Why Is Glass Transparent? The Physics of Light Passing Through

A window passes roughly nine-tenths of the visible light that reaches it, yet the sand it is made from is an opaque heap. Nothing about the chemistry changes when sand is melted, so the explanation has to lie in how the atoms are arranged and what energy light carries. Why is glass transparent? Three conditions must hold at once: visible photons cannot be absorbed, almost none are scattered sideways, and the surfaces and impurities steal only a few percent. This article walks through each condition and shows where it fails.

What transparency requires

When light enters a solid, only a handful of things can happen to it. It can be absorbed, so its energy turns into heat or excites the material. It can be scattered, so it leaves in a new direction. It can be reflected at a boundary. Or it can be transmitted, continuing in roughly the same direction. A material looks transparent when transmission dominates and the light that exits is still organized enough to form an image.

That last part matters. Frosted glass and a sheet of wax paper both let light through, but they scramble its direction, so you see brightness without a picture. Transparency is therefore two separate questions: how much light survives, and how well its direction is preserved. Glass scores well on both, for different reasons.

Each reason maps to a physical property. Absorption is governed by the electron energy levels of the material, which is the band gap story. Scattering is governed by how uniform the material is on a scale comparable to the wavelength of light, roughly 400 to 700 nanometers (nm). Reflection is governed by the refractive index, which sets how sharply the medium differs from air.

Why sand is opaque and glass is not

A common explanation says sand is opaque because it is polycrystalline, and glass is transparent because it is amorphous. The second half is right. The first half needs a correction, because a single grain of beach sand is usually a small crystal of quartz, and a clean quartz crystal is itself transparent. Rock crystal, the clear form of quartz, has been carved into lenses and jewelry for centuries.

What makes a pile of sand opaque is the pile. Light entering a grain meets a surface, where the refractive index jumps from about 1 (air) to about 1.54 (quartz). Some light reflects, the rest bends, and then it reaches the next grain and does the same thing again. After dozens of such interfaces in random orientations, no direction is preserved, and the light has either been scattered back out or absorbed by the tinted films and impurities that coat real grains.

The same thing happens with other transparent materials. Snow is white even though ice is clear, because snow is a pile of tiny ice crystals with air between them. Crushed window glass is white for the same reason. A bag of sugar looks white while a rock-candy crystal looks clear. The material has not changed. The number of interfaces has.

Pure silica glass and a single quartz crystal have the same chemical formula, SiO₂, and are both transparent. A powder of either one looks white, because every grain surface scatters light.

Solid glass is a different case. It is a single continuous piece with no internal grain surfaces to scatter light. The Wikipedia entry on glass notes that glass is often transparent because it lacks the grain boundaries that diffusely scatter light in polycrystalline materials (Wikipedia: Glass). Polycrystalline ceramics show the opposite behavior: they are usually opaque because grain boundaries and residual pores act as scattering centers when their size is comparable to the wavelength of visible light (Wikipedia: Transparent ceramics).

For readers who want to know what an amorphous solid actually is, why glass is not a normal solid covers the glass transition and the old myth about flowing windows. Here the relevant point is narrower: a melt that is cooled fast enough to skip crystallization ends up with disordered atoms but no internal boundaries.

The band gap: why visible photons are not absorbed

Electrons in an atom sit at specific energy levels, and when atoms bond into a solid those levels broaden into bands. In an insulator like silica, the highest filled band is separated from the next empty band by a band gap, an energy range with no available electron states. A photon can be absorbed by exciting an electron across that gap only if the photon carries at least the gap energy. Anything less passes by as if the electron were not there.

For silicon dioxide the gap is about 9 electron volts (eV), a value listed in the band-gap table on Wikipedia (Wikipedia: Band gap). An electron volt is the energy an electron gains crossing a potential difference of one volt, a convenient unit for photon energies. Visible light is far below that threshold, as the table shows.

Color	Wavelength (nm)	Photon energy (eV)
Violet	380 to 450	2.75 to 3.26
Blue	450 to 485	2.56 to 2.75
Green	500 to 565	2.19 to 2.48
Yellow	565 to 590	2.10 to 2.19
Orange	590 to 625	1.98 to 2.10
Red	625 to 750	1.65 to 1.98

Values from the color table in Wikipedia: Visible spectrum. Cyan (485 to 500 nm, 2.48 to 2.56 eV) is omitted for brevity.

