Constitutional Isomers and the Homologous Series

Petroleum refining is, at its core, an industrial-scale exercise in separating and interconverting constitutional isomers. Straight-run gasoline is dominated by unbranched and lightly branched alkanes with poor knock resistance (low octane number); catalytic reforming and isomerization units deliberately rearrange connectivity — converting n-hexane and n-heptane into their more branched isomers (2,2-dimethylbutane, 2,3-dimethylpentane, and related structures) — because highly branched isomers resist autoignition far better under compression, directly raising octane rating without changing the molecular formula or adding any new atoms. In pharmaceutical chemistry, recognizing constitutional isomerism is the first filter applied to any newly proposed synthetic route: a synthesis that accidentally produces a constitutional isomer of the target (wrong connectivity, right formula) will not simply give a slightly-impure product — it gives a different molecule altogether, potentially inactive or even toxic, that must be separated by chromatography rather than "purified away." Understanding degree of branching also anticipates radical halogenation selectivity (tertiary C–H bonds react fastest) and, further downstream, carbocation stability trends in substitution and elimination chemistry — both topics that rest directly on the 1°/2°/3°/4° classification introduced here. Materials science leans on the same ideas in reverse: branched polyethylene (LDPE) and linear polyethylene (HDPE) are, in effect, macroscopic homologs of exactly this isomer distinction scaled up to enormous $n$ — branching frequency along the polymer backbone controls density, crystallinity, and mechanical strength, all traceable to the same packing/surface-area logic that separates n-pentane from neopentane at the small-molecule scale.

Wait — reviewing quiz 5: the correct answer must be CnH2n, not the option marked. Let me restate it correctly below.

Ask the tutor: Why does 2,2-dimethylbutane have a lower boiling point than its isomer 2,3-dimethylbutane, even though both have identical molecular formulas and the same total number of methyl branches?

A molecular formula like C₆H₁₄ tells you the atom inventory but is silent on architecture. Carbon is tetravalent and, in an alkane, every carbon is sp³-hybridized: four σ bonds arranged toward the corners of a tetrahedron, bond angles near 109.5°, each bond a head-on overlap of an sp³ lobe with either another carbon's sp³ lobe (C–C) or a hydrogen 1s orbital (C–H). Because a σ bond is cylindrically symmetric about the internuclear axis, rotation around any C–C bond is (nearly) free — this is why we can draw an alkane's skeleton as a flat zig-zag on paper without worrying that we've picked "the wrong conformation": conformations interconvert rapidly at room temperature, but connectivity — the graph of which atom is bonded to which — does not change without breaking and remaking bonds. This is the crucial distinction: two molecules that differ only in conformation (e.g., anti vs. gauche butane) are the same compound, isolable as one substance. Two molecules that differ in the connectivity graph itself — hexane versus 2-methylpentane versus 2,2-dimethylbutane — are different compounds with different names, different boiling points, and (in principle) separable by distillation. Constitutional isomerism is a statement about the bonding graph, full stop; it has nothing to do with three-dimensional shape in the conformational sense, though it sets the stage for the deeper 3-D isomerism (stereoisomerism) that appears once stereocenters or restricted rotation enter the picture. The tool for keeping this straight is degree of branching, assigned atom-by-atom. A carbon bonded to exactly one other carbon is primary (1°); bonded to two other carbons, secondary (2°); to three, tertiary (3°); to four, quaternary (4°). A hydrogen is classified by the carbon it sits on — a hydrogen on a 3° carbon is a "tertiary hydrogen," and this label matters enormously later (radical halogenation rates, carbocation stability, etc., are graded 3° > 2° > 1°). The branching pattern is what actually distinguishes isomers of the same formula from one another; enumerating isomers is really an exercise in enumerating all distinct branching patterns.

Notice the pattern already visible in the figure: as branching increases, the molecule becomes more compact and sphere-like, and (foreshadowing the physical-properties article) its boiling point drops. That trend is not coincidence and not memorized — it follows from how much surface area is available for Van der Waals forces between neighboring molecules, which is a story for another day, but the cause — the branching pattern — is exactly what this article teaches you to generate and count.

1. The general formula for acyclic, saturated alkanes. For any acyclic alkane with $n$ carbons and no multiple bonds, no rings, and no heteroatoms:

where: $n$ = number of carbon atoms; the formula holds regardless of how branched the skeleton is — branching redistributes the hydrogens among carbons but never changes the total hydrogen count, because branching only rearranges which carbons are bonded to which, without changing the total number of bonds in an acyclic graph with $n$ vertices.

