How OLED Displays Actually Work: The Physics Explained
Most display explainers treat OLED as a black box — better blacks, punchier colors, thinner panels. That is useful shorthand for a buying guide, but it skips the genuinely interesting part: understanding how OLED displays work at the physics level reveals that every major strength and every major weakness of the technology flows from a single architectural fact. In an OLED panel, each pixel is its own light source. There is no shared backlight. No filter bank blocking most of what a lamp produces. Each organic molecule in each subpixel either emits a photon right now or it does not. That per-pixel emission is the whole story. True blacks exist because unneeded pixels are simply off. Burn-in exists because pixels that are always on age faster than pixels that are rarely on. Pull on that one thread and the rest of the technology unravels in a satisfying way.
What this covers: the physics of electroluminescence in organic materials, the full OLED device stack layer by layer, the three main methods of making color, how the AMOLED thin-film transistor backplane actually drives each pixel, why burn-in happens at the chemistry level, and a grounded comparison with LCD and microLED.
What Makes OLED Different: Per-Pixel Light
To appreciate why per-pixel emission matters, it helps to know what it replaced. A liquid-crystal display (LCD) works by shining a bright backlight — usually an array of LEDs — through a sheet of liquid crystals that can be electrically twisted to block or pass light, then through a color filter. The backlight is always on. When you want a black pixel, the liquid-crystal layer tries to block the light, but it never blocks it perfectly. Light leaks. That residual glow is why LCD black levels measure in the range of 0.01 to 0.05 nits rather than zero. The contrast ratio is real but bounded.
An OLED pixel does not filter or block light from somewhere else. It either produces light or it does not. Switch it off and it contributes exactly zero nits to the image — true black is the absence of emission, which is simply the absence of power. This is what the display industry means by infinite contrast ratio: the denominator in that ratio is genuinely zero, not a very small positive number.
That same fact — per-pixel emission — is what makes burn-in possible. If you leave a static element on an OLED screen for thousands of hours, the organic molecules in those pixels gradually degrade and emit less light for a given drive current. Adjacent pixels that were less active remain brighter. You now have a ghost image baked into differential aging. The strength and the weakness are the same property seen from two directions.
The Physics of Light from an OLED
Electroluminescence in organic molecules
The formal name for what OLED relies on is electroluminescence — light generated by electrical excitation, as opposed to photoluminescence (light from light) or incandescence (light from heat). The organic materials used are conjugated small molecules or polymers with alternating single and double carbon bonds. That alternating structure produces a set of delocalized pi electrons that can be promoted to excited electronic states with relatively modest applied voltages.
When you apply a voltage across an OLED device, two populations of charge carriers move toward the emissive layer from opposite directions. Electrons, injected from the cathode, move through an electron transport layer. Holes — the quantum-mechanical absence of an electron in the valence band, behaving as positive charge carriers — are injected from the anode and move through a hole transport layer. When an electron and a hole meet in the emissive layer, they are coulombically attracted and form a bound state called an exciton. The exciton is a quantum of stored energy sitting on an organic molecule.
The molecule then relaxes to its ground state and releases that energy. In a well-engineered organic material, a significant fraction of that relaxation happens radiatively — a photon is emitted whose wavelength corresponds to the energy gap of the particular molecule used. The color of light produced is set by the chemistry of the emitter: alter the molecular structure and you shift the emission peak. This is why OLED panels can produce saturated, accurate colors without subtractive color filters stealing efficiency.
There is one quantum-mechanical wrinkle that kept OLED efficiency low for the first generation of devices. Excitons form with either singlet or triplet spin configurations in roughly a 1:3 statistical ratio. Simple fluorescent emitters can only harvest singlet excitons radiatively; triplets relax non-radiatively and generate heat instead, capping efficiency at around 25%. Modern OLED devices use phosphorescent emitters or thermally activated delayed fluorescence (TADF) emitters that can harvest triplet excitons too, raising internal quantum efficiency substantially.
The device stack
Understanding the layer structure of an OLED device makes the physics tangible.