The most energetic visible photon carries about 3.26 eV, roughly a third of what silica needs. A useful analogy is a staircase with a missing flight. An electron standing on the floor can only climb if it receives exactly enough energy to reach the next occupied level, and in glass that level is a very long way up. Photons that cannot supply the climb are ignored.

Converting the gap to a wavelength shows where silica would start to absorb. Photon energy in eV equals about 1,240 divided by the wavelength in nm, so 9 eV corresponds to roughly 138 nm. That is deep in the vacuum ultraviolet, far beyond what the eye can see. Real fused silica starts losing transmission slightly before that: a 1 cm piece of UV-grade fused quartz transmits about 50% at 170 nm, and its overall cutoff is near 160 nm (Wikipedia: Fused quartz).

Why ordinary window glass blocks more ultraviolet than fused silica

You may have read that glass is opaque to ultraviolet light because UV photons have enough energy to cross the band gap. For pure silica that is only true far into the vacuum ultraviolet. The ultraviolet cutoff of window glass sits at much longer wavelengths, and the reason is not silica itself. Soda-lime glass contains added sodium and calcium oxides plus trace impurities, and such additives and impurities can add absorption at shorter wavelengths that pure silica lacks.

Fused silica, which is almost pure SiO₂ with no added fluxes, is therefore used for ultraviolet optics, while ordinary glass is not. The Wikipedia entry on fused quartz notes that aluminum and titanium impurities affect UV transmission, and that hydroxyl groups from water reduce infrared transmission (Wikipedia: Fused quartz).

Why metals are different

Metals have no gap. Their outer electrons occupy a partly filled band and can absorb photons of nearly any energy, then re-emit them. That process reflects light strongly at the surface instead of letting it pass, which is why metals look shiny and are opaque in all but extremely thin films. The absence of a gap is the reason a mirror works, a topic covered in how mirrors work.

What the electrons do instead: slowing light down

If the electrons do not absorb visible light, they still respond to it. The electric field of a light wave pushes on the electron cloud around each atom, making it oscillate at the light's frequency. Those oscillating charges radiate their own waves, and the sum of the incoming wave and all the re-radiated waves is a new wave that travels more slowly through the material than through vacuum.

The ratio of the speed of light in vacuum to its speed in the material is the refractive index, written n. For flat soda-lime glass n is about 1.520, and for fused silica it is 1.4585 at 587.6 nm (Wikipedia: Soda-lime glass; Wikipedia: Fused quartz). Light in window glass therefore travels at about 1/1.52, or 66%, of its vacuum speed. This is a delay in the wave, not a collision.

The refractive index depends slightly on wavelength, so blue light slows a bit more than red. That variation, called dispersion, is why a prism fans white light into a spectrum, a story told in Newton's prism experiments. Fused silica has a high Abbe number of 67.82, which means low dispersion, one reason it is prized for lenses (Wikipedia: Fused quartz).

For a deeper look at what light is, what is a photon explains the particle picture and why photon energy is tied to wavelength.

Why light does not scatter sideways

Slowing light is not the same as scattering it. In a perfectly uniform medium, the re-radiated waves from billions of atoms interfere destructively in every direction except forward. The sideways contributions cancel, and only the forward beam survives. Scattering appears when something breaks the uniformity: a particle, a bubble, a crack, a boundary, or a region where the density differs from its neighbors.

The size of the irregularity decides how strong the scattering is. Particles much smaller than the wavelength of light produce Rayleigh scattering, whose strength falls off as the fourth power of the wavelength, so blue light scatters far more than red (Wikipedia: Rayleigh scattering). That is why the daytime sky is blue, as covered in why the sky is blue, and why Tyndall's experiments with suspended particles produced a blue haze (Tyndall's blue-sky experiment). Irregularities comparable to or larger than the wavelength scatter all colors, which is why clouds, milk, and fog look white.

Glass still scatters a little, and the cause is frozen-in disorder

Even perfect glass is not perfectly uniform. As molten silica cools through the glass transition, its atoms stop rearranging, and whatever small density fluctuations exist at that moment are locked in. The temperature at which the structure freezes is called the fictive temperature, and a glass frozen at a lower temperature contains smaller density fluctuations.

