Core ConceptsIn this article, you will learn about the chemical and physical principles underlying semiconductors, using LEDs as a case study. You will also learn about how doping a semiconductor’s p–n junction with various alloys can alter its electroluminescent properties, such as the color of light emitted. By understanding doping and p-n junctions, you will understand how and why diodes work.What are conductors and insulators?We can classify different materials according to how easily electrons can pass through them. A conductor is a material that facilitates electron flow, readily permitting the conduction of electricity. Abundant in electronic devices and appliances, conductors have a central role in how we utilize electricity in our everyday lives. At the opposite end of the spectrum are insulators, which restrict electron flow and, as a result, minimize electric current. Using insulators is vital to ensuring safety by preventing direct contact between dangerous electrical currents and the people handling them. Insulators can also make electrical applications more energy-efficient, as they enclose current within the conductors carrying it, reducing energy loss.The only way to make electrons move within a circuit is to apply an electric potential (voltage) to the circuit. If the circuit contains a conductor, the electrons will easily pass through the conductor and continue onward in the circuit. However, if the circuit contains an insulator, the insulator will block some of the electrons. This limits the amount of electricity that can pass through the circuit, although no insulator prevents 100% of current flow. Even in the most robust insulators, a small amount of electrons can slip through.Semiconductors: A Happy MediumCompared to conductors and insulators, semiconductors have an intermediate ability to conduct electron flow. More electrons can pass through a semiconductor than an insulator, but less can pass through a semiconductor than a conductor. This makes semiconductors an ideal option for devices that require a controlled electricity supply or feature unidirectional electron flow. For example, a particular device using a conductor may get overwhelmed by too much electric current, whereas using an insulator may not supply the device with enough electricity. In this scenario, using a semiconductor is the compromise that leads to the device’s best performance.When a semiconductor contains two terminals, we refer to it as a diode. Within a diode is a p–n junction, a special structure made of semiconductor material that conducts electricity through the diode. Light-emitting diodes (LEDs) are one common type of diode that you may have encountered in day-to-day electronics and commercial applications. Compared to other diodes, an LED’s p–n junction has the unique ability to release photons. We’ll cover LEDs in more detail soon, but you should understand that this is what makes LEDs “light-emitting.” For now, let’s break down the intricate inner workings of the p–n junction.The p-n JunctionWith so many practical uses, we can find p–n junctions far and wide: in diodes, transistors, solar cells, light sensors, and more. What characterizes a p–n junction is the sequence of materials that comprise it, each of which contributes important electrochemical functions to the diode. In this article, we’ll mainly focus on LEDs as a prominent example of how p–n junctions can produce both light and color: an important function in electronics and similar devices.First, let’s envision an electrical circuit that includes a diode. By applying a voltage to this circuit, electrons in the circuit become excited and produce electricity. Electricity flows, in the form of current, from one of diode’s electrodes (the anode) to the other (the cathode). This is the only direction of current flow within a diode.An electrical circuit featuring a diode. Note that current flows from the voltage source (battery) and reaches the anode first. Current passes through the diode, through the cathode, and continues onward in the circuit.A Tour of the p-n Junction’s StructurePhysically located between the diode’s electrodes are the p–n junction’s two main components: the p side and the n side. Rich with electrons, the negatively-charged n side is functionally analogous to the cathode. Likewise, the p side lacks electrons, and therefore carries a positive charge like an anode. However, just because it lacks electrons doesn’t mean there’s nothing there. Instead of electrons, the p side has electron holes, which are essentially spaces that electrons can occupy.Individually, the n and p sides are each made of semiconductor materials. We can refer to them broadly as an n-type semiconductor and a p-type semiconductor. The junction forms from the sequence of a p-type semiconductor and an n-type semiconductor together.In the center of the p–n junction, connecting the p and n sides, is the depletion region. Unlike the rest of the junction, which is made of semiconductors, the depletion region serves as an insulator, so it boasts very little electrical charge. When an applied voltage stimulates current flow, this region is where electrons and electron holes cross paths with one another. As it turns out, this encounter is critical to the inner workings of the semiconductor.The layout of a p–n junction as seen in a semiconductor diode.How does the p-n junction work?At rest, the electrons in a diode’s n side are in their ground state: not excited and not moving meaningfully. Voltage, when applied to a circuit containing a diode, introduces electricity to the diode. This excites some of the electrons on the n side and sets them in motion.As these electrons reach the p side, they occupy the spaces (holes) there. We can also think of this as if the electrons that reach the p side are pushing the p side’s electron holes toward the n side, or the electron holes are “flowing” to the n side