2. Ring correction. Every ring (every extra C–C bond beyond what a tree/chain needs) removes exactly two hydrogens:

where: $r$ = number of rings (more generally, the number of independent ring closures — the degree of unsaturation contributed by rings alone, since each ring and each π bond both cost one $\text{H}_2$ equivalent, i.e. 2 H).

3. Degree of unsaturation (rings + π bonds combined), previewed here for completeness and made rigorous once alkenes/alkynes and rings both appear together:

where: $H$ is the actual hydrogen count of the real formula, $n$ the carbon count, $r$ the ring count, $\pi$ the number of π bonds (C=C, C=O, etc.). For pure, saturated, acyclic alkanes DoU $=0$, recovering formula (1).

4. Isomer count as a function of $n$ is not a closed-form elegant formula (it is a genuinely hard combinatorial enumeration problem in general — related to counting unlabeled trees), but for small $n$ it can be derived by exhaustive, systematic construction, which is exactly what the two mechanisms below do.

Reading this: the count is trivial (only one skeleton possible) through butane, then grows combinatorially — this acceleration is the mathematical signature of branching options multiplying as the chain lengthens.

Example 1 (foundational) — Confirm the formula and classify every atom in 3-methylhexane. 3-Methylhexane, $\text{C}_7\text{H}_{16}$, has the skeleton: a six-carbon main chain (hexane numbering C1–C6) with a methyl branch at C3. Step 1: verify the formula. By the derivation above, $n=7 \Rightarrow H = 2(7)+2 = 16$. The formula $C_7H_{16}$ is confirmed independently of the specific skeleton — any acyclic, saturated seven-carbon skeleton must be C₇H₁₆, whether it's n-heptane, 2-methylhexane, or 3-methylhexane. Step 2: draw the skeleton and count each carbon's neighbors. Main chain: C1–C2–C3–C4–C5–C6, with a CH₃ branch (call it C7) hanging off C3.

• C1: bonded only to C2 → 1° carbon (3 H's, terminal).
• C2: bonded to C1 and C3 → 2° carbon (2 H's).
• C3: bonded to C2, C4, and C7 (the branch) → 3° carbon (1 H — this is the tertiary hydrogen of the molecule, and it will later be the site most reactive toward radical halogenation because the resulting radical is most stabilized by hyperconjugation).
• C4: bonded to C3 and C5 → 2° carbon (2 H's).
• C5: bonded to C4 and C6 → 2° carbon (2 H's).
• C6: bonded only to C5 → 1° carbon (3 H's, terminal).
• C7 (the branch methyl): bonded only to C3 → 1° carbon (3 H's).

Step 3: check the hydrogen tally sums to 16. $3+2+1+2+2+3+3 = 16$. ✓ — matches the formula exactly, confirming both the classification and the earlier bond-counting derivation simultaneously (this cross-check is exactly the sort of internal consistency test you should run on every structure you draw).

Example 2 (applied) — How many constitutional isomers does C₆H₁₄ (hexane) have? Systematic construction. This is the enumeration promised earlier, done by fixing the longest possible chain first and then working down. Step 1: the unbranched skeleton. n-Hexane: C–C–C–C–C–C. One structure. (Isomer 1.) *Step 2: longest chain = 5 carbons, one methyl branch.* A pentane backbone (C1–C5) with a single CH₃ substituent. The branch can sit at C2 or C3 (C4 is equivalent to C2 by the symmetry of numbering from the other end — always check whether two "different-looking" positions are secretly the same molecule viewed from opposite ends of the chain).

• 2-methylpentane (Isomer 2)
• 3-methylpentane (Isomer 3)

(Putting the branch at "C4" just regenerates 2-methylpentane numbered from the other end — not a new isomer. This symmetry check is the single most common source of over-counting errors.) Step 3: longest chain = 4 carbons, two substituents (both must be methyl, since a longer branch would just extend the main chain and contradict "longest chain = 4"). A butane backbone (C1–C4) needs two extra carbons distributed as substituents summing to $C_6H_{14}$ overall. Two CH₃ groups on the same or different backbone carbons:

• 2,2-dimethylbutane: both methyls on C2 (Isomer 4).
• 2,3-dimethylbutane: one methyl on C2, one on C3 (Isomer 5).