The Wikipedia article on optical fiber describes the fictive temperature as the key variable controlling scattering from density fluctuations, so that a lower fictive temperature gives a more homogeneous glass with less Rayleigh scattering (Wikipedia: Optical fiber). In a window you never notice this, because the path is only millimeters long. In a fiber carrying light for tens of kilometers it becomes the main limit.

A useful analogy is a lake and a rough sea. Calm water carries a reflection almost undisturbed, while choppy water scatters it in every direction. Glass is calm water that froze in place, with only tiny ripples left.

Where a window loses light: reflection at the surfaces

Even with no absorption and no scattering, a pane of glass cannot transmit everything, because light is partly reflected whenever the refractive index changes. For light hitting a boundary head on, the fraction reflected is R = ((n₁ − n₂)/(n₁ + n₂))². For air (n = 1.0) meeting glass with n = 1.5, that works out to (0.5/2.5)² = 0.04, or about 4% (Wikipedia: Fresnel equations).

A window has two surfaces, so the simple estimate is about 8% lost to reflection. A more careful calculation that counts light bouncing back and forth between the two surfaces gives a total transmission of (1 − R)/(1 + R) for a lossless pane. The results below are calculated from the refractive indices quoted above and ignore absorption.

Material	Refractive index	Reflection per surface	Maximum transmission of a lossless pane
Fused silica	1.4585	about 3.5%	about 93.3%
Flat soda-lime glass	1.520	about 4.3%	about 91.8%
Quartz crystal (ordinary ray)	1.544	about 4.6%	about 91.3%

That ceiling explains why window glass is often described as passing about 90% of visible light: roughly 92% is the reflection limit, and a few more percent are lost to absorption in the bulk. Treat the 90% figure as a typical value for clear glass, not a constant. Thickness, iron content, and coatings all change it.

Anti-reflection coatings undo the loss

A coating can cancel the reflection by using interference. A layer one quarter of a wavelength thick, with an index between air and glass, makes the reflections from its top and bottom surfaces arrive half a wavelength apart, so they cancel. Magnesium fluoride (n about 1.38) is the classic single layer, lowering a crown glass surface from about 4% reflectance to about 1%, and multilayer stacks can reach 0.1% at one wavelength (Wikipedia: Anti-reflective coating).

The history is oddly accidental. In 1886 Lord Rayleigh noticed that old, tarnished glass transmitted more light than clean new glass, because the tarnish layer had an index between air and glass. Harold Dennis Taylor developed a chemical method in 1904, and Olexander Smakula at Carl Zeiss invented interference coatings in 1935 (Wikipedia: Anti-reflective coating). The same thin-film interference makes colors on soap films, explained in why soap bubbles show rainbow colors.

Where glass loses light inside: absorption and the green edge

Look at the edge of a thick sheet of ordinary glass and it appears green. The cause is iron. Flat soda-lime glass is typically about 73% silica, 14% sodium oxide, and 9% calcium oxide, with iron oxide at around 0.1% by weight in flat glass and about 0.04% in container glass (Wikipedia: Soda-lime glass). Iron is present in the sand and other raw materials, and it absorbs light.

The Wikipedia entry on glass explains that iron(II) oxide gives a green tint and iron(III) oxide gives yellow or yellow-brown (Wikipedia: Glass). The Wikipedia entry on float glass adds that ordinary float glass looks green in thicker sheets because of ferrous iron impurities (Wikipedia: Float glass). Absorption follows the Beer-Lambert law: each extra millimeter removes a constant fraction of the light, so the tint that is invisible in a 3 mm pane becomes obvious across a 30 cm glass block seen edge on.

Glassmakers lower the iron content to make clearer glass. Low-iron glass is used in aquariums, display cases, and solar panel covers, where the green cast or the lost percent matters. Another trick is to add a little manganese dioxide, which gives a purple tint that partly cancels the green (Wikipedia: Glass).

How impurities make colored glass

Color in glass is a deliberate version of the same absorption. Ions of certain metals have electron levels that sit inside the otherwise empty gap, and photons with exactly the right energy are absorbed, while the rest pass through. The color you see is what is left after the missing wavelengths are removed.

