Steric Strain and Gauche Interactions

Steric strain accounting is the invisible arithmetic behind nearly every prediction of stereochemical outcome in modern chemistry. In drug design, the preferred bioactive conformation of a flexible pharmacophore is routinely estimated by summing gauche-type interactions across every rotatable bond, exactly as in the worked examples above, before any expensive computation is run — medicinal chemists use these back-of-envelope strain counts to decide which analogs are worth synthesizing. In polymer science, the persistence length and crystallinity of polyethylene chains are controlled by the same anti-vs-gauche energetics: polyethylene chains overwhelmingly adopt the all-anti zig-zag backbone for exactly the reason worked out in Example 3, and the small residual gauche population sets the amorphous-region fraction that governs a plastic's flexibility and melting behavior. In force-field-based molecular modeling (MM2, MM3, MMFF, AMBER, CHARMM), the Lennard-Jones parameters $\varepsilon$ and $\sigma$ introduced here are literally tabulated per atom type and used to compute the steric-energy term for every conformer of every molecule simulated, from small-molecule drug candidates to full proteins. And directly downstream in the stereochemistry curriculum, this exact gauche-interaction bookkeeping becomes the quantitative tool for ranking chair conformers in Cyclohexane Chair Conformation, for assigning A-values in Axial and Equatorial Positions, and for rationalizing why equatorial substituents are almost always preferred — each axial substituent reintroduces a pair of 1,3-diaxial gauche-like interactions strikingly analogous to the CH₃–CH₃ clash in gauche-butane itself.

Ask the tutor: How do I use gauche-interaction counting to predict which chair conformer of a disubstituted cyclohexane is more stable?

Every atom behaves, for non-bonding purposes, as if it occupies a soft sphere of a characteristic van der Waals radius: about 1.20 Å for hydrogen, 1.70 Å for carbon (as a methyl group), 1.52 Å for oxygen, 1.55 Å for nitrogen. Two non-bonded atoms feel essentially nothing when they are far apart, a weak attraction (London dispersion, induced-dipole/induced-dipole) as they approach the sum of their radii, and then a violently steep repulsion the instant that sum is breached. This last regime — repulsion at "contact plus a little closer" — is what chemists mean by steric strain. The attraction at intermediate range comes from correlated electron fluctuations: an instantaneous dipole on one atom polarizes the other, and the resulting induced-dipole/induced-dipole attraction scales as $r^{-6}$. The repulsion at short range is fundamentally different in origin — it is not electrostatic at all in the leading term, but a consequence of the Pauli exclusion principle. When the filled valence orbitals of two atoms begin to overlap, electrons of the same spin must avoid each other by occupying higher-energy antibonding-like combinations; the total electronic energy rises steeply, faster than any power-law attraction can cancel. This repulsive wall is often modeled empirically as $r^{-12}$ (chosen for mathematical convenience — it is the square of the $r^{-6}$ attractive term — not because it is derived from first principles), giving the classic Lennard-Jones potential. Gauche-butane is the perfect minimal system to see this. Anti-butane (dihedral = 180°) places the two methyl groups as far apart as the fixed bond angles allow — no steric penalty, only the baseline torsional (bond-eclipsing) terms present in any staggered conformation. Rotate 120° to the other staggered well and you reach gauche-butane (dihedral = 60°), where the methyl-methyl through-space distance shrinks to roughly 3.1 Å — comfortably inside the sum of two methyl van der Waals radii (~2 × 2.0 Å ≈ 4.0 Å, using the effective radius of a rotating methyl group), so the methyls genuinely feel each other. Push on another 60° to full eclipse (dihedral = 0°) and the methyl-methyl distance collapses further, producing the single highest point on the whole rotational profile.

The green dashed line in the figure marks the actual through-space methyl-methyl distance realized in gauche-butane: it sits well up on the repulsive wall, left of the energy minimum, which is exactly why gauche-butane is destabilized relative to anti-butane rather than stabilized. Steric strain of this kind is strictly a through-space, non-bonded effect — it must be distinguished from torsional strain (also called eclipsing strain), which arises even between small atoms like hydrogens directly across an eclipsed bond, from filled-σ/antibonding-σ* (or filled/filled) orbital interactions along the bond axis rather than genuine van der Waals crowding. Gauche-butane's 0.9 kcal/mol penalty is essentially pure steric strain (the two conformers being compared are both staggered, so torsional strain is approximately constant between them); eclipsed butane's much larger ~4.5–6.3 kcal/mol penalty (relative to anti) is a mixture of steric strain (severe, since the methyls are now much closer) plus intrinsic torsional strain from the eclipsed C–H/C–H and C–H/C–CH₃ bonds.

