Sulfur can form six bonds because it sits in the third row of the periodic table, where atoms have access to orbitals that allow bonding arrangements beyond the four-bond limit familiar from carbon and nitrogen. The classic textbook explanation credits sulfur’s empty 3d orbitals for this ability, but decades of computational chemistry have shown that picture is largely wrong. The real story involves a different kind of bonding that lets sulfur accommodate six partners while arguably never violating the octet rule at all. Understanding what actually happens in a molecule like sulfur hexafluoride reveals one of the more interesting gaps between what introductory chemistry teaches and what researchers have known for years.
What Textbooks Used to Say
For most of the twentieth century, the standard explanation went like this: elements in the second row of the periodic table (carbon, nitrogen, oxygen) have only s and p orbitals available, which max out at four bonds. Sulfur, sitting one row lower, has 3d orbitals that are close enough in energy to participate in bonding. By mixing its 3s, three 3p, and two 3d orbitals into a set of six equivalent hybrid orbitals (called sp³d² hybrids), sulfur can point a bond toward each of the six fluorine atoms in SF₆. This framework made the geometry easy to visualize and predict, and it explained why second-row elements like oxygen could not pull off the same trick.
The concept of “expanded octet” followed directly from this logic. If sulfur uses six bonds, each contributing two electrons, that puts twelve electrons in sulfur’s valence shell, well beyond the usual eight. The molecule was labeled “hypervalent,” meaning the central atom exceeded the octet. Linus Pauling’s hybridization framework, which proved remarkably useful for explaining molecular shapes and bond angles across organic chemistry, was extended to cover these cases too. Tools like natural bond orbital analysis continue to show that Pauling’s general concepts of hybridization and resonance hold up well in modern quantum chemistry, lending the old framework a veneer of respectability even as its specific application to hypervalent molecules was coming apart.
Why the d-Orbital Explanation Collapsed
Starting in the 1980s and accelerating through the 1990s, high-level quantum chemical calculations began telling a different story. When researchers computed how much the 3d orbitals actually contributed to bonding in molecules like SF₆, the answer was: not much. The d orbitals were too high in energy and too diffuse to play a significant bonding role. They acted more like mathematical “polarization functions” that improved the flexibility of the calculation rather than physical orbitals that housed bonding electrons. The sp³d and sp³d² hybridization schemes, while elegant on paper, did not match what the electrons were actually doing.
This finding has been confirmed repeatedly with increasingly powerful computational methods. The concept of an expanded octet has persisted in the general chemistry curriculum despite abundant theoretical work disputing its accuracy, as one analysis in the Journal of Chemical Education put it. The electronic structures of molecules traditionally called hypervalent, including sulfuric acid, phosphorus pentafluoride, and SF₆ itself, look quite different from what the expanded-octet model predicts when examined through quantum chemical calculations.
How the Bonds Actually Form
If d orbitals are not doing the heavy lifting, what holds SF₆ together? The leading modern explanation relies on a bonding arrangement called the three-center four-electron bond (often abbreviated 3c-4e). Instead of each sulfur-fluorine bond being its own independent two-electron affair, the bonds in SF₆ are organized in linear groups of three atoms: fluorine-sulfur-fluorine. Each of these linear units shares four electrons among the three atomic centers. Two of those four electrons sit in a bonding orbital spread across all three atoms, while the other two sit in a nonbonding orbital concentrated on the two fluorine ends. The sulfur atom in the middle ends up with less than a full pair of electrons from each of these units, which means its total electron count stays closer to eight than to twelve.
Research using ab initio calculations has shown that the structures and stabilities of hypervalent molecules with five or six electron pairs around the central atom can be satisfactorily explained by 3c-4e bonding combined with simpler hybridization schemes, without invoking d orbitals at all.1Journal of Molecular Structure: THEOCHEM. Application of three-center-four-electron bonding for structural and stability predictions of main group hypervalent molecules: the fulfillment of octet shell rule In SF₆, you can think of the molecule as having three perpendicular 3c-4e bonds, each running through the sulfur atom from one fluorine to the fluorine directly opposite it. The sulfur uses only its s and p orbitals, arranged in a way that points bonds along the x, y, and z axes. The 3c-4e model has become the standard framework among researchers who study these systems, and it is recognized as a central feature of hypervalent compound structure and reactivity.2Heteroatom Chemistry. Dynamic aspects of hypervalent compounds effected by the formation of three center‐four electron bond in heteroatoms
A Competing Explanation Worth Knowing
The 3c-4e model is not the only modern framework. A team studying the entire series of sulfur fluorides, from SF all the way up to SF₆, proposed a different mechanism called “recoupled pair bonding.” The idea is that sulfur has lone pairs of electrons that normally sit tight and do not participate in bonding. Under the right conditions, one of these pairs can be split apart, or “recoupled,” so that one of its electrons forms a new bond with an incoming fluorine atom while the other electron redistributes elsewhere in the molecule.