Figure 1: The OLED device stack. Electrons injected from the cathode meet holes injected from the transparent ITO anode in the emissive layer (EML), forming excitons that relax to emit photons.
Substrate. Usually glass, though flexible OLED panels use polyimide films. This is the mechanical backbone. In bottom-emission devices, light exits through the substrate; in top-emission devices (common in AMOLED), it exits through a thin semi-transparent cathode.
Anode. Typically indium tin oxide (ITO), a transparent conducting oxide. ITO is chosen because it is optically transparent across the visible spectrum and has a high work function, meaning it can efficiently inject holes into organic hole-transport materials.
Hole Transport Layer (HTL). An organic semiconductor chosen to have a highest occupied molecular orbital (HOMO) energy level close to the work function of ITO and to the HOMO of the emissive material. Its job is to act as a conduit for holes moving inward and as a partial electron blocker preventing electrons from overshooting the emissive zone.
Emissive Layer (EML). The heart of the device. Contains the light-emitting dopant molecules dispersed in a host material. The host transports charge and transfers energy to the dopant; the dopant is the actual emitter. The dopant concentration is carefully tuned — usually a few percent by mass — to avoid concentration quenching, where excitons on adjacent emitter molecules annihilate each other non-radiatively.
Electron Transport Layer (ETL). Mirrors the role of the HTL on the electron side, conducting electrons inward from the cathode and partially blocking holes from escaping.
Cathode. A low-work-function metal such as aluminum capped with a thin lithium fluoride (LiF) interlayer, or magnesium:silver alloys. The low work function is necessary because electrons need to be injected into the lowest unoccupied molecular orbital (LUMO) of the organic material at the cathode interface.
In a tandem OLED — now used in many premium panels to achieve higher brightness without over-driving any single organic layer — multiple EML units are stacked with charge generation layers between them. Samsung Display’s 2026 QD-OLED panels use a five-layer tandem stack (marketed as QD-Penta Tandem), which the company reports improves luminous efficiency by roughly 1.3 times and extends the panel’s operational lifetime by roughly two times compared with their four-layer predecessor, with peak brightness reaching approximately 4,500 nits in the latest generation.
Making Color: RGB, WOLED, and QD-OLED
Three distinct engineering approaches are used in commercial OLED panels to generate the full color gamut, and they represent genuinely different trade-offs in efficiency, color accuracy, manufacturability, and longevity.
Figure 2: The three main color strategies in OLED panels. RGB OLED deposits red, green, and blue emitters side by side. WOLED adds a color filter array over a white emitter. QD-OLED converts blue OLED light using quantum dots to generate red and green subpixels.
RGB side-by-side OLED
The conceptually simplest approach: deposit separate red-emitting, green-emitting, and blue-emitting organic materials in adjacent subpixels. Each subpixel produces only its own color; no filtering or conversion is involved. Efficiency is high because you use almost every photon generated. Color is accurate and saturated.
The manufacturing challenge is deposition. Small-molecule organic materials are typically deposited by thermal evaporation through fine metal shadow masks (FMMs) — physical stencils that define where each color’s molecules land. At phone-sized panels, this works well; at TV-sized substrates, the masks sag under their own weight and registration accuracy degrades. This is why Samsung’s OLED phone displays use RGB OLED but TV-sized panels require different approaches. High-resolution inkjet printing of RGB OLED materials is under active development and may change this constraint.
White OLED plus color filter (WOLED)
LG Display’s approach for large-panel OLEDs (used in LG, Sony, Panasonic, and Philips OLED TVs) is to deposit a white-emitting OLED stack — typically a tandem structure with complementary emitters that together produce broadband white light — and then place a conventional color filter array on top. Red, green, blue, and often a white or yellow unfiltered subpixel are defined by the filter pattern.
The advantage is manufacturability: one organic deposition step covers the whole panel, avoiding the FMM registration problem. The disadvantage is efficiency loss at the color filters (which absorb a large fraction of light) and some compromise in color saturation, since the filter passbands are necessarily broad to recover enough brightness. The unfiltered white subpixel helps maintain luminance on bright content but can shift white balance if not carefully calibrated.
