Silicon dioxide melts at roughly 1,713 °C (about 3,115 °F), far above most common materials, because it is not built from individual molecules the way water or carbon dioxide is. Instead, every silicon atom bonds covalently to four oxygen atoms, and every oxygen atom bridges two silicon atoms, creating a continuous three-dimensional framework that extends in all directions. To melt SiO2 you have to break enormous numbers of these strong covalent bonds simultaneously, and that requires a tremendous amount of thermal energy.
A Network, Not a Molecule
Most substances people encounter in everyday life are made of discrete molecules held together by relatively weak intermolecular attractions. Wax, alcohol, and dry ice are all like this: their individual molecules are sturdy enough, but the forces keeping one molecule next to the next are easy to overcome with modest heat. Silicon dioxide works on a completely different principle. Each silicon atom sits at the center of a tetrahedron of four oxygen atoms, and each of those oxygen atoms is shared with an adjacent silicon atom. The result is a vast, repeating lattice where covalent bonds stretch unbroken from one side of a crystal to the other. There are no isolated SiO2 “molecules” in the solid, just one enormous bonded structure.
Covalent bonds between silicon and oxygen are individually strong, with bond energies on the order of 450 kJ per mole. Because each silicon has four of them and each oxygen has two, the network is dense with bonds in every direction. Melting means disrupting enough of these bonds that atoms can slide past one another and flow. Compared to overcoming the weak van der Waals forces that hold molecular solids together, or even the ion-ion attractions in many salts, breaking a web of covalent bonds demands far more energy per unit of material.
Why Carbon Dioxide Is a Gas but Silicon Dioxide Is a Rock
Carbon and silicon sit in the same column of the periodic table, and both form dioxides. Yet CO2 sublimes at −78 °C while SiO2 stays solid until well above 1,700 °C. The reason comes down to size and bonding style. Carbon is small enough that it can form strong double bonds with oxygen, producing compact O=C=O molecules. These molecules are self-contained: once you have an O=C=O unit, there is no leftover bonding capacity reaching out to neighboring molecules. The only thing holding one CO2 molecule near another is the feeble van der Waals attraction between them, so dry ice evaporates easily.
Silicon atoms are larger, and their orbitals do not overlap efficiently enough with oxygen to form the same kind of stable double bonds. Instead, silicon satisfies its bonding needs by linking to four separate oxygen atoms through single bonds, and each oxygen bridges to the next silicon. The geometry forces a three-dimensional lattice rather than a tidy, self-contained molecule. This single structural difference, four bridging single bonds versus two terminal double bonds, is what separates a refractory ceramic from a gas you can buy at a grocery store for carbonating water.
The Many Faces of Solid Silica
SiO2 does not come in just one crystal form. Depending on the temperature and pressure at which it crystallizes, silica adopts several distinct arrangements known as polymorphs. At atmospheric pressure the familiar ones are quartz (stable at lower temperatures), tridymite (intermediate temperatures), and cristobalite (the highest-temperature low-pressure form). All three are built from the same SiO4 tetrahedra, but they differ in how those tetrahedra are oriented and connected.
Because the basic bond strength is similar across all three forms, their melting points cluster in the same general range, though not at exactly the same value. Calorimetry work covering temperatures up to 1,800 K measured the enthalpies of fusion at about 8.9 kJ per mole for cristobalite (melting near 1,999 K, or roughly 1,726 °C) and about 9.4 kJ per mole for quartz at around 1,700 K.1Geochimica et Cosmochimica Acta. Thermodynamic properties of quartz, cristobalite and amorphous SiO2: drop calorimetry measurements between 1000 and 1800 K and a review from 0 to 2000 K Those enthalpy values are surprisingly modest for a material with such a high melting point, reflecting the fact that melting does not require breaking every bond in the network. It only requires loosening enough of them that the rigid lattice collapses into a viscous liquid.
Astronomical observations have identified tridymite and cristobalite in the dusty disks surrounding young stars, where transient heating events push temperatures to 1,200–1,300 K and rapid cooling locks in these high-temperature crystal forms.2arXiv. Silica in Protoplanetary Disks That silica crystals survive and remain identifiable in such extreme environments is a testament to how thermally robust the SiO4 framework is.
What Happens When Silica Finally Melts
Melting silica does not produce a thin, runny liquid. Instead, it yields an extraordinarily viscous melt, so thick that it creeps rather than pours. The reason is that even in the liquid state, silicon and oxygen maintain much of their tetrahedral bonding. The network breaks and re-forms on short timescales, but at any given instant most atoms are still connected to neighbors through covalent bonds. This gives molten silica a syrupy, almost glassy character.
