Ether Nomenclature and Structure
Diethyl ether once filled operating rooms with its sweet smell, dripping slowly onto a gauze mask to carry a patient into surgical anesthesia. That single molecule, CH₃CH₂–O–CH₂CH₃, is the simplest useful example of an entire functional class built on one deceptively plain motif: an oxygen atom bonded to two carbon groups, with no O–H bond at all. Ethers sit quietly between alcohols and alkanes in reactivity, sharing the alcohol's bent, oxygen-centered geometry but none of its hydrogen-bond-donating ability, and none of the alkane's complete inertness either — that lone-pair-rich oxygen is a Lewis base waiting for the right acid. Understanding ethers matters because they are everywhere in the practicing chemist's life before they are anywhere in a product. Tetrahydrofuran (THF), diethyl ether, and 1,4-dioxane are the workhorse solvents of organic synthesis — they dissolve organic substrates, coordinate to organometallic reagents like Grignards and organolithiums, and yet mostly refuse to react with the very species dissolved in them. That combination — good solvating power, poor reactivity — is not an accident of history but a direct consequence of the electronic structure at the C–O–C linkage, and understanding why ethers behave this way is the real content of this article. The physical picture is built from the same orbital reasoning used for Alcohol Nomenclature and Classification: an oxygen with two σ bonds and two lone pairs, roughly -hybridized, sitting at the center of a bent C–O–C angle near tetrahedral geometry. What changes on going from alcohol to ether is simple but consequential — the O–H bond is replaced by a second O–C bond. That one substitution removes the hydrogen-bond donor, dramatically raises the boiling point gap relative to alcohols of similar size, and removes the acidic proton that made alkoxides so easy to generate. The oxygen lone pairs remain, however, fully available for Lewis basic behavior, solvation of cations, and — under harsh enough acidic conditions — protonation as the first step of ether cleavage. This article works through IUPAC and common naming conventions for ethers, the orbital and geometric picture of the C–O–C unit, the physical-property consequences of losing the O–H bond, and the reactivity landscape that explains why ethers are prized as solvents but are not without their own hazards (peroxide formation in ethers exposed to air is a real laboratory danger). It closes by setting up, without yet deriving, the two reactions that make ethers synthetically important: Williamson Ether Synthesis for building them, and Ether Cleavage by Strong Acids for eventually taking them apart.
- An ether is R–O–R′: an oxygen bonded to two carbon groups, with no O–H bond.
- IUPAC names ethers as alkoxyalkanes (substitutive) — the smaller R–O– chain becomes an "alkoxy" prefix on the larger parent chain.
- Common names list both alkyl groups alphabetically followed by "ether" (diethyl ether, ethyl methyl ether); symmetrical ethers get the "di-" prefix.
- The C–O–C bond angle is bent, roughly 104–112°, from -hybridized oxygen carrying two lone pairs.
- Ethers cannot hydrogen-bond to each other (no O–H), giving them boiling points far below isomeric alcohols, but they accept H-bonds from water and alcohols, giving useful water solubility at low molecular weight.
- Ethers are kinetically inert toward bases, oxidants, and most nucleophiles/electrophiles, which is exactly why they are the default solvent for reactive organometallics — but they are cleaved by strong acids (HI, HBr) and can slowly autoxidize to explosive peroxides on storage.
- Key idea: an ether is an alcohol with the acidic, hydrogen-bond-donating O–H replaced by a second, unreactive C–O bond — same bent oxygen, same lone pairs, but a fundamentally different reactivity profile.
Ethers are R–O–R′: the oxygen keeps its two lone pairs and bent, roughly geometry from alcohols, but loses the O–H bond — so it cannot hydrogen-bond as a donor, cannot be deprotonated to an alkoxide, and is kinetically unreactive except toward strong Brønsted or Lewis acids that can protonate or coordinate a lone pair and trigger C–O bond cleavage.