The total non-bonded interaction energy between two atoms is modeled by the Lennard-Jones 6-12 potential:

where:

• $r$ is the through-space distance between the two non-bonded atoms (Å),
• $\varepsilon$ is the depth of the attractive well (kcal/mol), atom-pair specific,
• $\sigma$ is the distance at which $E_{\text{vdW}} = 0$ (related to, but slightly smaller than, the sum of van der Waals radii),
• the $r^{-12}$ term is the steep repulsive wall, the $r^{-6}$ term the attractive dispersion tail.

The energy minimum occurs at $r_{\min} = 2^{1/6}\sigma \approx 1.122\sigma$, where $E_{\text{vdW}} = -\varepsilon$. For butane's rotational profile, the total torsional energy as a function of the C2–C3 dihedral angle $\phi$ is well described (to lowest order) by a truncated Fourier expansion:

where:

• $V_3$ (the three-fold term) captures the intrinsic, near-constant eclipsing/torsional barrier common to all C–C single bonds (~2.9–3.0 kcal/mol per fully eclipsed conformation, as in ethane),
• $V_1$ and $V_2$ encode the extra, conformation-specific steric and electronic asymmetry (methyl vs. hydrogen substituents) that makes gauche different from anti and makes the two eclipsed forms different from each other,
• $\phi = 180°$ is anti, $\phi = \pm60°$ are the two gauche wells, $\phi = 0°$ and $\phi = \pm120°$ are the two distinct eclipsed maxima.

The population ratio between conformers follows the Boltzmann distribution:

where:

• $\Delta G^{\circ} = G^{\circ}_{\text{gauche}} - G^{\circ}_{\text{anti}}$ is the free-energy difference (kcal/mol),
• $R = 1.987 \times 10^{-3}$ kcal/(mol·K),
• $T$ is the absolute temperature (K).

Example 1 (Foundational): Populate butane's conformers at physiological temperature. Find the anti and total-gauche populations of butane at $T = 310$ K (body temperature), given $\Delta G^{\circ}_{\text{gauche-anti}} = 0.9$ kcal/mol. Compute $RT = (1.987\times10^{-3})(310) = 0.6160$ kcal/mol.

Each gauche well: 23.2% of anti's population. Both wells together: $2 \times 0.2320 = 0.4640$.

So at 310 K the mixture shifts very slightly toward more gauche population (31.7% vs. 30.4% at 298 K) — exactly as expected, since raising temperature always pushes populations toward statistical (energy-independent) parity, here 25:50 in the high-T limit.

Example 2 (Applied): Estimating a steric A-value analog — 2,3-dimethylbutane vs. butane. 2,3-Dimethylbutane, $(CH_3)_2CH$–$CH(CH_3)_2$, has a central C2–C3 bond with two methyl groups on each carbon. In the anti-like staggered conformation that minimizes strain, count the number of gauche-type methyl–methyl interactions and estimate the added rotational strain relative to a hypothetical strain-free reference. Label C2's substituents as CH₃(a), CH₃(b), H, and C3's substituents as CH₃(c), CH₃(d), H. In the lowest-energy staggered conformation, place the two hydrogens anti to each other (this is the conformation analogous to butane's "anti," now generalized since there is no single unique anti pair of methyls possible — every staggered conformation here necessarily puts some methyls gauche to each other). Working through the Newman projection: with H(C2) anti to H(C3), the remaining four methyls are distributed so that CH₃(a) is gauche to both CH₃(c) and CH₃(d) is gauche to CH₃(a)... more carefully, each staggered arrangement places every substituent on C2 gauche to two of the three substituents on C3 (since gauche means 60° away, and each carbon has three positions 120° apart, two of which are 60° from any given group on the other carbon, one of which — the anti one — is 180° away). Concretely, in the best staggered conformer (H anti to H), each of the two methyls on C2 is gauche to each of the two methyls on C3, giving

Using the calibration that one such interaction costs ≈0.9 kcal/mol (the same steric penalty as in gauche-butane, since the local geometry — a methyl 60° from another methyl across a C–C bond — is essentially identical),

above a strain-free reference. This estimate (obtained by simple additive counting of gauche interactions, the same method used throughout conformational analysis) is close to experimentally/computationally derived steric-energy estimates for 2,3-dimethylbutane's ground conformer, and it demonstrates the generalizable counting rule: total gauche steric strain ≈ (0.9 kcal/mol) × (number of gauche substituent–substituent pairs beyond hydrogen–hydrogen), a rule used constantly when comparing chair conformers of substituted cyclohexanes.

