Table of contents
Light-emitting diodes (LEDs) are semiconductor devices that spontaneously emit incoherent, omnidirectional output when forward biased. They are available with operating wavelengths across the electromagnetic spectrum, from the infrared band to the ultraviolet.
Compared to incandescent, fluorescent, or even halogen lighting, LEDs are more efficient, more compact, and more robust. Because they are fabricated using batch processing, they are also inexpensive. For many applications, they provide very effective solutions.
How are LEDs used?
LEDs support a variety of use cases:
- Indication lighting: LED indicator lights communicate device status to the user. Depending on the circuit design, an LED can be driven in a number of modes: lit versus unlit, flashing versus steady-state, etc. Color LEDs, especially multicolor devices, enable more sophisticated communications, e.g., red for alerts, green for operational, blue for Bluetooth connection, etc.
- Functional illumination: From lighting up buttons to illuminating informational displays like meters on a battery charger, LEDs help guide user operations and communicate progress on a variety of devices.
- Design lighting: LEDs can be used in conjunction with overlays to create sophisticated lighting effects such as illuminated logos or other types of lighting modulation when equipment is in operation.
- Sensors: Infrared LEDs play essential roles in proximity sensors and presence/absence sensors for everything from automatic garage doors to smart refrigerators.
- Display lighting: LEDs are commonly used for display lighting in electronics ranging from smart phones to wearables to other portable devices. They are also used for large-scale displays and signage.
- General illumination: With the push to eliminate incandescent bulbs, white-light LEDs have made great inroads in the general lighting market. Their high efficiency, long lifetime, and general robustness have made them increasingly popular alternatives to incandescent, halogen, and even compact fluorescent bulbs.
What industries do LEDs serve?
LEDs support the communications industry throughout the vertical stack, from networking gear like routers, switches and transponders to point-of-use equipment like servers and handsets. They are primarily used for indication lighting and display backlighting/site lighting.
The healthcare industry depends heavily upon electronics and its electronics depend heavily on LEDs. From wrist-mounted heart rate monitors to large-frame diagnostic equipment like bone-density scanners, LEDs enhance user interfaces, illuminate displays, and assist healthcare professionals with the functionality of medical equipment. They’re effective in point-of-care equipment ranging from hospital beds to lifts in wheelchairs, as well as lab instrumentation.
Industrial automation and control
The durability and long lifetime of LEDs makes them an excellent fit for the industrial environment. LED indication lighting is in wide use for equipment like motor drives and controllers. The color and controllability of LEDs makes them effective for machine-vision lighting and they also serve as backlights for instrument panels and human-machine interfaces (HMIs).
EVs, autonomous vehicles, automotive, and transportation
Efficient and insensitive to shock and vibration, LEDs are widely used in automotive, autonomous vehicles, and EVs, not to mention infrastructure like charging stations and gas pumps. LEDs support user interfaces and instrument panels, backlight displays, and add exotic design elements while minimizing power consumption.
Robust, high-efficiency, and just a few millimeters on a side, LEDs are ideal sources for portable devices ranging from consumer electronics to the multitude of devices that make up the Internet of Things (IoT). LEDs are robust enough to survive impact and vibration while maximizing operating lifetime per charge.
How do LEDs work?
LEDs are semiconductor devices that emit light through electroluminescence. Electroluminescence occurs when current flows through the p-n Junction of a direct-bandgap material (see figure 1). In a p-n junction, a p-type region (material with excess holes) is placed in contact with an n-type region (material with excess electrons). A depletion region—an area depleted of charge carriers—forms immediately on either side of the junction.
Because the depletion region has no mobile charge carriers, it acts as an insulator rather than a conductor, preventing the flow of electrons. This is the steady-state condition. Now, an LED is just a special variety of diode, so it can only pass current when forward biased. When we forward bias the p-n junction, the free electrons in the n-type material gain sufficient energy to cross the depletion region and recombine with the holes in the p-type region. (Note: Like any other diode, an LED will block current when reverse biased, until it exceeds maximum voltage, after which breakdown will occur.)