Where this lives today
Diethyl ether and THF remain the two most-used solvents for generating and reacting Grignard reagents and organolithiums; the ether oxygen's lone pair coordinates to the electropositive magnesium or lithium center, stabilizing the organometallic and keeping it in solution without itself being attacked — a direct, everyday exploitation of the Lewis-basic-but-inert duality discussed above. 1,4-Dioxane, a six-membered ring ether with two oxygens, is similarly used as a high-boiling, water-miscible reaction solvent, though its regulatory status is complicated by suspected carcinogenicity at high exposure. Cyclic and acyclic polyethers show up throughout biology and materials science: crown ethers (cyclic polyethers such as 18-crown-6) selectively bind alkali metal cations inside a ring of ether oxygens, each lone pair pointing inward to chelate the cation — a macrocyclic amplification of the same Lewis basicity that lets simple diethyl ether coordinate a single Mg²⁺ or Li⁺ ion. Polyethylene glycol (PEG), a linear polyether, is used industrially and pharmaceutically as a biocompatible, water-soluble polymer for drug delivery (PEGylation of protein therapeutics extends their circulation half-life), leveraging exactly the hydrogen-bond-accepting (but not donating) character of ether oxygens established mechanistically above. Fentanyl and many other pharmaceuticals carry ether linkages as metabolically stable "spacer" groups precisely because ethers resist the oxidative and hydrolytic enzymes that would rapidly cleave an ester or amide in the same position — the same kinetic inertness that makes ethers good laboratory solvents also makes them good, slowly-metabolized substructures in drug design. On the hazard side, the same α-C–H bonds that make ethers useful (activated slightly by the adjacent oxygen lone pair) are the site of slow autoxidation to peroxides on long-term storage, which is why bottles of old diethyl ether or THF are treated with real caution in any working laboratory, and why commercial ethers are typically sold with a radical-trapping stabilizer such as BHT.
Ask the tutor: How exactly does protonation of an ether oxygen by HI trigger versus cleavage, and what decides which mechanism operates for a given ether?
The physical picture
Start from the atom itself. A neutral oxygen has six valence electrons. In an ether R–O–R′, oxygen forms two σ bonds (to the two carbon substituents) and retains two lone pairs, giving it four electron domains total — the same electron count and hybridization picture as water or an alcohol. Valence-shell electron-pair repulsion (VSEPR) reasoning, refined by orbital-mixing arguments, places these four domains at roughly tetrahedral positions, built from -hybridized oxygen orbitals: two hybrids form σ(C–O) bonds by overlap with carbon orbitals, and two hybrids hold the lone pairs. The bond angle is not a clean 109.5°, and understanding why is itself instructive. Lone pairs are held closer to the nucleus and spread into more diffuse, "fatter" orbital lobes than bonding pairs, because a lone pair is not constrained by a bonding partner pulling electron density away. Extra lone-pair–bond-pair repulsion compresses the R–O–R′ angle below the ideal tetrahedral value in the parent case (water is bent at 104.5°), but as the substituents R and R′ grow bulkier, steric repulsion between the alkyl groups pushes the angle back up. Dimethyl ether sits near 111–112°, and diethyl ether is close to 112° as well — both above tetrahedral, because two methyl or ethyl groups repel each other more than two O–H hydrogens do. This angle-widening-with-bulk trend is a recurring theme: the same lone-pair/bond-pair competition that gives water its bent shape gives ethers a slightly more open version of that same bend. Two structural features follow directly and explain almost all of ether chemistry. First, both C–O bonds are polar (oxygen is more electronegative than carbon, ), so the molecule has a net dipole moment — smaller than an alcohol's because the two C–O dipoles partially oppose each other around the bent oxygen, but still enough to give ethers modest polarity and real solvating power for polar and ionic solutes. Second, and most consequentially, there is no O–H bond anywhere in the molecule. That single fact is the hinge on which the entire reactivity comparison between alcohols and ethers turns.
Compare this directly to methanol or ethanol: replace one alkyl group with H, and you regenerate an O–H bond, restoring hydrogen-bond-donor ability and an acidic proton with –18. Removing that O–H (going from alcohol to ether) removes both properties simultaneously — this is precisely the trade at the heart of the alcohol/ether comparison.
