In its most familiar bonding situation, sulfur carries two lone pairs. That is the answer for hydrogen sulfide, for the sulfur in amino acids like cysteine, and for simple organic sulfur compounds. But sulfur is unusually flexible among the common nonmetals, and the number of lone pairs it holds shifts dramatically depending on how many bonds it forms. A sulfur atom bonded to two neighbors keeps two lone pairs; one bonded to four neighbors may have just one; and one bonded to six neighbors can end up with none at all.
Two Lone Pairs in Divalent Sulfur
When sulfur forms two bonds and nothing else, the arithmetic is straightforward. Sulfur has six valence electrons. Two of those go into two bonds, leaving four electrons, which sit as two lone pairs. This is the arrangement in hydrogen sulfide (H₂S), in thiols (the -SH group found on the amino acid cysteine), and in thioethers like dimethyl sulfide. If you have ever noticed the smell of rotten eggs or certain cooked vegetables, you have encountered divalent sulfur and its two lone pairs.
This two-lone-pair setup is often the version students encounter first, and for good reason: it mirrors the arrangement in water, where oxygen also has two lone pairs. The similarity is not a coincidence. Sulfur sits directly below oxygen on the periodic table, and both elements default to two bonds in their simplest compounds. The lone pairs give these molecules a bent shape, pushing the bonded atoms closer together than a straight line would allow.
When the Count Drops
Sulfur does not always stop at two bonds. It routinely forms three, four, five, or even six bonds in different compounds, and each additional bond eats into the supply of electrons available for lone pairs.
- Three bonds (e.g., sulfonium ions): Sulfur uses three of its electron pairs for bonding and retains one lone pair. The geometry around sulfur becomes roughly pyramidal, much like ammonia.
- Four bonds (e.g., sulfur dioxide, sulfoxides, SF₄): In sulfur dioxide, sulfur forms bonds to two oxygens plus a double bond, leaving one lone pair. In sulfur tetrafluoride, sulfur bonds to four fluorines and still holds one lone pair, which forces the molecule into a distinctive seesaw shape rather than a simple square.
- Six bonds (e.g., sulfur hexafluoride, SF₆): All six of sulfur’s valence electrons are involved in bonding. The count of lone pairs drops to zero, and the molecule is a perfect octahedron with sulfur at the center.
The progression is not unique to sulfur, but sulfur reaches higher bond counts more readily than oxygen does. Oxygen essentially never forms six bonds. Sulfur does so routinely, and the reasons are rooted in sulfur’s larger atomic size and the accessibility of orbitals beyond the usual four.
The Hypervalency Debate
That last point, sulfur forming more bonds than the classic octet rule would seem to allow, has sparked a long-running argument among chemists. The traditional explanation is that sulfur can expand its octet by using higher-energy d orbitals. But modern quantum-chemical calculations cast doubt on that picture. A 2019 study revisiting sulfate and related anions concluded that hypervalency “may be regarded as an artifact” of the way we draw Lewis structures, which assume all electrons are neatly localized in two-center bonds and lone pairs.1PubMed. Revisiting a Historical Concept by Using Quantum Crystallography: Are Phosphate, Sulfate and Perchlorate Anions Hypervalent? In other words, the “extra” bonds may be better understood as highly polar bonds with significant ionic character, rather than true covalent bonds requiring an expanded electron shell.
For practical purposes, this debate does not change the lone-pair count you would assign when predicting molecular geometry. Whether the bonding is “truly” hypervalent or not, the electron arrangement around sulfur in SF₆ still behaves as though there are six bonding pairs and zero lone pairs, and the molecule still takes an octahedral shape. But the debate matters if you are trying to understand where the electrons actually are, rather than just using a bookkeeping model.
How Lone Pairs Determine Shape
The reason anyone cares about lone pairs is that they take up space. Lone pairs repel bonding pairs, and that repulsion dictates the three-dimensional shape of the molecule. For divalent sulfur with two lone pairs, the bond angle in H₂S is about 92 degrees, noticeably tighter than the 104.5 degrees in water. The difference matters: it tells us the lone pairs on sulfur are squeezing the bonded hydrogens into a narrower wedge than oxygen’s lone pairs do. Sulfur’s larger size means its lone pairs sit farther from the bonds and exert a somewhat different repulsive geometry.