If we consider an energy diagram for the material, the free electrons populate the conduction band, while the holes populate the valence band (see figure 2). Between the two is the bandgap, which represents energies outside of the allowed quantum states for the atoms.
Applying a forward bias to the junction narrows the depletion region, which enables the electrons in the n-type region to cross the junction and recombine with holes in the p-type region. Because the conduction band is at a higher energy level than the valence band, recombination releases energy. For the right compound semiconductor, this release takes the form of a photon and heat.
From physics, we know that
where ƛ is wavelength, h is Planck’s constant, c is the speed of light in a vacuum, and E is energy. Producing an LED in a specific color just requires developing a compound semiconductor with the corresponding bandgap energy. That’s easier said than done, of course, but LEDs are now available with output across the electromagnetic spectrum, from the infrared to the ultraviolet wavebands (see table).
|Color||Wavelength [nm]||Semiconductor material|
|Infrared||λ > 760||Gallium arsenide (GaAs)
Aluminium gallium arsenide (AlGaAs)
|Red||610 < λ < 760||Aluminium gallium arsenide (AlGaAs)
Aluminium gallium indium phosphide (AlGaInP)
Gallium arsenide phosphide (GaAsP)
Gallium(III) phosphide (GaP)
|Orange||590 < λ < 610||Aluminium gallium indium phosphide (AlGaInP)
Gallium arsenide phosphide (GaAsP)
|Yellow||570 < λ < 590||Gallium arsenide phosphide (GaAsP)
Aluminium gallium indium phosphide (AlGaInP)
Gallium(III) phosphide (GaP)
|Green||500 < λ < 570||Traditional green:
Gallium(III) phosphide (GaP)
Aluminium gallium indium phosphide (AlGaInP)
Aluminium gallium phosphide (AlGaP)Pure green:
Indium gallium nitride (InGaN) / Gallium(III) nitride (GaN)
|Blue||450 < λ < 500||Zinc selenide (ZnSe)
Zinc selenide (ZnSe)
Indium gallium nitride (InGaN)
Synthetic sapphire, Silicon carbide (SiC) as substrate with or without epitaxy,
Silicon (Si) as substrate—under development (epitaxy on silicon is hard to control)
|Violet||400 < λ < 450||Indium gallium nitride (InGaN)|
|Ultraviolet||λ < 400||Indium gallium nitride (InGaN) (385-400 nm)
Diamond (235 nm)
Boron nitride (215 nm)
Aluminium nitride (AlN) (210 nm)
Aluminium gallium nitride (AlGaN)
Aluminium gallium indium nitride (AlGaInN)—down to 210 nm
LED output characteristics
Several key metrics are used to express LED output:
- Luminous flux—luminous power emitted by the LED, in units of lumens (lm)
- Luminous intensity—luminous power emitted per unit solid angle in a specific direction, in units of candelas (cd)
- Luminance or brightness (cd/m2)—luminous intensity per unit area
LED output has traditionally been specified in terms of luminous intensity or luminance. With the rise of general lighting applications, however, LEDs are increasingly specified in terms of lumens.
The higher the drive current, the brighter the output of the LED—to a point. When the current goes too high, it can cause multiple problems. First, it will shorten lifetime. Second, past a certain current level, the dominant recombination mechanism shifts from radiative recombination to a non-radiative process known as Auger recombination. This change in the amount of energy converted to light causes a significant drop in conversion efficiency known as droop. Finally, when current goes high enough, it will burn out the LED entirely.
Figure illustrates energy levels as uniform. In reality, energy levels vary slightly from electron to electron, depending on which orbitals they occupy. Because of this, the energy released by recombination—and the wavelength of the emitted photons—also varies. This broadens the output bandwidth of an LED to between 50 and 100 nm. In contrast, laser diodes, which operate based on stimulated emission, generate narrowband output (1 to 2 nm).
Even in a well-controlled process, LED output can vary from batch to batch in terms of characteristics like dominant wavelength/color, luminance, and voltage. As a result, LEDs are typically binned—sorted for a specific characteristic—after fabrication.