Both molecules have identical molecular formula C₄H₁₀O and similar molar mass, yet 1-butanol boils at 117 °C while its ether isomer, diethyl ether, boils at 35 °C — an 82 °C gap driven entirely by the intermolecular hydrogen-bonding network available to the alcohol's O–H and unavailable to the ether. Ethers can still accept hydrogen bonds from water or alcohols using their lone pairs (they just cannot donate one), which is why low-molecular-weight ethers such as diethyl ether and THF retain meaningful water solubility despite their nonpolar-feeling alkyl skeletons.
The core ideas
The naming of ethers rests on treating the smaller alkoxy fragment as a substituent, in exact parallel to how a hydroxyl is treated as a substituent prefix in the IUPAC name of a polyfunctional alcohol.
IUPAC substitutive nomenclature. For an unsymmetrical ether R–O–R′ with R the longer/senior chain:
where: — the shorter or lower-priority chain R′, together with the oxygen, becomes the prefix "alkoxy" (methoxy, ethoxy, propoxy, isopropoxy, tert-butoxy, phenoxy for R′ = aryl), — the longer chain R is numbered to give the lowest locant to the point of attachment of the alkoxy group, — for a chain that also bears other substituents, alphabetization of prefixes follows the usual IUPAC rule (treat "ethoxy" under "e," "methoxy" under "m," and so on).
Common (radicofunctional) nomenclature. List the two alkyl/aryl groups attached to oxygen, alphabetically, as separate words, followed by the class name "ether":
where a symmetrical ether (R = R′) is named with the prefix "di-" attached once, e.g. "diethyl ether," not "ethyl ethyl ether." Reading these two equations together: IUPAC nomenclature always treats an ether as a substituted alkane (an "-oxy" group hanging off a parent chain, exactly as "-hydroxy" or, more commonly, the suffix "-ol" is used for alcohols), while common nomenclature treats the ether as a compound built from two named alkyl pieces joined by the word "ether," with no parent-chain selection needed at all. Both are IUPAC-sanctioned for simple ethers; common names dominate everyday lab speech (diethyl ether, "ether"; methyl tert-butyl ether, MTBE), while substitutive names are required once the molecule carries additional functional groups that need a proper parent chain and locants (e.g., 2-methoxyethan-1-ol, 1-ethoxy-2-methylpropane). Cyclic ethers are named either as substitutive "oxa-" replacement names (replacing a CH₂ in a cycloalkane with O, e.g. oxirane, oxetane, oxolane/tetrahydrofuran, oxane/tetrahydropyran) or by long-entrenched common/trivial names (ethylene oxide for oxirane, THF for tetrahydrofuran, dioxane for the six-membered ring with two oxygens at the 1,4-positions). Three-membered cyclic ethers (epoxides) are structurally and reactively distinct enough — highly strained, highly electrophilic — that they are treated as their own functional class rather than as "ordinary" ethers; that strain-release reactivity is the subject of its own article and is only the endpoint, not the topic, here.
Goal. Given the structure CH₃–O–CH(CH₃)–CH₂–CH₃ (a four-carbon chain bearing a methoxy group on C2), produce the correct IUPAC substitutive name.
Step 1 — Identify the two carbon fragments joined by oxygen. On one side of the ether oxygen sits a single methyl group, CH₃–. On the other side sits a four-carbon chain, CH₃–CH(–)–CH₂–CH₃, attached to oxygen at its second carbon. There is no competing rule about which side is "senior" for a simple ether — by convention, and because it yields the more informative name, the longer carbon chain is chosen as the parent, and the shorter O–R′ fragment becomes the alkoxy substituent.
Step 2 — Name the alkoxy substituent. The smaller fragment is O–CH₃. As a substituent prefix this is "methoxy" (compare "hydroxy" for –OH; the "-oxy" suffix is the general pattern for any –O–R substituent prefix, with "meth-" specifying R = CH₃).