In this framework, adding fluorine atoms to sulfur alternates between two kinds of bonds. Some are ordinary strong polar covalent bonds, with bond-breaking energies in the range of roughly 90 to 110 kcal/mol. Others are weaker recoupled pair bonds, requiring only about 40 to 55 kcal/mol to break. This alternating pattern explains why each successive fluorine is not equally easy to add and why the bonding energies are not all the same.3ACS Publications (The Journal of Physical Chemistry A). Theory of Hypervalency: Recoupled Pair Bonding in SFn (n = 1−6) The recoupled pair bonding model and the 3c-4e model are not necessarily incompatible; they describe the same underlying quantum reality from different theoretical angles. Both agree that d orbitals are not the answer.
Why Fluorine Is Almost Always Involved
You might wonder: if sulfur can form six bonds, why do we almost always see it bonded to fluorine in these examples? The answer has everything to do with electronegativity. Fluorine is the most electronegative element on the periodic table, which means it pulls electron density away from sulfur very effectively. This matters because in the 3c-4e bonding model, the nonbonding electrons in each three-center unit are concentrated on the outer atoms. The more electronegative those outer atoms are, the more willing they are to hoard those electrons, and the more stable the whole arrangement becomes.
This is why SF₆ is extraordinarily stable and well-characterized experimentally, while a hypothetical molecule like H₆S (sulfur bonded to six hydrogen atoms) has never been observed. Hydrogen is far less electronegative than fluorine, so it cannot stabilize the charge distribution needed for six bonds around sulfur. Even mixed hydrogen-fluorine hypervalent sulfur species have never been experimentally detected.4PubMed Central. Electronic structure of H2S, SF2, and HSF and implications for hydrogen-substituted hypervalent sulfur fluorides SF₄, with four fluorines, is known but far less stable than SF₆. The pattern holds broadly: hypervalent molecules across the periodic table almost always feature highly electronegative ligands like fluorine, chlorine, or oxygen.
Is Sulfur Really “Expanding” Its Octet?
This is where the terminology gets contentious. Many chemists now argue that calling these molecules “hypervalent” or saying sulfur has an “expanded octet” is misleading, because in the 3c-4e bonding picture, sulfur never actually has more than eight electrons in its immediate vicinity. The extra electron density is parked on the fluorine atoms, not on sulfur. By this reading, SF₆ obeys the octet rule after all; it is just that three of its six “bonds” are delocalized multi-center bonds rather than simple two-center two-electron bonds.
Quantum chemical calculations support this interpretation. When researchers examined the electronic structure of SF₆ and similar molecules, they found that the hypervalent label is inaccurate.5Journal of Chemical Education. Addressing the Hypervalent Model: A Straightforward Explanation of Traditionally Hypervalent Molecules The practical geometry, an octahedron with six equivalent S–F bonds, is real. The idea that sulfur is somehow breaking the rules by cramming twelve electrons into its valence shell is not. Still, the expanded-octet language remains deeply embedded in textbooks, standardized exams, and introductory chemistry courses. It is one of those cases where the simplified model works well enough for predicting shapes and reactivities that educators keep teaching it, even though the underlying explanation is wrong.
The Sulfate Question
SF₆ is the showpiece, but sulfur also appears to exceed its octet in much more common molecules: sulfate ions (SO₄²⁻), sulfuric acid (H₂SO₄), and sulfur dioxide (SO₂). In textbook Lewis structures, sulfuric acid is often drawn with two S=O double bonds, which would put ten or twelve electrons around sulfur. Does the same bonding story apply?
Charge density studies of crystalline potassium sulfate paint a clear picture. The sulfur-oxygen bonds in sulfate are highly polarized covalent bonds, but the “single bond” description fits them far better than the “double bond” picture. The hypervalent description of sulfur in the sulfate group is effectively ruled out by the combined experimental and computational evidence.6PubMed. Testing the concept of hypervalency: charge density analysis of K2SO4 A separate quantum crystallography study confirmed that the S–O bonds are highly polarized and covalent, and that any increase in valence population from 3c-4e bonding is not sufficient to make the sulfur atom genuinely hypervalent.7PubMed. Revisiting a Historical Concept by Using Quantum Crystallography: Are Phosphate, Sulfate and Perchlorate Anions Hypervalent?
The takeaway: whether you are looking at SF₆ or sulfuric acid, the idea that sulfur has physically burst past the eight-electron limit does not hold up under scrutiny. What sulfur does have is a remarkable ability to distribute charge across its bonds, and to form bonding arrangements that are more complex than the simple two-center bonds of introductory chemistry. That flexibility, not a literal electron-count violation, is what makes sulfur’s chemistry so rich.