as the electrons flow to the p side. This flow represents the movement of electric charge; in other words, it represents a current.At this stage, the n side has more positive charge and the p side has more negative charge, compared to when the electrons had been in their ground state. This crisscrossing movement of electrons to the p side and electron holes to the n side creates the depletion zone at the center of the p–n junction. Since it has nearly no charge, the depletion zone impedes the electrons and electron holes from any further movement.Meet the Electricity Gatekeeper: The Depletion ZoneAfter having introduced a voltage to stimulate current flow, why would we want the depletion zone to later interfere with that current flow? This principle may seem counterintuitive, but it’s actually a special property of the semiconductor that gives us additional control over the diode. We can harness the depletion zone in two different ways in order to start and stop electricity flow as desired.Let’s say the voltage source in our example is a battery, which has a positive and a negative side. Usually, in an electric circuit, we connect the battery’s positive side to the diode’s p-type semiconductor, and the negative side to the n-type semiconductor. In this format (forward bias) the voltage helps the electrons override the depletion zone and move toward the p side. This makes the depletion zone somewhat smaller — easier for electrons to overcome — so current can flow through the diode.In this circuit diagram, current is able to flow through this diode, and through the rest of the circuit, because the diode is forward-biased.However, we could also flip either the battery or the diode around. In this circuit scenario, the battery’s negative side connects to the p-type semiconductor and its positive side to the n-type semiconductor. We’re effectively applying the voltage in the opposite direction, so we describe this as reverse bias. Unlike forward bias, reverse bias draws electrons away from the center of the semiconductor’s p–n junction, making the depletion zone larger than before. A larger depletion zone presents a stronger barrier to electron movement, so current stops flowing through the diode.Since the diode in this circuit is reverse-biased, current flowing from the battery is unable to flow through the diode.To make a semiconductor conduct and interrupt electricity flow at will, all we have to do is invert either the circuit’s diode or voltage source. As such, a p–n junction provides us with easy control over a circuit, including greater command over the current’s flow and direction.It’s important to note that the depletion region isn’t completely electrically neutral. The side of the depletion region that borders the n side has positive charge, and the side bordering the p side has negative charge. This dynamic results in an electric field. During this cross-diffusion process, the electric field provides a force opposite the direction of diffusion. If the electric field’s force is strong enough to stop electrons and electron holes from diffusing, the depletion region has achieved equilibrium.Bands and Band GapsAn excited electron ascends from its original energy state, the valence band, to a higher one. When this happens, the valence band has an electron hole where the now-excited electron used to be. This is how electron holes arise in the first place. Upon applying a potential difference to the electric circuit, the electrons become excited. This makes the valence bands of the semiconductor’s atoms electron-poor, but rich in electron holes.The higher energy state, which accepts an excited electron, is the conduction band. Being of two different energy levels, the valence band and the conduction band have an energy difference between them. This energy difference is the band gap. Band gaps also represent the amount of energy necessary to excite electrons from the valence to the conduction bands.Valence bands and conduction bands exist on two different energy levels. The energy cost of promoting an electron from the valence band to the conduction band is the band gap.Electrons can only move if there are electron holes available to move into. So, a semiconductor only conducts electricity if at least some of the electrons are excited and in the conduction band. A semiconductor’s band gap is smaller than an insulator’s band gap, but larger than a conductor’s.Recombination in the p-n JunctionYou’ve learned that an excited electron transcends from the valence band to the conduction band and has a higher energy than it used to. We also know that excited electrons are drawn toward the p side of the junction.Excited electrons from the n-type semiconductor and electron holes from the p-type semiconductor aren’t just crossing paths as they flow toward the opposite side. They’re actually joining together in a process called recombination.When an excited electron recombines with an electron hole by “filling” the hole (occupying the empty space), something exceptional happens. The negatively-charged electrons and the positively-charged electron holes have equal, but opposite, charges. Since the magnitude of their charge is the same, their charges cancel each other out. This is why the depletion zone, the region where most of the electrons and electron holes encounter each other, has practically no charge. And because the depletion zone has relatively little charge, there’s a weaker electric field there. This makes it difficult for electrons and electron holes to continue flowing toward the opposite side, so that’s why the depletion zone is like a barrier to current flow.Let’s consider the moment an excited electron fits into an electron hole, and the immediate aftermath. The amount of energy that went into exciting the electron now gets released. Most types of diodes release this energy as thermal energy, or heat. But in LEDs, this energy instead