Step 3 — Number the parent chain to give the substituent the lowest locant. The parent chain is butane: C1–C2–C3–C4. The methoxy group sits on the carbon that is the second carbon counting from one end (CH₃–CH(OCH₃)–CH₂–CH₃). Numbering from the correct end places methoxy at C2 rather than C3, since IUPAC rules require the lowest possible locant set. So the parent is "butane," numbered so the substituent is at position 2.
Step 4 — Assemble the name. Combine locant, substituent prefix, and parent chain name:
Step 5 — Cross-check with the common name. The same molecule, named radicofunctionally, is "sec-butyl methyl ether": the two groups on oxygen are a methyl group and a sec-butyl group (CH₃CH(–)CH₂CH₃ attached through its 2-position), listed alphabetically (methyl before sec-butyl is not alphabetical — "butyl" precedes "methyl" alphabetically ignoring the "sec-" locant prefix), giving "sec-butyl methyl ether." Both names describe the identical connectivity; the IUPAC name is unambiguous and locant-based, while the common name is a quick shorthand built from recognizable alkyl-group names.
Go deeper · Derivation B · 3 minWhy the O–H bond's absence controls both intermolecular forces and reactivity
Goal. Show mechanistically/orbital-wise why removing the O–H bond (going from R–OH to R–O–R′) simultaneously (a) collapses hydrogen-bond donation and (b) removes the low-barrier deprotonation pathway, leaving ethers as comparatively inert, weakly Lewis-basic solvents.
Step 1 — Catalogue what an alcohol's O–H bond provides. In R–OH, the O–H σ bond is strongly polarized ( on H, on O) because of the ~1.4-unit electronegativity difference between O and H combined with oxygen's small size and high charge density. This polarized bond does two jobs: it is a hydrogen-bond donor (the electron-poor H can be shared with a lone pair on a neighboring O), and it is a Brønsted acid (a base can remove H⁺, generating an alkoxide with (ROH) 16–18).
Step 2 — Remove the O–H bond and replace it with a C–O bond. In R–O–R′, both positions on oxygen that could have carried an ionizable, hydrogen-bond-donating H now instead carry carbon. A C–H bond has a much smaller electronegativity difference () and the carbons are buried inside alkyl groups, not exposed the way the alcohol's H was. There is consequently no acidic proton directly on oxygen and no hydrogen available to donate into a neighboring lone pair as a classical O–H···O hydrogen bond.
Step 3 — Track what is retained. The oxygen itself is untouched: it is still -hybridized (approximately), still bent, and still carries two full lone pairs, exactly as in the alcohol or in water. These lone pairs are why an ether remains a competent hydrogen-bond acceptor — it can still solvate the acidic O–H or N–H protons of another molecule (water, an alcohol, an ammonium salt) even though it cannot supply an acidic proton of its own. This asymmetry (acceptor but not donor) is the direct structural reason diethyl ether is water-soluble at low concentration despite having no way to hydrogen-bond to itself.
Step 4 — Follow the consequence for boiling point. Alcohols form extended hydrogen-bonded networks in the liquid phase (each O–H both donating to, and accepting from, neighboring molecules), which costs substantial extra energy to break on vaporization. Ethers cannot form this network among themselves (no donor bond), so their only significant intermolecular attractions are weaker dipole–dipole interactions and London dispersion forces. Lower intermolecular cohesion means a much lower boiling point for a given molar mass, exactly the 82 °C gap seen between 1-butanol and diethyl ether.
Step 5 — Follow the consequence for chemical reactivity. Because there is no acidic O–H, ethers cannot be deprotonated by ordinary bases to generate a reactive alkoxide — there is simply no accessible low-energy site for base-mediated chemistry to start. The two lone pairs on the ether oxygen remain nucleophilic/Lewis basic, so an ether can react with a sufficiently strong electrophile or Brønsted acid: protonation of oxygen by concentrated HI or HBr (a strong enough acid to overcome the poor basicity of an ether oxygen, ) generates a good leaving group (a neutral ether becomes a positively charged oxonium), which then permits or attack by the halide nucleophile at carbon — this is the entry point into Ether Cleavage by Strong Acids. Absent such a strong activator, the C–O bonds of an ether are simply not electrophilic enough, and the oxygen is simply not acidic enough, for the molecule to react with the mild acids, bases, oxidants, and nucleophiles present in a typical reaction flask — which is exactly the property that makes ethers valuable, chemically "invisible" solvents.