Photons are emitted from the LED die over range of angles. Those that are at near-normal incidence with respect to the input face of the case will propagate across the interface. At larger angles, they may refract or scatter. For sufficiently large angles, they will undergo total internal reflection and never escape the LED (see figure 3).
For this reason, maximizing light extraction is a primary focus of LED design. As part of packaging, the LED die is placed in a reflective cone designed to direct as much output as possible in the forward direction (see below).
Even when photons propagate into the case or lens, they can still exit over a variety of angles. Several metrics are used to characterize LED output. The first is viewing angle. We define viewing angle as the full width at half maximum (FWHM) of the luminous intensity. Viewing angle may be expressed as either full angle or half angle. LEDs are available with full viewing angles ranging from 30° to 130°.
The ideal viewing angle depends upon the application. For user interfaces designed for front viewing, brightness is most important, so an LED with a narrow viewing angle (30°) will provide best results. For interfaces designed to be viewed from the side or for a variety of illumination applications, a broad viewing angle will deliver the most uniform and effective performance.
A directivity radiation diagram provides another way to characterize LED output. These diagrams display normalized output as a function of angle (see figure 4). Basically, it provides an at-a-glance measure of how much of the output is forward directed.
How are LEDs constructed?
We’ve talked about the physics of LED operation. Now let’s look at how the technology is applied to produce a practical device. LEDs are designed from the chip level on up to maximize light extraction and minimize loss. Both recombination and photon emission take place on the p-region side of the p-n junction. To minimize loss, LEDs are fabricated with the n-type layer next to the substrate and the p-type layer on top, reducing the distance photons travel from the junction to the surface.
LED dies are around 0.25 mm on a side. To be forward biased, they are connected to a two-part lead structure. This consists of a cathode (negative lead), which is topped by a structure called the anvil, and the anode (positive lead), also called the post (see figure 5). The actual LED die is placed in a reflective cone on the top of the anvil and connected to the post by a very thin wire.
The actual semiconductor die of an LED is very small and the leads are fragile. To protect the structure and simplify handling, LEDs are either encapsulated in epoxy-resin structures or mounted in ceramic substrates (more on this later).
Multicolor LEDs are in widespread use for indication lighting, displays, and design illumination. They can switch between two or three distinct colors or output a continuum of shades. Because LED output color is determined by bandgap, the devices are not tunable in the same sense as a laser. Instead, multicolor LEDs are based on two or more co-packaged LEDs that are individually biased to generate the desired output.
Bicolor LEDs are two-pin devices that incorporate two co-packaged dies connected anode to cathode (see figure 6). Consider a green/red LED. When the green LED is forward biased, the green LED will emit light. When the red LED die is forward biased, the device will emit red light.
In a tricolor LED, two LEDs are once again co-packaged. This time, however, each has a separate anode but a common cathode (or vice versa). As with the bicolor LED, when either of the LEDs is forward biased, it emits light (see figure 7). The common cathode another operating mode, which is forward biasing both LEDs simultaneously. In this case, color mixing of the light can create third color.
The final type of multicolor LED is the RGB LED. These devices contain co-packaged red, green, and blue LEDs. They are available as four-pin designs with a common cathode, four-pin designs with a common anode, or six-pin designs that incorporate anode and cathode pairs for each die.
The operating principle is similar to that of the tricolor LED. By using a microcontroller to precisely adjust the drive current and, hence, the brightness of each die, it’s possible to generate any of a broad continuum of colors.
LEDs are available in a variety of mounting styles, making it easy to optimize designs to meet the performance, form factor, and environmental requirements of the application.
The through-hole form factor and mounting style is the one classically associated with LEDs. Through-hole LEDs are available in standardized sizes of 1.8 mm, 3 mm, and 5 mm. In this mounting style, the pins are passed through holes in the PCB and soldered on the back side. The bond provides a high degree of structural stability, making them good fits for devices that will be subject to high degrees of shock and vibration.
As previously noted, through-hole LEDs are encapsulated in epoxy resin cases (see figure 8). These cases provide mechanical and environmental protection for the LED dies. Although the packaging is often colored, this is typically more for easy identification rather than to modify the output wavelength. Remember, the color of the light output is determined by the semiconductor bandgap. A blue LED encapsulated in clear plastic will still shine bright blue.