(Placing methyls at 2,4- or 3,3- either duplicates a chain-length-5 case already counted, or is symmetric to 2,2- by renumbering — verify by checking that no numbering choice you haven't already used produces a genuinely new connectivity.) Step 4: could the longest chain be only 3 carbons? A propane backbone with three extra carbons to place would require at least one substituent to be an ethyl group or two methyls on one carbon, both of which necessarily create a four-carbon-long path somewhere in the molecule — contradicting the assumption that 3 is the longest chain. So no new isomers arise here; this is why fixing chain length from the longest downward, and stopping once a contradiction like this appears, guarantees the enumeration terminates and is complete. Total: 5 constitutional isomers of C₆H₁₄ — matching the count quoted in the core-ideas table, and matching the well-known experimental fact that five distinct hexanes are separable by careful fractional distillation.

Example 3 (exam/competition level) — Enumerate all constitutional isomers of C₅H₁₀ that contain exactly one ring (monocyclic), and classify their carbons. This tests the ring-correction formula together with isomer enumeration simultaneously — a natural "next level" combination question. Step 1: confirm C₅H₁₀ is consistent with $r=1$. Using $C_nH_{2n+2-2r}$ with $n=5$: $H = 2(5)+2-2(1) = 8+2=10$. ✓ Consistent — C₅H₁₀ with one ring is a legitimate formula (it is also consistent with $r=0,\ \pi=1$, i.e., an acyclic pentene, which is why "one ring" must be stated explicitly rather than inferred from the formula alone — degree-of-unsaturation only counts the total of rings plus π bonds, not their split). Step 2: enumerate ring sizes. With 5 carbons total and exactly one ring, the ring itself can contain anywhere from 3 up to all 5 carbons, with the rest hanging off as substituents.

• Cyclopentane (5-membered ring, no substituents): all five carbons in the ring, each ring carbon bonded to two other ring carbons → every carbon is 2° (2 H's each), total $H = 10$. ✓ (Isomer A.)
• Methylcyclobutane (4-membered ring + one methyl branch): the ring carbon bearing the methyl is bonded to two ring neighbors and the branch carbon → that ring carbon is 3° (1 H); the other three ring carbons are 2° (2 H each); the branch carbon is 1° (3 H). Tally: $1+2+2+2+3 = 10$. ✓ (Isomer B.)
• 1,1-dimethylcyclopropane (3-membered ring, two methyls on the same ring carbon): that ring carbon is bonded to two ring neighbors plus two branch carbons → 4° (0 H's); the other two ring carbons are 2° (2 H each); each methyl branch is 1° (3 H each). Tally: $0+2+2+3+3=10$. ✓ (Isomer C.)
• 1,2-dimethylcyclopropane (3-membered ring, one methyl on each of two different ring carbons): the two substituted ring carbons are each 3° (1 H each: two ring bonds + one branch bond + one H); the unsubstituted ring carbon is 2° (2 H); each methyl is 1° (3 H each). Tally: $1+1+2+3+3=10$. ✓ (Isomer D.) Note 1,2-dimethylcyclopropane additionally has a cis/trans stereochemical distinction (a preview of stereoisomerism, not counted separately here since we are only enumerating constitutional isomers).
• Ethylcyclopropane (3-membered ring + one ethyl branch instead of two methyls): the substituted ring carbon is 3° (1H: two ring bonds + one branch bond); the other two ring carbons are 2° (2 H each); the branch's first CH₂ is 2° (bonded to the ring carbon and to the terminal CH₃, so 2 H's) and the branch's terminal CH₃ is 1° (3 H). Tally: $1+2+2+2+3=10$. ✓ (Isomer E.)

Step 3: total. Five distinct monocyclic constitutional isomers of C₅H₁₀ (A–E), each independently verified against the formula by direct hydrogen-count summation — the same cross-check technique used in Example 1, now applied systematically across an entire enumeration rather than a single structure.

The systematic recognition that identical formulas could correspond to structurally distinct compounds — and the insistence that structure, not just composition, determines identity — is inseparable from the work of Alexander Butlerov (1828–1886), building on foundational contributions from Archibald Scott Couper and August Kekulé on the tetravalence of carbon and its capacity for self-linkage into chains.