QD-OLED (quantum dot color conversion)
Samsung Display’s QD-OLED architecture, now in its fifth generation with cumulative shipments exceeding five million units as of early 2026, takes a different path. A blue OLED emitter stack covers the entire panel. Over the red and green subpixels, a layer of quantum dots absorbs the blue photons and re-emits them at precisely tuned red and green wavelengths. The blue subpixel passes through directly.
The quantum-dot color conversion step is the key to understanding both the strengths and limitations. Quantum dots are semiconductor nanocrystals — typically cadmium selenide or, in newer commercial products, indium phosphide — whose emission wavelength is set by their physical size. Because quantum confinement shifts the bandgap with particle diameter, you can synthesize dots tuned to virtually any visible wavelength. The emission peaks are narrow, producing vivid, accurate colors and wide color gamut coverage.
The efficiency of the conversion step is high but not perfect: some blue photons are absorbed and do not convert, and the conversion itself is not lossless. On the positive side, QD-OLED panels avoid the large filter absorption losses of WOLED. The resulting color volume — the intersection of brightness and color gamut — is generally superior to WOLED at the same brightness level. The 2026 generation, reaching approximately 4,500 nits peak, demonstrates how tandem blue emitters combined with efficient quantum dot conversion layers can push brightness well beyond what earlier OLED architectures could reach.
To understand how quantum dots achieve tunable emission from size alone, see the companion piece on the physics of quantum dots.
Driving the Pixels: The AMOLED Backplane
A display is not just an array of emitters — it is also a control system that must independently address potentially tens of millions of subpixels at refresh rates of 60 to 240 times per second. How that is achieved separates passive-matrix OLEDs from the active-matrix OLEDs in every modern smartphone and television.
Figure 3: The AMOLED 2T1C pixel drive circuit. A switch transistor (T1) gates the data signal onto a storage capacitor. The capacitor voltage then controls the gate of a drive transistor (T2), which sets the current flowing through the OLED for the entire frame period.
PMOLED: the simple case
In a passive-matrix OLED (PMOLED), rows of anodes and columns of cathodes form a grid. To illuminate a pixel, you apply voltage to the appropriate row and column simultaneously — like selecting a cell in a spreadsheet by row and column address. Driving happens sequentially, one row at a time. To achieve adequate average brightness, each pixel must be driven at high instantaneous current during its brief on-time.
This works well for small, low-resolution displays — wearable secondary displays, indicator panels, small audio equipment screens. It fails for high-resolution panels because the instantaneous drive current required to compensate for a short duty cycle degrades the organic materials rapidly. It also makes individual pixel brightness control imprecise at scale.
AMOLED: a transistor per pixel
Active-matrix OLED (AMOLED) solves both problems by giving each pixel its own drive circuitry embedded in a thin-film transistor (TFT) backplane — a layer of transistors and capacitors fabricated on the substrate beneath the organic layers. The canonical minimal circuit is called 2T1C: two transistors and one capacitor per pixel.
The switch transistor (T1) is normally off. When the row select line (gate line) pulses high, T1 turns on and the data voltage on the column (data line) charges the storage capacitor. T1 then turns off and the row is released. The storage capacitor now holds the data voltage for the entire frame period, continuously biasing the gate of the drive transistor (T2). T2 controls the current flowing from the power supply line through the OLED diode to the common cathode. Because T2 sets current rather than voltage, and because OLED brightness is roughly proportional to current density, the storage capacitor voltage directly encodes the pixel’s gray level.
This architecture means every pixel is driven continuously throughout the frame, not just during its brief row-select moment. High average brightness requires only modest drive currents, reducing stress on the organic layers. Independent pixel control is precise.