Additive	Effect on color	Notes
Iron(II) oxide	Green	Common impurity in raw sand
Iron(III) oxide	Yellow to yellow-brown	Form depends on furnace chemistry
Cobalt oxide (0.025 to 0.1%)	Deep blue	Very small amounts are enough
Copper(II) oxide	Turquoise
Manganese dioxide	Purple	Can cancel iron's green
Cadmium sulfide	Red
Sulfur with carbon and iron salts	Amber

Source: Wikipedia: Glass.

Colored glass is still transparent in the sense that matters here. The material stays amorphous and unscattering, and it simply has a narrower transmission window. Depth matters, again through the Beer-Lambert law, so a thin cobalt glass looks pale blue while a thick one looks nearly black.

Gold ruby glass works differently

Not every color comes from dissolved ions. Gold ruby glass, also called cranberry glass, is red because tiny particles of colloidal gold are dispersed through it. Richard Zsigmondy, who won the Nobel Prize in Chemistry in 1925, explained that small colloids of gold are responsible for the red color (Wikipedia: Cranberry glass). Here the color comes from how conduction electrons in each nanoparticle respond to light, not from an energy level in the glass.

The most famous example is the Lycurgus Cup, a 4th-century CE Roman glass cage cup in the British Museum. It contains gold and silver particles about 70 nm across and appears green when lit from the front and red when light passes through it (Wikipedia: Lycurgus Cup). The Roman craftsmen could not have known the mechanism, and neither could the 17th-century glassmaker, Johann Kunckel or Antonio Neri depending on the account, who rediscovered the technique (Wikipedia: Cranberry glass).

The nanoparticles are well below the wavelength of light in size, which is why the cup is still a clear object and not a milky one. They absorb and scatter selectively, yet they are too small and too sparse to turn the glass into a fog. That difference between a few tiny particles and many large ones is the same boundary that separates a clear sky from a cloud.

How sand becomes clear: melting, cooling, and floating

Glassmaking has a long history. Wikipedia reports that the earliest known glass objects, mostly beads, date to the mid-third millennium BC, and that sustained glass production appeared around 1600 BC in Mesopotamia and around 1500 BC in Egypt (Wikipedia: Glass). That puts routine glassmaking at more than 3,500 years old.

Modern window glass is almost all soda-lime glass, which Wikipedia says accounts for about 90% of manufactured glass (Wikipedia: Soda-lime glass). Pure silica melts and softens at extreme temperatures. Its glass transition is near 1,200 °C, and fused quartz melts at roughly 2,200 °C (Wikipedia: Silicon dioxide; Wikipedia: Fused quartz). Adding sodium oxide lowers the working temperature a great deal, with soda-lime glass starting to soften at as low as 700 °C and showing a glass transition near 564 to 573 °C, though it also disrupts the silica network and adds absorbing impurities (Wikipedia: Soda-lime glass).

The float process

Flat, distortion-free window glass is made by the float process, developed at Pilkington Brothers by Sir Alastair Pilkington and Kenneth Bickerstaff between 1953 and 1957. Near-perfect float glass became widespread in the 1960s (Wikipedia: Float glass; Wikipedia: Glass). Raw materials are melted at about 1,500 °C, then flow onto a bath of molten tin about 3 to 4 m wide, 50 m long, and 6 cm deep, where they spread into a flat ribbon under their own weight (Wikipedia: Float glass).

The glass cools from about 1,100 °C on the bath to about 600 °C before it is lifted onto rollers, then passes through an annealing kiln about 100 m long to relieve stress before cutting (Wikipedia: Float glass). The molten tin surface is nearly perfectly flat, so both faces of the sheet come out optically flat. Older windows, which were cast or blown, have waves and bubbles that you can see as a rippled view. Those are defects of the old process, not of glass as a material.

The cooling step has a transparency role of its own. Cooling too slowly invites crystallization, which creates grains that scatter light and leave a cloudy product called devitrification. Cooling unevenly leaves stress, which does not block light but weakens the sheet and can change how it refracts. Annealing is the compromise: fast enough to stay amorphous, controlled enough to relax stress.

When glass stops being clear

Glass is transparent only as long as nothing breaks its uniformity or adds absorbers. Several familiar glass products work by deliberately breaking one of the conditions.