Step 6 — State the resulting hierarchy of reactivity.
where the alcohol reacts as both a weak acid and a nucleophile/electrophile precursor, the ether reacts only as a weak Lewis base (toward strong acids, or as a ligand toward metal cations and organometallics), and the alkane is essentially unreactive toward all polar reagents. This hierarchy is why THF and diethyl ether are the default reaction media for Grignard and organolithium chemistry: the ether oxygen's lone pair coordinates to and stabilizes the electron-deficient, highly reactive R–Mg–X or R–Li center (Lewis acid–base adduct formation), while the ether's own C–O bonds remain untouched by the very same organometallic reagent it is solubilizing.
Worked examples
Example 1 (foundational): name and classify an ether from its skeletal structure. Consider the molecule CH₃CH₂–O–CH₂CH₂CH₃ (ethyl propyl ether by common naming). Work out both names and check the C–O–C bond count. Step 1. Identify the two alkyl fragments attached to oxygen: an ethyl group (CH₃CH₂–) and a propyl group (CH₃CH₂CH₂–). Both are simple, unbranched alkyl chains — there is no ambiguity about which fragment is which. Step 2 — common name. Alphabetize the two group names: "ethyl" before "propyl." The common name is ethyl propyl ether. Step 3 — IUPAC name. The longer chain (propyl, three carbons) becomes the parent "propane"; the shorter chain (ethyl, two carbons) plus oxygen becomes the substituent "ethoxy." Numbering propane to give the substituent the lowest locant places it at C1 (the terminal carbon where oxygen attaches):
Step 4 — sanity check on oxygen's bonding. Oxygen here has exactly two σ bonds (to the ethyl carbon and to the propyl carbon) and two lone pairs — four electron domains, consistent with hybridization and a bent C–O–C angle expected near 111–112° by analogy to diethyl ether, since both flanking groups are simple, unbranched, sterically unremarkable alkyl chains. Step 5 — limiting case. If the propyl group were replaced by a second ethyl group, the molecule becomes symmetrical, CH₃CH₂–O–CH₂CH₃, and the common name collapses correctly to "diethyl ether" (not "ethyl ethyl ether"), verifying that the "di-" convention is simply the R = R′ special case of the general two-groups-alphabetized rule.
Example 2 (applied): predicting relative water solubility and boiling point among three C₄H₁₀O-type/near-isomers. Compare 1-butanol (CH₃CH₂CH₂CH₂OH, bp 117 °C), diethyl ether (CH₃CH₂OCH₂CH₃, bp 35 °C), and pentane (CH₃CH₂CH₂CH₂CH₃, bp 36 °C, included only for contrast as it contains no oxygen at all). Rank their boiling points and water solubilities, justifying each ranking mechanistically. Step 1 — catalogue intermolecular forces available to each. 1-Butanol has an O–H bond: it is both hydrogen-bond donor and acceptor, allowing extensive self-association in the neat liquid. Diethyl ether has two ether-oxygen lone pairs (acceptor only, no donor) plus dipole–dipole and dispersion forces from its polar but O–H-free C–O–C unit. Pentane has only London dispersion forces, since carbon and hydrogen have essentially no electronegativity difference and there is no heteroatom at all. Step 2 — rank boiling points. Hydrogen-bond networks (donor + acceptor, as in the alcohol) demand the most energy to disrupt on vaporization, dipole-only interactions (the ether, acceptor-only, weaker net cohesion) demand less, and pure dispersion forces (the alkane) demand the least for a molecule of comparable size:
Notice the ether and the alkane end up nearly coincident: once the alcohol's hydrogen-bonding network is switched off (ether) or the oxygen is removed altogether (alkane), the boiling point is governed almost entirely by molecular size and shape, not by the small extra dipole–dipole contribution the ether retains over the alkane. Step 3 — rank water solubility. 1-Butanol can donate and accept hydrogen bonds with water, giving substantial (though not unlimited, since the butyl chain is fairly hydrophobic) miscibility. Diethyl ether can only accept a hydrogen bond from water's O–H — no donation possible — giving a smaller but still real solubility (about 6–7 g per 100 mL water at room temperature, enough to matter in extraction chemistry). Pentane, with no heteroatom and no hydrogen-bond-accepting site at all, is essentially insoluble in water (parts-per-million level).