gets released in the form of a photon. Photons are basically particles of light energy, so the LED emits light.Note that, although we describe the movement of electrons from a relatively negative terminal to a relatively positive terminal, we consider electric current to flow from positive to negative. In other words, current moves in the opposite direction that electrons move. This is why, in a diode, current flows from the positively-charged anode to the negatively-charged cathode.A circuit diagram comparing the flow of current (from positive to negative) and the flow of electrons (from negative to positive).In an LED, the semiconductors comprising the p and n sides are often alloys, as we’ll discuss soon. First, let’s put alloys and semiconductors into context using the concept of doping.What is doping?Not all semiconductors are created equal. Different semiconductors can conduct electricity differently depending on which chemical elements make up the p–n junction. Electrical engineers can change a semiconductor’s elemental composition using a technique called doping. Doping a semiconductor means adding small amounts of new elements to the semiconductor, in order to change its electrochemical behavior.The new elements that get introduced are known as dopants. Since elements have various electron configurations, they have varying capabilities to conduct electric current. Tweaking the semiconductor’s elemental composition is a small adjustment — but, as we’ll see shortly, it’s one that requires careful consideration of the electrochemical consequences.So why do we dope semiconductors? Doping is another strategy for manipulating a semiconductor so it has a desired effect, or behaves precisely as intended. It improves a semiconductor’s functionality so we can extend the diode’s function to new applications and contexts. By being selective with our choice of dopants, we can change how electricity flows through the material, and optimize the semiconductor’s performance for whatever its application. Later, we’ll explore light emitting diodes (LEDs) as a specific case study for how doping can customize a semiconductor.How does doping work?An electrical engineer named Nick Holonyak invented the forerunner of the LED: a semiconductor diode that glowed red. This red light was due to doping the semiconductor with an alloy called gallium arsenide phosphide (GaAsP).Ultimately, the band gap determines an LED’s color, but the alloys used in the semiconductor dictate the band gap. The elements in a particular alloy have unique electron configurations, and thus it takes different amounts of energy for an electron in that element’s atom to ascend to the conduction band. The band gap’s energy eventually gets released as a photon, which carries a specific wavelength of light. As we know, distinct wavelengths distinguish different colors in the visible light region.If the band gap between alloys is large, they release a higher-energy photon. Photons with high energy have shorter wavelengths, so their colors appear violet or blue to us. Alternatively, if the band gap is small, a lower-energy (higher-wavelength) photon is emitted, producing colors like orange or red. An intermediate band gap yields photons with intermediate wavelengths, yellow or green in color.Doping an LED’s semiconductor with different alloys is how we change the color of the light it emits. GaAsP is a common, but not the only, alloy used for producing red light. Other alloy options like aluminum gallium arsenide (AlGaAs) and gallium(III) phosphide (GaP) achieve the same color. We can combine alloys in numerous ways to make a wide range of colors spanning the entire rainbow.But how do we determine which alloys to combine? The most crucial point to remember is that elements — their atoms, their electrons — interact with one another when combined. A dopant could provide electron-donor or electron-acceptor properties to the semiconductor’s original material. Which properties, specifically, hinges upon whether we’re doping the n or p side (or both), and the elemental composition of the original material. Historically, semiconductors most frequently feature silicon, so as a basis for understanding this chemistry, we tend to consider how a dopant would react when combined with silicon.N-Type Doping vs. P-Type DopingAs a semiconductor, silicon allows some electrons to flow through it, but in certain cases we may want a diode to conduct more electricity. Doping either the n side or the p side can bestow different properties, such as an improved ability to move electric current, upon the semiconductor.When we dope the semiconductor’s n side, we introduce dopants that have more electrons than the original material (say, silicon). These surplus valence electrons are able to move freely, carrying electric charge along the way. As a result, N-type doping generally makes a semiconductor conduct electricity better. The dopant used in N-type doping is an electron donor to the elements in the original material.Contrarily, the dopant used in P-type doping is an electron acceptor because it has fewer electrons than the original material. Fewer electrons means more electron holes. Remember that electron holes can flow through the semiconductor, too — as positive charge carriers. As the dopant accepts electrons from the original material, it must break some of the existing bonds. This bond-breaking forms electron holes.Although their mechanisms are essentially opposites, N-type doping and P-type doping both achieve the same end goal: supporting the flow of electric current through a semiconductor. Let’s examine the factors that influence which elements and alloys we choose for our dopants.Putting Doping in Context: The Periodic TableMost alloys incorporate elements from Groups III, IV, and V of the periodic table. Group III, IV, and V elements are those found in group (column) 13, 14, and 15 of the periodic