• Frosted glass. Sandblasting or acid etching roughens the surface. The glass underneath is unchanged, but the rough surface refracts light in random directions at every tiny facet. Wet the frosted side with water and it turns clearer, because water (n about 1.33) fills the pits and reduces the index contrast.
• Opal and milk glass. These contain dispersed particles or droplets of a second phase, such as tiny crystals or gas pockets, whose refractive index differs from the surrounding glass. Light bounces off each one, so the glass looks white.
• Glass wool. Each fiber is clear, but a mass of fibers with air between them is a pile of interfaces, just like sand or snow.
• Bubbles and seeds. Trapped gas bubbles act as scatterers. The Wikipedia entry on fused quartz notes that trapped air bubbles can make the material translucent (Wikipedia: Fused quartz).
• Devitrified glass. If the melt partly crystallizes, grain boundaries return and the glass turns cloudy.

Tempered glass, which is cooled quickly to place the surface under compression, stays transparent. Stress does not create scattering centers, so the pane keeps its clarity while gaining strength. The tradeoff is that stress changes the polarization of transmitted light slightly, which is why you can sometimes see faint patterns in a car window through polarized sunglasses.

The clearest glass ever made: optical fiber

The purest transparency belongs not to windows but to optical fiber. In 1965 Charles Kao and George Hockham at Standard Telephones and Cables proposed that losses below 20 dB/km were achievable if impurities were removed from silica glass. In 1970 a team at Corning Glass Works, led by Robert Maurer, Donald Keck, and Peter Schultz, reached the threshold, producing a fiber with 17 dB/km loss. Kao later shared the 2009 Nobel Prize in Physics for this work (Wikipedia: Optical fiber).

A decibel is a logarithmic unit of power loss, where 10 dB is a factor of ten, so 20 dB/km means only 1% of the light survives one kilometer. Commercial low-loss fiber today reaches about 0.148 dB/km at a wavelength of 1,550 nm in the infrared, according to the same source. Over 100 km that totals 14.8 dB of loss, which leaves about 3% of the input power. That is why optical amplifiers can span oceans.

A record-low-loss fiber analysis cited by Wikipedia breaks down the remaining loss at 1,550 nm into 0.12 dB/km from Rayleigh scattering, 0.015 dB/km from infrared absorption, 0.0047 dB/km from impurity absorption, and 0.001 dB/km from waveguide imperfection (Wikipedia: Optical fiber). In other words, the best glass is limited mostly by the frozen-in density fluctuations described earlier, not by dirt. The story of how this technology works is told in how fiber optic cables transmit data and the history of the fiber optic cable.

Fiber operates in the infrared on purpose. Rayleigh scattering weakens as the wavelength grows, and silica's infrared absorption does not become serious until still longer wavelengths, so 1,550 nm sits in a sweet spot where both losses are small.

Other routes to transparency

Glass is not the only way to get a clear solid. Single crystals such as sapphire and quartz are transparent because a perfect crystal has no internal boundaries. Wikipedia reports that polycrystalline ceramics can also be made transparent if their grains and pores are much smaller than the wavelength of light, which became practical in the early 2000s with nanopowder processing. Examples include aluminum oxynitride, magnesium aluminate spinel, and yttrium aluminum garnet (Wikipedia: Transparent ceramics). Aluminum oxynitride and spinel are major candidates for transparent armor.

Plastics follow the same rules. A semicrystalline polymer contains small crystalline regions embedded in amorphous material, and those regions can scatter light, so polyethylene film looks hazy. Fully amorphous polymers such as acrylic lack that order and can be very clear. In every case, the question is the same two-part test: does the material absorb visible light, and is it uniform on the scale of the wavelength?

Diamond is an interesting comparison. Its band gap is 5.5 eV, according to the same band-gap table on Wikipedia (Wikipedia: Band gap). That is smaller than silica's gap but still well above visible photon energies, so a flawless diamond is transparent, and its sparkle comes from a high refractive index and careful cutting rather than from any difference in the absorption story.

Common confusions about glass transparency

"Light passes through the empty space between atoms." Atoms are mostly empty space, but light is a wave interacting with the electron clouds, not a particle threading through gaps. Metals are also mostly empty space and are opaque. What counts is whether the electrons can absorb or reflect the light.

"Glass is a slow-flowing liquid." The flowing windows story has been refuted. Glass is an amorphous solid, and the waviness in old panes comes from how they were made. This is covered in why glass is not a normal solid.