Step 4 — the practical payoff. This is exactly why diethyl ether is the classic solvent for liquid–liquid extraction of organic compounds from an aqueous layer: it is polar enough (via its lone pairs and dipole) to dissolve many organic solutes and to itself carry a small, controllable trace of water, yet nonpolar enough, and poor enough a hydrogen-bond donor, to remain a distinct, immiscible layer from bulk water.
Example 3 (exam/competition level): rationalize a bond-angle trend and connect it to reactivity toward a Lewis acid. Dimethyl ether has a C–O–C angle of about 111.7°, while di-tert-butyl ether — with two bulky tert-butyl groups — has a C–O–C angle widened to roughly 132°, one of the largest ether bond angles known. Explain the trend mechanistically, and then predict, with reasoning, which ether would form a stronger Lewis acid–base adduct with BF₃: dimethyl ether or di-tert-butyl ether. Step 1 — set up the competing effects on the bond angle. The oxygen in any dialkyl ether carries two lone pairs and two bonding pairs to carbon, in an approximately arrangement. Two effects push against the idealized 109.5°: (i) lone-pair/lone-pair and lone-pair/bond-pair repulsion, which is largest for the two lone pairs and tends to compress the C–O–C angle below tetrahedral (as it does in water, 104.5°, and to a lesser extent even in dimethyl ether, which is still slightly above tetrahedral because two methyls already repel each other more than two lone hydrogens would), and (ii) steric repulsion directly between the two alkyl substituents, which widens the angle as those substituents get bulkier. Step 2 — apply this to the two specific ethers. In dimethyl ether, the two methyl groups are small; steric widening is modest, and the angle (111.7°) sits only slightly above the tetrahedral value. In di-tert-butyl ether, each substituent is a bulky, three-methyl-branched tert-butyl group; the steric repulsion between these two large groups, when forced to sit at a ~110° angle, becomes severe (A-value-scale strain), so the molecule relaxes toward a much wider angle (~132°) to relieve that steric clash — even though this simultaneously worsens the geometric overlap available for the O lone pairs relative to an idealized arrangement. Step 3 — connect geometry to Lewis basicity. Coordination to BF₃ requires the ether oxygen to donate a lone pair into boron's empty orbital, forming a new dative O→B bond and pulling the oxygen toward a more pyramidalized, roughly tetrahedral-with-adduct geometry (four groups around O: two alkyl carbons, the new B, and the remaining lone pair). This adduct formation is opposed by steric clash between the incoming BF�3/its substituents and the bulky groups already on oxygen — exactly the same kind of steric repulsion that widened the C–O–C angle of the free ether in the first place. Step 4 — predict and justify the ranking. Di-tert-butyl ether, despite having "more" alkyl character intuitively suggesting stronger electron donation, is in fact a much weaker Lewis base toward bulky acids such as BF₃ than dimethyl ether, because the severe steric bulk around oxygen (already reflected in its wide 132° resting angle) blocks close approach of the boron center — a textbook case of steric control overriding a naive electronic (more alkyl substitution → more electron-rich oxygen) argument:
Step 5 — limiting-case check. If both alkyl groups shrank to hydrogen (the ether "limit" of water itself), steric widening vanishes entirely and the angle relaxes to the smallest value in the series (104.5°), while Lewis basicity toward small acids like H⁺ becomes comparatively unhindered — consistent with water being an excellent, sterically unencumbered Lewis base/H-bond partner, confirming that the steric-widening mechanism identified above, not some independent electronic effect, is what is driving the entire angle-and-basicity trend across the series H₂O → Me₂O → t-Bu₂O.