table, respectively. We use this (albeit confusing) notation because the realm of semiconductor physics does not describe elements in the periodic table’s modern IUPAC terms. An easy way to conceptualize this is that Group III elements have three valence electrons, Group IV elements have four, and Group V elements have five. Silicon is a Group IV element.An overview of the periodic table. IUPAC group numbers are labeled along the top of the image. Semiconductor dopants are typically p-block elements (depicted here in yellow).However, newer semiconductors use alloys of one Group III element and one Group V element. The crystal lattice structures found in these III-V semiconductors have a wider range of band gaps and, consequently, can produce a greater assortment of colors. III-V semiconductors using some alloys, like indium gallium nitride (InGaN), can even span infrared and ultraviolet wavelengths. This gives LEDs new potential applications even beyond the visible light region.GaAsP is an alloy of two distinct semiconductor materials: gallium arsenide and gallium phosphide. Looking at the periodic table above, we can see that gallium is located in Group III (IUPAC column 13), while arsenic and phosphorous are each in Group V (IUPAC column 15). This means that gallium arsenide and gallium phosphide are both examples of III-V semiconductors. Basically, using a III-V semiconductor dopes the diode with both electron donor atoms and electron acceptor atoms. This facilitates electricity flow and, by extension, alloys like GaAsP expand an LED’s emissive properties and overall utility.Group III elements work best for doping the p side. With room to spare in their valence shell, a Group III atom can accept an electron. If the Group III atom accepts an electron from an adjacent silicon atom, the silicon atom will have an electron hole in its valence band. Conversely, Group V elements are effective dopants for the n side. While four of their valence electrons form bonds with adjacent silicon atoms in the lattice, the fifth valence electron can move about and will readily ascend into the conduction band.Using a silicon semiconductor as an example, each individual silicon atom has a full valence shell. There are no extra electrons and no electron holes.P-type doping: Doping the silicon semiconductor with a Group III element, such as boron, creates an electron hole (void) because boron’s valence shell is not full.N-type doping: Doping the semiconductor with a Group V element, like antimony, introduces an extra valence electron because antimony has five valence electrons.If we didn’t dope some semiconductors, all of our LEDs would be red, just like Holonyak’s original. But with the versatility that doping provides, we can produce LEDs of all colors. This allows us to communicate different messages using different colors and create vibrant visual displays. Take a moment to picture the red, yellow, and green LEDs that comprise a traffic stoplight, or all of the individual hues in your smartphone’s pixels. That’s doping at work.Doping in LEDs: ElectroluminescenceOne of the most significant characteristics of a light-emitting diode is — well, that it emits light. Electroluminescence is a substance’s ability to emit light upon interacting with an electric current or field. As we’ve seen, this is what happens in an LED. Electron excitation within the LED’s semiconductor causes recombination between the electrons and electron holes. Recombination results in emitting photons, which manifests as electroluminescence.As electrons move between energy states, they must overcome the semiconductor’s band gap. In an LED, we can manipulate the band gap by changing the chemical composition of the semiconductor’s alloy. Doping involves purposely infusing the semiconductor’s n side or p side with different elemental impurities to alter the band gap. These impurities are the dopants.Decreasing the band gap makes it easier for electrons to become excited. By doping a semiconductor, we can replace the atoms in its crystal lattice structure with the dopant’s atoms. The dopant’s atoms have their own energy levels, so doping can either decrease or increase the band gap, depending on how the dopant’s energy band compares to the existing atoms’ bands. If the dopant’s energy band is closer to the conduction band, then the energy gap has decreased, and electron excitation is more readily possible.Similarly, increasing the band gap hinders excitation because more energy input is needed to excite the electron. Doping the n side of the semiconductor can lead to occupying energy levels close to the conduction band. When more energy is required to overcome this, it’s harder for electron excitation (and subsequent light production) to occur.Since different energies (wavelengths) result in different colors of light, doping doesn’t just impact an LED’s light production, but its color production too. When designing an LED, we choose which alloy to use according to the size of its band gap and the corresponding energy cost of promoting an electron to the conduction band.ConclusionSemiconductors, such as those used in the p–n junction of diodes, allow a carefully-controlled supply of electric current to a device. The p–n junction’s unique structure features electron and electron hole flow between the junction’s n and p sides in response to an applied voltage. This represents electron excitation, by which excited electrons transcend different energy levels, or bands. Electrical engineers can modify these bands and the amount of energy between them by doping the semiconductor with various chemical elements or alloys. In the case of a light-emitting diode (LED), the excited electrons release light as photons upon returning to their ground state. Doping an LED’s semiconductor changes the wavelength of the photon, thereby changing the color of light that the LED emits through electroluminescence.