In sulfur tetrafluoride, the single remaining lone pair sits in the equatorial plane and pushes the four fluorines into a seesaw arrangement that is neither tetrahedral nor square planar. The lone pair effectively claims the most spacious position available. Removing that lone pair entirely, as in sulfur hexafluoride, restores full symmetry to the molecule. This progression from bent to seesaw to octahedral is a textbook demonstration of how lone-pair count sculpts molecular architecture.
Not All Lone Pairs on Sulfur Are Equal
Even when sulfur carries two lone pairs, those two pairs are not identical. Work on intramolecular hydrogen bonding in solution found that the lone pairs on divalent sulfur are best described as nonequivalent: one sits in an orbital with more s character (roughly a 3s orbital) and the other in an orbital with more p character (roughly a 3p orbital).2Canadian Journal of Chemistry. Concerning lone-pair stereospecificity of intramolecular OH hydrogen bonds to oxygen and sulfur in solution That study estimated the 3s lone pair forms a hydrogen bond at least 8 kJ/mol weaker than the 3p lone pair. In practical terms, the two lone pairs are not interchangeable: one is more accessible and more chemically reactive than the other.
This nonequivalence shows up again in radical chemistry. Computational studies of the cysteine thiyl radical found that the remaining lone-pair orbitals on the sulfur atom are “almost pure sulfur 3p orbitals,” and their energy separation from the singly occupied orbital is extremely small, on the order of a few thousand wavenumbers.3PubMed. Electronic structure of the cysteine thiyl radical: a DFT and correlated ab initio study That near-degeneracy makes sulfur radicals behave quite differently from oxygen or carbon radicals, which is one reason sulfur-centered radicals play such a distinctive role in biochemistry and materials science.
Why Hydrogen Bonds to Sulfur Are Weaker and Stranger
One of the most practically important consequences of sulfur’s lone pairs is how they participate in hydrogen bonding. Sulfur can act as a hydrogen-bond acceptor, donating lone-pair electron density to an approaching O-H or N-H group. But these hydrogen bonds are consistently weaker than the analogous bonds to oxygen, and they approach from a different angle. A crystallographic survey found that hydrogen bonds to sulfur show “a marked preference for a more perpendicular direction of approach” compared to hydrogen bonds to oxygen.4Journal of the American Chemical Society. Directionality of Hydrogen Bonds to Sulfur and Oxygen
This perpendicular preference ties directly back to the nonequivalent lone pairs described above. The 3p lone pair, which sticks out roughly perpendicular to the plane defined by sulfur and its two bonded neighbors, is the better hydrogen-bond acceptor. So donors approaching sulfur tend to come in at steep angles rather than along the direction you might naively expect. For anyone designing drug molecules or engineering protein interactions, this geometry matters: a hydrogen bond to a sulfur atom does not sit where a hydrogen bond to oxygen would, and treating them as interchangeable leads to wrong predictions about binding poses and molecular recognition.
Lone Pairs and Chirality at Sulfur
When sulfur forms three bonds to three different groups, it keeps one lone pair. That lone pair acts as a fourth substituent around the sulfur atom, making the geometry tetrahedral. And if the three bonded groups are all different from one another, the sulfur becomes a stereogenic center: it can exist in two mirror-image forms that are not superimposable, just like a carbon atom bonded to four different groups.
This is the basis of chirality in sulfinyl compounds, sulfoxides, and sulfonium salts. A review of chiral sulfur chemistry noted that tetrahedral sulfur “can become a stereogenic center when three or four different groups are attached to it,” with the lone pair filling the role of a fourth substituent.5ACS Publications. Modern Stereoselective Synthesis of Chiral Sulfinyl Compounds – Section: Chiral Organic Compounds with a Tetrahedral Sulfur Stereocenter The lone pair is invisible in most structural drawings, but it is physically real and occupies space, enough to make one enantiomer of a sulfoxide behave differently from the other in biological systems. The proton-pump inhibitor esomeprazole, for example, is the single-enantiomer form of omeprazole, and the stereocenter is a sulfur atom whose chirality depends on the spatial orientation of its lone pair.
Chirality at sulfur is generally more stable than chirality at nitrogen. A nitrogen atom with three different groups and a lone pair can rapidly invert, flipping inside out like an umbrella, which scrambles the two mirror-image forms at room temperature. Sulfur inverts much more slowly because the energy barrier to inversion is higher. This makes chiral sulfur centers practical targets for drug design and asymmetric catalysis.