"Transparent means nothing happens to the light." Light slows down, partly reflects, and is slightly scattered and absorbed in even the best glass. Transparent means those effects are small enough to ignore for a given purpose.

"A thicker pane is less transparent because of scattering." In clean glass, the extra loss comes mostly from absorption, and in colored glass it can be dramatic. A thick block of clear glass looks green at the edge because iron absorbs a bit more of the red end of the spectrum over a long path.

"Perfect transparency is possible with the right glass." Reflection at a boundary is unavoidable whenever the index changes, and scattering from frozen-in density fluctuations has a floor set by the glass transition. Coatings can reduce reflection to a fraction of a percent at chosen wavelengths, but never to exactly zero across all colors and angles.

Frequently asked questions

Q: Why is glass transparent but sand is not? A: A single grain of sand is often a clear quartz crystal. A pile of sand is opaque because light meets many grain surfaces, reflecting and refracting at each one until no direction survives. Melted and cooled into a solid piece, the same silica has no internal surfaces, so light passes through.

Q: Why can't visible light excite electrons in glass? A: Silicon dioxide has a band gap of about 9 eV, while visible photons carry only about 1.65 to 3.26 eV. Photons without enough energy to lift an electron across the gap are not absorbed, so they continue through the material.

Q: Why does glass look green at the edge? A: Iron impurities in the raw materials absorb some light, and iron(II) gives a green tint. Flat soda-lime glass contains around 0.1% iron oxide by weight. In a thin pane the absorption is small, but along the edge the light travels through a much longer path and the tint becomes visible.

Q: Does glass block ultraviolet light? A: Ordinary window glass transmits less ultraviolet than fused silica because of its additives and impurities, while pure fused silica transmits ultraviolet light down to roughly 160 nm. The ultraviolet cutoff therefore depends on the glass composition, not on silica alone.

Q: Why does frosted glass look cloudy? A: The surface is roughened by etching or sandblasting, which scatters light in many directions. The bulk glass is unchanged. Adding water to the surface fills the roughness and makes the glass look clearer, because the index difference at each facet shrinks.

Q: How much light does a window pass? A: About 4% of light reflects at each surface for an index of 1.5, so a lossless pane could pass roughly 92%. Absorption in the bulk lowers that by a few more percent. Anti-reflection coatings can raise it by reducing surface reflection.

Q: What is the most transparent material made by people? A: Silica optical fiber is the benchmark. Commercial low-loss fiber loses about 0.148 dB/km at 1,550 nm, and the rest of the loss is dominated by Rayleigh scattering from density fluctuations frozen into the glass.

The takeaway

Glass is transparent because three separate conditions line up. Its wide band gap means visible photons cannot be absorbed by electrons, its disordered but uniform structure leaves no grain boundaries to scatter light, and careful refining keeps absorbing impurities such as iron to a few hundredths of a percent. What remains are small, predictable losses: about 4% reflection at each surface, a faint green edge, and a tiny amount of scattering that matters only over kilometers.

The idea that ties it together is that transparency is a statement about scale. Light responds to structure at the scale of its own wavelength, and to energy levels at the scale of its own photon energy. A material that is smooth at 500 nm and silent at 2 eV will let light through, whether it is a window, a lens, a fiber, or a carefully sintered ceramic. Take either away, by roughening the surface, grinding it to powder, or adding the right ion, and the clarity goes with it.

Sources

• Wikipedia: Glass, history, composition, colorants, grain boundaries and transparency.
• Wikipedia: Soda-lime glass, composition, refractive index, softening range, share of production.
• Wikipedia: Fused quartz, refractive index, Abbe number, ultraviolet cutoff, impurities.
• Wikipedia: Fresnel equations, normal-incidence reflectance.
• Wikipedia: Anti-reflective coating, quarter-wave coatings and history.
• Wikipedia: Optical fiber, attenuation, Kao and Corning milestones, Rayleigh scattering.
• Wikipedia: Band gap and Wikipedia: Visible spectrum, band gaps and photon energies.
• Wikipedia: Float glass, the Pilkington process.
• Wikipedia: Transparent ceramics, grain boundaries and pores in polycrystals.
• Penn State MATSE 81: Why is glass transparent?, the structure and energy-level explanation.

---

Read next:

• Why Is the Sky Blue? The Physics Explained
• How Mirrors Work: The Physics of Reflection
• How Electricity Actually Works