Limiting cases, scope & common traps
| Situation | What happens | Why |
|---|---|---|
| R = R′ (symmetrical ether) | Common name uses "di-" once: "diethyl ether," never "ethyl ethyl ether" | Alphabetization/listing collapses to a single repeated group name |
| One R = H | Not an ether at all — reverts to an alcohol, R–OH | The defining feature of an ether is two carbon substituents on oxygen, zero O–H bonds |
| Both R, R′ bulky (e.g., di-tert-butyl ether) | C–O–C angle widens well past tetrahedral (~130°+); Lewis basicity toward bulky acids drops sharply | Steric repulsion between substituents dominates over the modest lone-pair compression effect |
| Ether stored in air, especially THF or diethyl ether, for long periods | Slow radical autoxidation at the α-C–H forms explosive peroxides; never distill to dryness | The C–H bonds α to ether oxygen are weakened (captodative-like stabilization of the resulting radical by the adjacent O lone pair), making them susceptible to slow O₂ radical-chain attack |
| Strong Brønsted acid present (HI, HBr, conc.) | Ether oxygen is protonated, generating a leaving group; C–O bond cleaves via / | Overcomes the ether's poor intrinsic basicity (); sets up Ether Cleavage by Strong Acids |
| Ordinary base or mild nucleophile/electrophile present | No reaction | No acidic O–H to remove, no sufficiently electrophilic carbon or leaving group under mild conditions |
| Epoxide (three-membered cyclic ether) | Highly reactive, opposite of "typical" ether behavior | Ring strain (~27 kcal/mol) makes the C–O bonds unusually electrophilic; treated as a separate functional class |
The chemist behind it
Ether nomenclature and the electronic description of the C–O–C bond are the product of decades of collective structural chemistry rather than a single discovery, but one figure looms over the interpretive framework used throughout this article: Linus Pauling, whose hybridization and electronegativity concepts gave chemists the vocabulary — oxygen, polar covalent bonds, lone-pair repulsion — to explain why ethers are bent, weakly polar, and weakly basic, rather than simply cataloguing that they are.
Origins · 1 minLinus Pauling and how this was discovered
Pauling's 1931 paper "The Nature of the Chemical Bond" introduced hybrid orbitals (, , ) as a way to rationalize observed molecular geometries, including the bent shape of ethers, water, and alcohols, in terms of orbital mixing rather than ad hoc geometric rules. His electronegativity scale, developed the same decade, gave chemists a quantitative handle on bond polarity — exactly the tool used above to explain why the C–O bond in an ether is polar enough to matter for solvation but the C–H bonds nearby are not.
Is an ether the same thing as an alcohol with the H replaced by an alkyl group?
Yes, structurally that is exactly the relationship: take R–OH and replace the hydroxyl hydrogen with a second carbon substituent R′, and you obtain the ether R–O–R′. The oxygen's hybridization, lone pairs, and general bent geometry carry over almost unchanged from the alcohol. What changes completely is the reactivity: the acidic, hydrogen-bond-donating O–H bond is gone, replaced by an inert C–O bond, so the resulting ether cannot be deprotonated by ordinary bases and cannot hydrogen-bond to itself. This is why ethers are often introduced as "capped" or "protected" alcohols in synthesis — converting a reactive alcohol into a comparatively inert ether is in fact a standard protecting group strategy used to shield an –OH from unwanted reactions elsewhere in a molecule.
Why do chemists use ethers like THF and diethyl ether as solvents instead of, say, water or alcohols?