Measurements of silica’s viscosity at high temperature show Arrhenius-type behavior with an activation energy near 515 kJ per mole, meaning the liquid’s resistance to flow drops in a predictable, exponential way as temperature rises.3ScienceDirect (Elsevier). Viscosity-temperature relation based on the evolution of medium-range structures of silica That activation energy is high compared to most liquids, consistent with the idea that atoms in molten silica must break and remake strong covalent bonds just to move past one another. Even several hundred degrees above the melting point, pure silica melt is orders of magnitude more viscous than, say, molten iron.
If you cool molten silica quickly enough, it never finds time to rearrange into a crystal. Instead, the liquid freezes into an amorphous solid, the familiar material known as fused silica or silica glass. Calorimetry places the glass transition of amorphous SiO2 at roughly 1,480 K (about 1,207 °C), the temperature below which the disordered structure essentially locks into place.1Geochimica et Cosmochimica Acta. Thermodynamic properties of quartz, cristobalite and amorphous SiO2: drop calorimetry measurements between 1000 and 1800 K and a review from 0 to 2000 K Fused silica retains most of the thermal resilience of crystalline silica precisely because the same strong Si–O bonds persist in the amorphous state, just without long-range order.
How Impurities Dramatically Lower the Melting Point
Pure silica is remarkably hard to melt, but in practice most silica-based materials contain other elements, and those additions have an outsized effect on thermal behavior. When atoms like sodium, potassium, or calcium enter the silica network, they do not slot into the tetrahedral framework. Instead, they break bridging oxygen bonds, converting a bridging oxygen (one shared between two silicon atoms) into a non-bridging oxygen that hangs off just one silicon. Each broken bridge cuts a link in the continuous network.
As more of these network modifiers are added, the three-dimensional lattice progressively fragments into smaller chains and clusters. The viscosity plummets, and the melting point drops with it. Progressive addition of network modifiers leads to network depolymerization through the formation of non-bridging oxygen atoms; at a certain limit, the liquid becomes so depolymerized that its viscosity is extremely low and it can no longer even form a glass when quenched.4Comprehensive Nuclear Materials. Material Performance and Corrosion/Waste Materials This is exactly the principle behind soda-lime glass, the ordinary glass in windows and bottles. Pure silica melts above 1,700 °C, but adding sodium oxide and calcium oxide brings the working temperature down to around 1,000–1,100 °C, making mass production economically feasible.
Even small amounts of modifiers make a measurable difference. Research on the crystallization resistance of oxide glasses found that for strong glass formers like silica, the addition of even small amounts of modifiers rapidly reduces glass stability.5Journal of Non-Crystalline Solids. Effect of network formers and modifiers on the crystallization resistance of oxide glasses The network’s integrity is sensitive: you do not need to replace a large fraction of silicon atoms to meaningfully weaken the structure. This sensitivity is why geological magmas, which are silica-rich but laced with sodium, potassium, iron, magnesium, and aluminum, flow as liquids at temperatures hundreds of degrees below the melting point of pure quartz.
Silica Under Extreme Pressure
Everything discussed so far applies near atmospheric pressure, but deep inside Earth’s mantle, pressures reach tens of gigapascals and the rules change. Under those conditions, the familiar SiO4 tetrahedron is no longer the most stable geometry. Silicon atoms can accommodate more oxygen neighbors, shifting from four-fold to five-fold and even six-fold coordination.
The best-known high-pressure polymorph is stishovite, found naturally in meteor impact craters and synthesized in the lab. In stishovite, each silicon is surrounded by six oxygens in an octahedral arrangement rather than four in a tetrahedron. This denser packing makes stishovite significantly harder and denser than quartz, and it pushes the melting point even higher. Molecular dynamics studies show that when stishovite melts, silicon coordination drops from six back toward four, indicating that the melt reverts toward a more open tetrahedral arrangement even though the crystal was octahedral.6Geochimica et Cosmochimica Acta. Molecular dynamics of silica at high pressures: Equation of state, structure, and phase transitions Along the melting curve, the coordination number of silicon in the liquid increases with pressure, blurring the boundary between the ordered crystal and the disordered melt.
At pressures above roughly 30 GPa, researchers have documented even more exotic phases. Experiments compressing coesite (another silica polymorph) beyond 30 GPa produced a new triclinic phase called coesite-IV, containing a mixture of four-, five-, and six-coordinated silicon. Further compression above 50 GPa yielded coesite-V, which retains only five- and six-coordinated silicon in a framework of face- and edge-sharing octahedra.7Nature Communications. Metastable silica high pressure polymorphs as structural proxies of deep Earth silicate melts These high-pressure phases matter to geophysicists because they serve as structural proxies for what silicate melts look like deep in the mantle, where direct observation is impossible.