In practice, real AMOLED backplanes use more than 2T1C — typically four to eight transistors per pixel — to compensate for transistor threshold voltage variation across the panel (which would otherwise create brightness non-uniformity), to enable local dimming control, and to support features like high-frame-rate or always-on display modes. Low-temperature polycrystalline silicon (LTPS) and oxide semiconductor (IGZO) are the two main TFT technologies used. LTPS offers higher electron mobility and is preferred for high-refresh-rate displays; IGZO has lower leakage current and is better suited for always-on modes where the storage capacitor must hold its charge for longer periods.
The TFT backplane also determines the physical form factor possibilities. Because the OLED emitters are deposited on top of the TFT layer, and because modern OLED emitters can be deposited on flexible polymer substrates, the entire stack can be manufactured on flexible foil. This enables curved displays, rollable displays, and foldable phones — none of which are possible with rigid glass LCD panels. The physics of flexible light emission from an organic thin film is what makes the foldable phone segment possible, and it is the same physics described above, now fabricated on a substrate that bends.
For a comparison with how CMOS image sensors manage per-pixel signal acquisition in an analogous active-matrix architecture, see how CMOS image sensors work.
Burn-In and the Limits of Organic Emitters
Burn-in is the defining long-term limitation of OLED technology, and its mechanism is worth understanding at the chemistry level — not because it will change your purchasing decision dramatically, but because it clarifies what you can and cannot do about it.
Differential aging
Every organic emitter degrades over time. The degradation is predominantly driven by the cumulative dose of exciton formation events in the emissive layer. Each time an exciton forms and either emits a photon or decays non-radiatively, there is a finite probability of a chemical transformation — bond breaking, molecular rearrangement, or production of quencher species that can migrate through the organic layer and deactivate nearby emitter molecules. The degradation rate is non-linear: it accelerates with drive current density, so pixels running at high brightness degrade faster than pixels running dimly.
In a typical use scenario, the pixels displaying a static user interface element — a navigation bar, a keyboard, a news ticker logo, a game HUD — accumulate a higher cumulative exciton dose than pixels displaying changing video content. Over time those high-use pixels become measurably dimmer than their neighbors. Because the eye is remarkably sensitive to spatial brightness nonuniformity, the ghost image of the static element can become visible under certain conditions (a solid gray background is the classic reveal test).
The blue OLED lifetime problem
The situation is made more complicated by the fact that red, green, and blue organic emitters do not age at the same rate. Blue organic emitters have historically been the shortest-lived component of any OLED stack. The photon energy of blue light is higher than red or green — approximately 2.6 to 2.8 eV for deep blue emission versus roughly 2.0 to 2.1 eV for green and approximately 1.8 to 1.9 eV for red. Higher photon energy means the excited molecular state holds more energy, and more energetic excited states are more chemically reactive. The probability of a degradative side reaction per exciton formation event is higher for blue emitters.
This is precisely why QD-OLED’s architecture is interesting from a longevity standpoint: the quantum dot conversion layer is photoluminescent and does not itself undergo electroluminescent stress. The blue OLED emitter still ages, but only the blue subpixel of each pixel is an electroluminescent blue emitter. The red and green output comes from quantum dots that are excited optically — by the blue OLED light — and quantum dots are generally far more photochemically stable than organic emitters under typical use conditions. Whether this translates to meaningfully longer panel life in practice is a product-generation question; Samsung’s reported doubling of operational lifetime in the 2026 tandem stack is encouraging but represents a manufacturer’s own characterization under specific test conditions.
Pixel-level compensation
Panel manufacturers address burn-in through several engineering countermeasures. Automatic brightness limiting (ABL) circuits reduce drive current when large areas of the display are at high brightness — reducing peak stress at the cost of peak luminance on bright images. Panel-level calibration routines (sometimes called pixel refresh or OLED care cycles) run during idle periods and attempt to measure the response of each pixel and adjust its drive current to compensate for differential aging, extending the perceptible uniformity of the panel for longer. Screensavers, pixel-shift routines that slightly jitter the image to spread usage, and enforced screen dimming after periods of static content are all mitigation strategies baked into modern OLED devices.
These mitigations are effective but not infinite. A panel running a static video feed or a fixed HUD for many hours per day will develop nonuniformity that compensation algorithms cannot fully conceal indefinitely. Understanding that the mechanism is chemistry — not software — sets appropriate expectations.