Sulfur’s Range of Oxidation States
The ability to form six bonds is just one manifestation of a broader trait: sulfur is unusually versatile in the number and type of bonds it can form. Unlike oxygen, which almost always sits in the –2 oxidation state, sulfur spans the entire range from –2 to +6. It shows up as a sulfide (–2) in amino acids like cysteine, as elemental sulfur (0) in volcanic deposits, as a sulfoxide (+2) or sulfone (+4) in pharmaceutical molecules, and as sulfate (+6) in minerals and metabolic intermediates. This range is made possible by the large, polarizable 3p electron shell, low-lying empty orbitals that can accept electron density through effects like negative hyperconjugation, and sulfur’s ability to stabilize multiple geometries.8Journal of Medicinal Chemistry. Turning the Sulfur Dial: Strategic Oxidation-State Manipulation in Modern Drug Discovery
Drug designers exploit this versatility constantly. By oxidizing a sulfur-containing drug from a thioether (sulfur bonded to two carbons) to a sulfoxide and then to a sulfone, chemists can systematically change the molecule’s shape, polarity, metabolic stability, and how it interacts with biological targets. Each oxidation state has distinct geometry and electronic properties, so the same basic molecular scaffold can be tuned for different purposes just by adjusting the sulfur. This “sulfur dial” concept has become a recognized strategy in modern drug discovery.
Why SF₆ Matters Outside the Lab
The extraordinary stability that makes SF₆ a fascinating bonding puzzle also makes it a serious environmental problem. SF₆ has been the standard insulating gas in high-voltage electrical equipment for decades, prized for its ability to quench electrical arcs and resist breakdown under extreme voltages.9PubMed. Theoretical Characterizations on the Eco-Friendly Gas Tetrafluoropropyne for Electrical Insulation to Replace Sulfur Hexafluoride It is also used in magnesium casting, semiconductor manufacturing, and even as a tracer gas in atmospheric studies. These applications depend on the same bonding features we have been discussing: six strong, symmetrical bonds that make the molecule chemically inert and thermally robust.
The flip side of that stability is that once SF₆ escapes into the atmosphere, virtually nothing breaks it down. Its atmospheric lifetime is estimated at roughly 3,200 years, and its 100-year global warming potential is about 23,800 times that of carbon dioxide.10Chemical Engineering Journal. Recent advances in degradation of the most potent industrial greenhouse gas sulfur hexafluoride Three-dimensional atmospheric modeling has produced broadly consistent numbers, with the 100-year global warming potential calculated at roughly 23,800 and the 500-year value rising to about 31,300.11Atmospheric Chemistry and Physics. Determination of the atmospheric lifetime and global warming potential of sulfur hexafluoride using a three-dimensional model On a per-molecule basis, SF₆ is the most potent greenhouse gas tracked by climate agencies.
Efforts to find replacements are ongoing. Fluoronitrile and fluoroketone gas mixtures have emerged as leading candidates for some high-voltage applications, and researchers are investigating compounds like tetrafluoropropyne as eco-friendly alternatives. But SF₆ is deeply entrenched in the global electrical grid, and retrofitting existing equipment is expensive and slow. The very bonding properties that make SF₆ scientifically elegant, its perfect octahedral symmetry, chemical inertness, and thermal stability, are the same properties that make it an atmospheric pollutant with a multi-millennial hangover.
Why Second-Row Elements Cannot Do This
A natural question is why oxygen, sitting directly above sulfur on the periodic table, never forms six bonds. The most important factor is size. Oxygen’s atomic radius is much smaller than sulfur’s, which means trying to pack six fluorine atoms around an oxygen center would create severe steric strain: the fluorines would physically crowd each other. Sulfur’s larger atomic radius gives its bonding partners enough room to arrange themselves into an octahedron without bumping into one another.
There is also an energetic factor. Forming the 3c-4e bonds that allow hypercoordination requires decoupling electron pairs that would normally stay inert. In sulfur, the energy cost of doing this is offset by the energy gained from forming new bonds to highly electronegative partners. In oxygen, the valence electrons are held more tightly and closer to the nucleus, making the decoupling process prohibitively expensive in energy terms. The result is a sharp divide between the second and third rows: carbon, nitrogen, and oxygen stick to four bonds (or fewer), while sulfur, phosphorus, and chlorine can extend to five or six under the right conditions.
This row-based divide is not absolute, though. Silicon, which sits below carbon, can form five- and six-coordinate compounds under special circumstances, and even some second-row species can be coaxed into arrangements that look superficially hypervalent if you define the terms loosely enough. But the practical chemistry is clear: the third row and below is where the action is, and sulfur is the most chemically versatile member of the club.