Lone Pairs in Elemental Sulfur
Even in its elemental form, sulfur’s lone pairs play a structural role that is more complex than textbook pictures suggest. Crystalline sulfur typically forms eight-membered rings (S₈), where each sulfur atom bonds to two neighbors and carries two lone pairs. A high-resolution electron density study of these rings found that the lone pairs are not just passive bystanders. They can delocalize spatially up to roughly 1.3 ångströms from the sulfur nucleus, mixing with the bonding orbitals and inducing what the authors described as a “twisted σ-bond with substantial π-character.”6Journal of the American Chemical Society. A Twisted σ-Bond Mediated by a Lone-Pair in Elemental Sulfur
In plain terms, the lone pairs do not just sit in their assigned orbitals. They spread out enough to interfere with the bonds between neighboring sulfur atoms, bending and twisting those bonds in ways that a simple Lewis structure completely fails to capture. The measured electron density along the S–S bond showed a bimodal distribution that could not be explained by a conventional single bond. This is a case where the standard model of two lone pairs per sulfur atom is technically correct in its count but misleading about what those lone pairs are actually doing.
Sulfur’s Lone Pairs in Protein Biochemistry
Sulfur appears in two of the twenty standard amino acids: cysteine (with a thiol -SH group) and methionine (with a thioether -S- group). In both cases, the sulfur carries two lone pairs, and those lone pairs are central to how these amino acids interact with metal ions in proteins. Computational work comparing metal binding at cysteine and methionine found large differences in how strongly they bind copper, with the charged form of cysteine (the thiolate) binding copper far more tightly than methionine’s thioether sulfur.7PubMed Central. Comparison of metal-binding strength between methionine and cysteine residues: Implications for the design of metal-binding motifs in proteins
The difference in binding strength is enormous. When cysteine loses its thiol proton and becomes a thiolate (Cys⁻), its interaction with copper is roughly 250 kcal/mol stronger than methionine’s interaction with the same metal. This is not a subtle effect; it is the difference between a loose association and a near-covalent grip. The lone pairs on the thiolate sulfur are more available for donation to the metal because the negative charge raises the energy of the lone-pair electrons, making them better donors. Methionine’s sulfur, still neutral and flanked by two carbon groups, holds its lone pairs more tightly.
This distinction has practical consequences for protein engineering. If you want a metal-binding site that grabs copper and does not let go, you design around cysteine. If you want a site that binds more weakly, perhaps to shuttle a metal ion and release it elsewhere, methionine is the better choice. Both amino acids offer two lone pairs on sulfur, but the chemical environment around that sulfur changes how those lone pairs behave by orders of magnitude in binding energy.
Common Mistakes When Counting
The most frequent error is assuming sulfur always has two lone pairs. That is true only for divalent sulfur. Students drawing Lewis structures of sulfur dioxide, sulfuric acid, or sulfur hexafluoride will get the wrong geometry if they default to two lone pairs without checking how many bonds sulfur actually forms in that molecule. The rule is simple: start with six valence electrons, subtract the number used in bonds (counting each bond as one electron from sulfur in a simple Lewis model), and the remaining electrons form lone pairs.
A subtler mistake is treating sulfur’s lone pairs as equivalent to oxygen’s. They are not. Sulfur’s larger size, its nonequivalent 3s and 3p lone pairs, and its lower electronegativity all combine to make sulfur a weaker hydrogen-bond acceptor, a better soft-metal ligand, and a more spatially diffuse electron donor than oxygen. Swapping oxygen for sulfur in a molecule does not just change the atom; it changes how every lone pair around that atom interacts with the molecular environment.
A third common confusion involves resonance structures. In molecules like sulfur dioxide, you can draw resonance structures that place different numbers of lone pairs on sulfur. The “real” molecule is a blend of those structures, and the effective lone-pair count is somewhere between the extremes. For geometry prediction, you count the lone pairs in the structure that obeys the octet rule (one lone pair on sulfur in SO₂), but for understanding reactivity, the resonance picture gives a fuller answer. Sulfur’s willingness to participate in resonance and delocalization, including the twisted-bond behavior seen in elemental sulfur, means that lone-pair counts are best understood as useful approximations rather than rigid physical facts.