The shape of the packaging varies depending on the application. In some LEDs, the case is rectangular, which is well-suited for smooth surfaces. In other instances, it is domed. Domes can be designed to shape the output characteristics of the LED by controlling scatter and guiding the light to provide a wider or more narrow viewing angle.
The most basic through-hole LED mounting style is vertical. Specialty adapters include vertical adapters that elevate the LED from the PCB. For applications that require the output direction to be at right angles to the board, the LED is placed in a right-angle mount (see figure 9). Note, in these mounts, the leads need to be threaded through the mount and bent to go through the PCB holes.
Through-hole LEDs are simple and familiar devices. On the downside, they require holes to be drilled to the PCB, limiting board layouts and adding a step to the assembly process. They typically require manual installation, making the assembly both more time intensive and more expensive than their SMD counterparts. For applications requiring a robust solution that can stand up to punishment, however, through-hole LEDs are an ideal choice.
Surface-mount (SMD) LEDs
Surface-mount LEDs are packaged to be soldered directly to the board. In an SMD LED, the dies are packaged in a molded ceramic substrate (see figure 10). As with through-hole LEDs, the dies are placed in reflective cones and connected to anode and cathode. SMD LEDs are available as bicolor, tricolor, and RGB devices. They are driven in the same fashion as their through-hole counterparts. That’s where the similarity ends, however.
SMD LEDs are low-profile and can be smaller than their through-hole cousins. Surface mounting eliminates the need for drilling holes in the PCB and for manual insertion. Instead, pick-and-place robots can do the job, dramatically reducing both cost and assembly time. Fewer holes mean fewer constraints on board layout, simplifying PCB design.
On the downside, the soldering step is more complex than for through-hole LEDs. Although the solder provides a tight bond to the board, SMDs do not have the extra mechanical strength provided by the pins of through-hole LEDs.
For applications in high temperature environments, screw mounting provides a useful tool for thermal management. In a screw-mounted LED, the housing of the device includes threaded holes for conductive screws to improve heat dissipation. Alternatively, the device can be mounted to the board using thermal adhesive.
Screw mounting is a slow and labor-intensive process. For heat-intensive applications in which reliability trumps costs, screw mounted LED may be appropriate.
Getting LED output to point of use
Space on the PCB is always in short supply. The open location may be far from the point of use or even on a different plane. This is where light pipes come into play. A light pipe is an optically transparent or translucent structure that transmits LED output from the source to the user interface with minimal optical loss. They can be paired with LEDs to meet countless application requirements.
Light pipes can be classed as rigid or flexible:
- Rigid light pipes: Typically made of polycarbonate, these monolithic structures typically transmit light over straight lines or limited angles. They are available in both panel press-fit and board-mount styles. They provide simple, easy to install solutions for distances of 3 inches or less. With proper design, they can offer ingress protection and reliability even in the face of shock and vibration.
- Flexible light pipes: Made of optical fiber, these components must be mounted to the board using specialized adapters. They are very effective for transmitting light over complex paths and longer distances. They can be configured to provide ingress protection and ruggedized to provide shock and vibration resistance.
To simplify assembly and enhance performance, light pipes are available as integrated subassembly systems. In the case of adapters for board-mounted light pipes, for example, adapters are available with integrated SMDs. The device only requires the subassembly system adapter/ LED to be soldered to the board. If the light pipe is rigid, it just needs to be press fit into the adapter. If the light pipe is flexible, it just needs to be attached.
LEDs play essential roles in a vast range of products across many industries, including communications, healthcare, industrial operations, transportation, portable devices and consumer electronics. They frequently act only as support technologies, but they are frequently the first thing the user sees and few devices could function without them.
LEDs provide compact, economical, robust, and flexible solutions for use cases like indication lighting, functional illumination, design illumination, display illumination, and sensors. By considering the design options discussed in this pillar page, OEMs can use LEDs to help realize their product visions and delight their customers. Contact Bivar to learn more.