Even before full phase transitions occur, the low-pressure polymorphs undergo gradual structural changes under compression. Infrared spectroscopy of quartz, coesite, and stishovite under pressure shows that below about 10–20 GPa, compression mainly bends the Si–O–Si angles between tetrahedra. Above 20 GPa, the tetrahedra themselves distort so severely that silicon begins picking up additional oxygen neighbors, and the crystalline phases gradually become amorphous.8Journal of Geophysical Research: Solid Earth. High‐pressure infrared spectra of α‐quartz, coesite, stishovite and silica glass This pressure-induced amorphization is a kind of structural collapse, distinct from thermal melting, but it highlights how central the tetrahedral geometry is to silica’s identity. Anything that distorts or destroys the tetrahedra, whether heat or pressure, changes the material’s properties dramatically.
Why Defects Matter for Melting
In a perfect crystal, every atom sits in its designated lattice site and the melting point reflects the energy needed to shake the entire structure apart. Real crystals are never perfect. They contain vacancies (missing atoms), grain boundaries (where two crystal domains meet at an angle), and free surfaces. These defects are weak points in the network, and they can initiate melting at temperatures below what an idealized calculation would predict.
Molecular dynamics simulations of stishovite illustrate this clearly. Pre-existing defects such as grain boundaries, vacancies, and free surfaces initiate or facilitate both solid-state disordering and melting. For stishovite with low defect concentrations, melting precedes disordering, but defect-rich stishovite can transform into a high-density amorphous silica that then undergoes a continuous transition into melt.9Journal of Physics: Condensed Matter. Solid-state disordering and melting of silica stishovite: the role of defects In other words, a heavily flawed crystal does not melt in the sharp, clean way a perfect crystal would. Instead, it gradually loses order, first around the defects, and eventually throughout the bulk. This has practical consequences: finely ground silica powder, with its abundance of surface defects, can begin softening at temperatures slightly below the nominal melting point of a large, well-ordered crystal.
Silica’s Behavior in Geological and Planetary Melts
Most people encounter SiO2 in the form of sand, glass, or electronics-grade wafers, but the compound’s thermal properties shape entire planets. Earth’s crust and mantle are dominated by silicate minerals, compounds where SiO4 tetrahedra combine with metals like magnesium, iron, and aluminum. The viscosity and melting behavior of these silicates control how magma rises, how volcanic eruptions unfold, and how tectonic plates recycle material back into the mantle.
Pure silica magma, called rhyolitic melt when it dominates a volcanic system, is the most viscous type of lava. Its high silicon-oxygen connectivity traps dissolved gases and builds pressure, which is why silica-rich eruptions tend to be explosive. Basaltic magma, poorer in silica and richer in iron and magnesium, flows freely by comparison. The same mechanism that lowers the melting point when network modifiers are added to glass, breaking bridging oxygens, also explains why basalt erupts calmly while rhyolite explodes.
Research into diffusion in high-temperature silicate melts has revealed that even at temperatures above 1,000 °C, quantum-mechanical effects at the atomic level influence how different isotopes of silicon and oxygen move through the liquid.10Nature. The Soret effect and isotopic fractionation in high-temperature silicate melts These isotopic signatures, detected in natural lava samples and laboratory experiments, give geochemists a tool for reconstructing the thermal history of rocks. The fact that quantum effects matter at all at such extreme temperatures is a reminder that silica’s bonding environment is unusual: the lightweight atoms and strong covalent bonds create conditions where zero-point energy differences between isotopes remain significant even when everything is glowing white-hot.
How the Si–O–Si Angle Contributes to Strength
The overall architecture of silica depends not just on the strength of individual Si–O bonds but on the angles between linked tetrahedra. The Si–O–Si angle, the angle at each bridging oxygen, is typically around 144° in quartz but varies among polymorphs and under strain. This angle is flexible to a degree: it can bend without the bond itself breaking, which is how silica accommodates small stresses or compressions. Molecular dynamics simulations of silica under mechanical strain show that at modest deformation rates, the network adjusts by rotating tetrahedra to increase the Si–O–Si angle in the direction of the applied strain.11ScienceDirect (Elsevier). Molecular dynamics studies of brittle failure in silica: effect of thermal vibrations Only when deformation exceeds this rotational flexibility do bonds begin to snap and the material fractures.
This angular flexibility is part of why silica glass, despite being amorphous, is still so thermally robust. The disordered network can absorb thermal vibrations by rocking tetrahedra back and forth without breaking bonds. Each Si–O–Si linkage acts like a flexible hinge connecting two rigid units. You need enough thermal energy to overwhelm that hinge mechanism and actually stretch bonds to their breaking point before the material transitions to a liquid. That threshold is what sets the melting point so high.
The same hinge-like flexibility also explains a quirky property of silica glass: it has a very low coefficient of thermal expansion. When heated, the rocking motion of the tetrahedra absorbs much of the added energy without translating it into the outward push that makes most materials expand. This is why fused silica can survive sudden temperature changes that would shatter ordinary glass, and why it is the material of choice for telescope mirrors, laboratory glassware, and semiconductor fabrication equipment where dimensional stability matters.