatch.

PHYS5320 Chapter Six 21 LED Characteristics The energy of an emitted photon from a LED is not simply equal to the bandgap energy because the electrons in the conduction band are distributed in energy and so are the holes in the valence band.

The electron concentration as a function of energy is asymmetrical and has a peak

at (1/2)kBT above Ec. The energy spread is ~2kBT. The rate of direct recombination is proportional to both the electron and hole concentrations at the energies involved.

PHYS5320 Chapter Six 22 LED Characteristics

The linewidth h is typically

between 2.5kBT to 3kBT.

  c /  hc / Eph  d hc   2 dEph Eph hc   Eph 2 Eph

asymmetrical PHYS5320 Chapter Six 23 LED Characteristics

The output spectra of actual LED devices are less asymmetrical than the idealized spectrum. There is a turn-on or cut-in voltage from which the current increases sharply with voltage. The turn-on voltage generally increases with the bandgap energy (blue LEDs: 3.5–4.5 V; yellow LEDs: ~ 2 V; infrared LEDs: ~ 1 V).

PHYS5320 Chapter Six 24 LEDs for Optical Fiber Communications

The type of light sources suitable for optical communications depends on both the communication distance and the bandwidth requirement. For short haul applications (local networks), LEDs are preferred as they are simpler to drive, more economic, have a longer lifetime, and provide the necessary power even though their output spectrum is much wider than that of a laser diode.

For long haul and wide bandwidth communications, laser diodes are invariably used because of their narrow linewidth, high output power, and higher signal bandwidth capability.

PHYS5320 Chapter Six 25 LEDs for Optical Fiber Communications

Photons emerge from an area on a crystal face perpendicular to the active layer. Photons emerge from an Greater intensity. area in the plane of the More collimated beam. recombination layer. Smaller linewidth.

PHYS5320 Chapter Six 26 LEDs for Optical Fiber Communications

Light is coupled from a surface emitting LED into a multimode A microlens focuses fiber using an index-matching diverging light from a surface epoxy. The fiber is bonded to emitting LED into a the LED structure. multimode optical fiber. Burrus type device

PHYS5320 Chapter Six 27 LEDs for Optical Fiber Communications

Light from an edge emitting LED is coupled into a fiber typically by using a lens or a graded-index (GRIN) rod lens.

PHYS5320 Chapter Six 28 LEDs  Emerging Light Sources

(A) Illustration of the nightly illumination of a gaslight with a thorium oxide- soaked mantle in the 1880s. (B) Replica of Edison’s lamp. (C) Contemporary compact . (D) High-pressure sodium lamp. E. F. Schubert, J. K. Kim, Science 2005, 308, 1274.

PHYS5320 Chapter Six 29 LEDs  Emerging Light Sources

PHYS5320 Chapter Six 30 LEDs  Emerging Light Sources A fluorescent lamp is filled with low-pressure Hg vapor and noble gases (Ar, Xe, Ne, Kr). The inner bulb surface is coated with a fluorescent (and often slightly phosphorescent) coating made of various metallic and rare earth salts. The cathode is typically made of W coated with a mixture of barium, strontium, and calcium oxides, which has a low thermionic emission temperature. When the light is turned on, the electric power heats up the cathode enough for it to emit electrons. These electrons collide with and ionize noble gas to form a by impact ionization. As a result of avalanche ionization, the conductivity of the ionized gas rapidly rises, allowing higher currents to flow through the lamp. Mercury atoms are then ionized, causing it to emit light in the UV region mainly at 254 nm and 185 nm. The efficiency of fluorescent owes much to the fact that low-pressure Hg plasma emit about 65% of their total light at 254 nm and about 10–20% at 185 nm. The UV light is absorbed by the fluorescent coating, which re-radiates the energy at lower frequencies (longer wavelengths in the visible region). The fluorescent coating controls the color of the light, and along with the bulb’s glass prevents the harmful UV light from escaping.

PHYS5320 Chapter Six 31 LEDs  Emerging Light Sources Luminous flux accounts for the sensitivity of the eye by weighting the power at each wavelength with the luminosity function, which represents the eye’s response to different wavelengths. Its unit is lumen (lm). The luminous flux is a weighted sum of the power at all wavelengths in the visible band. Light outside the visible band does not contribute.

  L  683.002lm/W yJ  d 0

L is the luminous flux in lumens. J() is the power spectral density in W/nm. y is the luminosity function (dimensionless).

Lighting efficiency: L Black: photopic luminosity function, for   daylight levels. IV Green: scotopic luminosity function, for low light levels.

PHYS5320 Chapter Six 32 LEDs  Emerging Light Sources The efficiency of fluorescent lamps is limited to ~90 lm/W by a fundamental factor: the loss of energy incurred when converting 250-nm UV photons to visible photons. The efficiency of incandescent lamps is limited to ~17 lm/W by the filament temperature with a maximum of ~3000 K, resulting, as predicted by blackbody radiation theory, in the utter dominance of invisible IR emission. The efficiency of sodium lamps is around 100–200 lm/W. In contrast, the efficiency of solid-state light sources (mostly LEDs) is not limited by fundamental factors but rather by the imagination and creativity of scientists who, in a worldwide concerted effort, are trying to create the most efficient light source possible. The high efficiency of solid- state sources already provides energy savings and environmental benefits in a number of applications.

Solid-state sources also offer controllability of their spectral power distribution, spatial distribution, , temporal modulation, and polarization properties. Such “smart” light sources can adjust to specific environments and requirements, a property that can result in tremendous benefits in lighting, automobiles, transportation, communication, imaging, agriculture, and medicine.

PHYS5320 Chapter Six 33 LEDs  Emerging Light Sources

Perfect materials and devices would allow us to generate the optical flux of a 60-W incandescent bulb with an electrical input power of 3 W.

PHYS5320 Chapter Six 34 LEDs  Emerging Light Sources

PHYS5320 Chapter Six 35 PHYS5320 Chapter Six 36 PHYS5320 Chapter Six 37 Reading Materials

D. A. Neamen, “Semiconductor Physics and Devices: Basic Principles”, Irwin, Boston, MA 02116, 1992, Chapter 14, “Optical Devices”, 14.4 & 14.5.

S. O. Kasap, “Optoelectronics and Photonics: Principles and Practices”, Prentice Hall, Upper Saddle River, NJ 07458, 2001, Chapter 3, “Semiconductor Science and Light Emitting Diodes”, 3.5 – 3.9.

PHYS5320 Chapter Six 38