OLED vs. LCD vs. MicroLED: Honest Trade-offs
No display technology is unconditionally superior; each is a different engineering compromise.
LCD remains the dominant display technology by unit volume and continues to improve. Mini-LED LCD — arrays of small LEDs that allow zone-level local dimming — closes the contrast gap with OLED significantly, though local dimming zones still produce halo artifacts around bright objects on dark backgrounds that per-pixel OLED emission cannot. LCD does not burn in. LCD backlight brightness can be extremely high (some mini-LED monitors exceed 2,000 nits across the full panel, and specialized units go higher), and the liquid-crystal layer itself does not degrade perceptibly at normal usage levels. For content that spends long periods displaying static elements, or for environments with very high ambient light, LCD remains a sensible choice.
MicroLED is conceptually the closest technology to OLED: it also uses per-pixel emission, so it also has true blacks and infinite contrast. The emitters are inorganic gallium nitride LEDs at micron scale rather than organic molecules, which gives them substantially higher efficiency, higher peak brightness, and far greater longevity — inorganic LEDs do not suffer the same exciton-driven degradation that organic emitters do. The major unsolved problem is manufacturing: transferring millions of microscopic LED chips onto a substrate with sufficient yield, speed, and uniformity remains extremely difficult and expensive. As of mid-2026, consumer microLED products exist but at price points that restrict them to large commercial and luxury applications. The technology is credible; the economics are not yet competitive.
OLED sits in between: better blacks and thinner form factors than LCD, existing at consumer price points today in both phone and TV formats, with the burn-in and peak brightness trade-offs described above. The relevant comparison is always against the specific OLED variant and use case. A QD-OLED panel at 4,500 nits peak is a different proposition than a first-generation WOLED from 2015.
For the analogy in silicon photonics — where per-photon detection and emission at the chip level raises similar engineering trade-offs — see how silicon photonics chips work.
Practical Recommendations
Understanding the physics leads directly to practical guidance that goes beyond “OLEDs are great, but watch out for burn-in.”
Before you buy:
– Match the technology to your content. OLED is exceptional for dark-room cinema, gaming with variable scene content, and mixed-use personal computing. It is a higher-risk choice for a dedicated broadcast monitor running a static ticker 12 hours a day.
– Compare WOLED and QD-OLED for your use case. WOLED’s mature manufacturing makes it available in more screen sizes; QD-OLED’s color volume advantage is most apparent in high-dynamic-range content with bright, saturated highlights.
– Do not over-index on peak nit figures. Sustained full-screen brightness is often much lower than peak brightness, which is measured on a small window. Read independent measurements that test sustained brightness, not just the manufacturer spec.
– On monitors, check whether the panel has automatic brightness limiting that significantly reduces brightness on white documents. Some OLED monitors throttle aggressively enough to be distracting for text work.
After you buy:
– Enable any built-in pixel refresh or OLED care features. They work, and they cost nothing.
– Use dynamic wallpapers and avoid leaving the same static image up for extended periods.
– Engage auto-brightness so the panel reduces drive current in ambient light conditions that don’t require maximum brightness.
– For monitors specifically: use a screensaver or sleep mode; the operating system’s taskbar and static application chrome are exactly the kind of static elements that accumulate differential aging.
– Do not run the panel at maximum brightness continuously if your content does not require it.
None of these steps prevents aging entirely — they only distribute it more evenly and slow the rate at which nonuniformity becomes visible.
FAQ
Does OLED burn-in still happen in 2026, or has it been solved?
Burn-in has been substantially mitigated in current-generation panels through tandem emitter stacks, improved blue OLED materials, and smarter compensation algorithms — but it has not been eliminated. The underlying chemistry of organic emitter degradation still applies. Panels used for typical home cinema and general computing workloads show far less burn-in than early generations did, but users running static content for very long daily hours — professional monitors, commercial signage, gaming with persistent HUDs — should still account for differential aging as a real constraint. The risk is lower than it was five years ago; it is not zero.