Ethers combine two properties that are hard to find together: enough polarity (from the C–O dipoles and lone pairs) to dissolve a wide range of organic substrates and to coordinate to metal centers, and enough chemical inertness to survive contact with very reactive reagents. Water and alcohols both carry an acidic, electrophilic O–H bond that reacts destructively with strong bases and organometallics (a Grignard reagent, for instance, is instantly protonated and destroyed by water or an alcohol's O–H). An ether has no such O–H bond, so it can dissolve and stabilize (via Lewis-base coordination to the metal) a Grignard or organolithium reagent without being decomposed by it — a chemical "spectator" role that few other polar solvents can fill.
Why is the C–O–C angle in ethers usually greater than the tetrahedral 109.5°?
Two effects compete to set this angle: lone-pair/bond-pair repulsion, which by itself would compress the angle below tetrahedral (as seen in water, 104.5°), and steric repulsion directly between the two alkyl substituents, which widens the angle as those substituents get larger. In dimethyl ether the steric effect already outweighs the lone-pair compression, giving about 111.7°; in bulkier ethers such as di-tert-butyl ether, steric repulsion dominates overwhelmingly, pushing the angle out to roughly 132°. The general rule is that bulkier substituents on ether oxygen push the C–O–C angle progressively further above tetrahedral.
Can ethers act as nucleophiles or bases in a reaction?
Yes, but only toward sufficiently strong electrophiles or acids — the ether oxygen's lone pairs are real and available, they are simply not reactive enough to engage mild reagents. Toward strong Lewis acids (BF₃, AlCl₃) or strong Brønsted acids (concentrated HI, HBr, or H₂SO₄ at elevated temperature), the oxygen lone pair readily coordinates or is protonated, generating an oxonium species that can then undergo carbon–oxygen bond cleavage. This is the mechanistic entry point for Ether Cleavage by Strong Acids, and it is also exactly the interaction that lets an ether solvent coordinate to and stabilize a Grignard reagent's electron-deficient magnesium center.
Why do old bottles of diethyl ether or THF pose an explosion risk?
The C–H bonds on the carbon directly attached to the ether oxygen (the α-carbons) are weakened relative to ordinary alkane C–H bonds, because the resulting radical, formed by homolysis, is stabilized by delocalization into the adjacent oxygen lone pair. Over months to years of exposure to atmospheric oxygen, especially with light or trace metal contamination, this bond is slowly abstracted in a radical chain autoxidation, building up organic peroxides that concentrate as the more volatile ether evaporates. These peroxides are shock- and heat-sensitive and have caused real laboratory explosions, particularly during distillation to dryness, which is why old ether bottles are tested for peroxides (or simply not opened) before use.
How does ether nomenclature relate to naming alcohols?
The logic is directly parallel. For an alcohol, the carbon chain bearing the –OH group is chosen as the parent, numbered to give the hydroxyl group the lowest locant, and named with the suffix "-ol." For an ether, there is no single "-ol"-type suffix in common use; instead, one of the two carbon fragments on oxygen is chosen as the parent chain (the longer one, by convention), and the other fragment, together with the oxygen, becomes an "alkoxy" substituent prefix — the same kind of prefix treatment used whenever a hydroxyl is not the principal characteristic group of a molecule (named as "hydroxy-" instead of "-ol"). Recognizing this shared logic makes it straightforward to name a molecule that contains both an alcohol and an ether functional group at once, such as 2-methoxyethan-1-ol.
Are epoxides just small-ring ethers, and do they behave the same way?
Structurally, yes — an epoxide is a three-membered ring ether, oxygen bonded to two carbons that are themselves bonded to each other. Reactively, no — epoxides are dramatically more reactive than ordinary (five-, six-membered-ring or acyclic) ethers because the three-membered ring imposes roughly 27 kcal/mol of angle strain, forcing the C–O–C angle down near 60° instead of the comfortable ~104–112° enjoyed by unstrained ethers. That severe strain makes the ring-opening reaction thermodynamically very favorable and kinetically accessible even to comparatively weak nucleophiles and mild acids, in sharp contrast to the strong-acid-only cleavage chemistry of ordinary ethers described here. Epoxides are consequently treated as their own reactive functional class rather than as typical ethers.