Why does OLED produce better blacks than QLED or mini-LED LCD?
QLED and mini-LED LCD still use a backlight that illuminates the entire panel or large zones of it. Even when a local dimming zone is suppressed to minimum, residual light leaks through the liquid-crystal layer and into adjacent zones, producing a visible glow on dark content. An OLED pixel that is told to display black receives no drive current and emits no light at all — the measured luminance of an off OLED pixel in a darkened room is effectively the panel’s ambient reflectance, not a light source. The difference is visible to the naked eye in dark room viewing.
What exactly is a quantum dot, and why does QD-OLED produce more vivid colors?
A quantum dot is a semiconductor nanocrystal, typically two to ten nanometers in diameter. At that scale, quantum confinement shifts the electronic energy levels in a size-dependent way, so the dot’s optical emission wavelength is determined by its physical size. The emission peaks of quantum dots are narrow — typically 20 to 30 nanometers full width at half maximum — compared to the much broader emission spectra of conventional color filters. Narrow emission peaks mean the red, green, and blue subpixels have less spectral overlap, which translates to a wider and more accurately defined color gamut. The full quantum dot physics story is covered in how quantum dots actually work.
Is OLED good for eyes? What about the PWM flicker concern?
The blue light output and flicker concerns raised about OLED are partially valid and partially overstated. Many OLED panels use pulse-width modulation (PWM) to dim the display — rapidly switching pixels on and off rather than reducing drive current, because organic emitters’ color point can shift slightly at very low current. At high brightness, duty cycles are long and flicker frequency is high, which is not perceptible. At low brightness, some panels use lower PWM frequencies that are perceptible to a minority of users, particularly in peripheral vision. High-PWM-frequency panels and DC dimming panels (which use current reduction rather than switching) exist and are noted in display reviews for users who are sensitive to flicker. Neither OLED nor LCD has a definitively better eye safety profile across all operating conditions; both emit the same visible and near-visible spectrum from comparable ambient levels.
Can OLED be made flexible or transparent?
Yes, and this follows directly from the physics. The organic thin films are genuinely thin — the active stack is on the order of a few hundred nanometers — and they can be deposited on flexible polymer substrates as well as glass. Foldable smartphones use flexible OLED panels laminated over a hinge mechanism. Transparent OLED is possible by replacing the opaque metal cathode with a thin semi-transparent electrode, allowing light to pass through both the front and back of the panel. Transparent OLED is used in commercial applications such as retail display windows, though transparency comes at the cost of some brightness and contrast since ambient light passes through and competes with the emitted image.
How does OLED compare with microLED for the future of displays?
MicroLED and OLED are both per-pixel emission technologies, so they share the fundamental advantage of true blacks and infinite contrast. The key differences are longevity (inorganic GaN microLED emitters degrade far more slowly than organic emitters), efficiency (inorganic LEDs are more efficient at high brightness), and manufacturing cost (OLED is a mature thin-film deposition process; microLED requires precise mass transfer of millions of individual chips). In mid-2026, consumer microLED products exist but at significantly higher cost than OLED. For most purchasing decisions in the near term, the relevant comparison is between OLED variants; microLED will likely become competitive in the consumer segment within this decade but is not there yet for mainstream use.
Further Reading
From this site:
– How Quantum Dots Actually Work: The Physics Explained — the nanocrystal physics behind QD-OLED’s color conversion layer
– How CMOS Image Sensors Work: The Physics Explained — per-pixel active-matrix signal acquisition, an instructive parallel to AMOLED pixel drive architecture
– How Silicon Photonics Chips Work: Moving Data with Light — photon generation and detection at chip scale
External sources:
– Society for Information Display (SID) — Display Week Proceedings — primary research literature on OLED device physics and backplane engineering
– IEEE Spectrum — Organic LEDs — peer-reviewed coverage of OLED materials advances and efficiency milestones
– FlatpanelsHD — QD-OLED 2026 panel specification reporting — independent hardware journalism on Samsung Display’s 2026 tandem stack specifications