Example 3 (Exam/competition level): Rank three staggered conformers of 3-methylpentane by strain energy and predict the ground state. 3-Methylpentane is $CH_3CH_2$–$CH(CH_3)$–$CH_2CH_3$; focus on rotation about the C3–C4 bond, where C3 bears (CH₃, H, and the C2 ethyl-chain carbon) and C4 bears (H, H, and the C5 ethyl-chain carbon) — but to make this a genuinely three-substituent-vs-three-substituent problem, consider instead rotation about the C2–C3 bond, where C2 = CH₃CH₂– (bearing H, H, and CH₃ as its three "arms" viewed from C3) and C3 bears CH₃, H, and CH₂CH₃ (the C4–C5 ethyl chain). Enumerate the three staggered conformers (dihedral 60°, 180°, 300°) by which two "large" groups (the ethyl on C2, the ethyl-chain on C3, and the methyl on C3) end up anti or gauche to each other, and rank them. Define on C2 (viewed toward C3): substituents are H, H, CH₃ (only the terminal methyl of the ethyl group projects backward from C2; C2 itself only has one non-H, non-C3 substituent, namely its own CH₃... to keep this concrete, treat C2's three substituents as {H, H, CH₃-terminal}). On C3: substituents are {CH₃ (the branch), H, CH₂CH₃ (continuing chain, call it Et)}. The three staggered conformers place the large groups (CH₃ on C2's side; CH₃-branch and Et-chain on C3's side) as follows:

• Conformer I: CH₃(from C2) anti to Et(chain continuation on C3). Then CH₃(from C2) is gauche to both CH₃-branch and H on C3; meanwhile the two hydrogens on C2 are gauche to the remaining groups. Count large–large gauche pairs: CH₃(C2)···CH₃-branch(C3) = 1 gauche large-large interaction (the anti pair, C2-methyl/Et, contributes zero strain by definition of anti).
• Conformer II: CH₃(from C2) anti to CH₃-branch (on C3). Then CH₃(from C2) is gauche to Et(chain) — a bulkier group than a methyl, so this gauche interaction, while still ≈0.9 kcal/mol at the carbon-carbon level (the immediate steric clash is still methyl-like, since it's the first CH₂ of the ethyl group that clashes, not the whole chain), is essentially the same order of magnitude. Count: one gauche interaction between CH₃(C2) and CH₂CH₃(C3).
• Conformer III: H(from C2) anti to Et(chain), leaving CH₃(from C2) gauche to both CH₃-branch and, separately, one of C3's remaining positions. This conformer forces the bulky CH₃(C2) to be gauche to CH₃-branch and does not relieve the Et group either, which ends up gauche to one of C2's hydrogens (cheap) — but critically, in this arrangement the two largest groups overall, if not carefully anti-positioned, can create a second, more severe steric clash if the geometry forces CH₃-branch and Et(chain) — both on C3 — to remain gauche to bulky groups on C2 simultaneously.

Careful enumeration (as always done via a full Newman projection rather than verbal description) shows that Conformer I (methyl on C2 anti to the ethyl-chain continuation on C3, i.e., the two chain-extending groups anti to each other) is the global minimum, because it is the conformation chemists universally draw as the "extended zig-zag" backbone — the one where the main carbon chain is anti at every internal bond. This reflects a general and testable principle: the lowest-energy conformation of any acyclic chain places the backbone carbons anti to each other at every rotatable bond, pushing branch substituents (here, the methyl branch) into gauche positions relative to backbone hydrogens rather than into gauche positions relative to other backbone carbons. This is exactly the reasoning used to justify why n-alkanes adopt the all-anti zig-zag as their global energy minimum, and it is the direct ancestor of the reasoning used later for equatorial preference in cyclohexanes: bulky groups avoid gauche relationships with other bulky groups, preferring gauche relationships with hydrogen wherever a choice exists.

The quantitative marriage of steric strain to conformational analysis is inseparable from Derek H. R. Barton, who in 1950 argued that molecular reactivity and stability in steroids and other cyclic systems could be predicted from conformational preferences — axial versus equatorial substitution, chair-chair interconversion — built directly on exactly this kind of non-bonded, van der Waals reasoning that had first been worked out quantitatively for simple open-chain systems like butane. Barton's insight, for which he shared the 1969 Nobel Prize in Chemistry with Odd Hassel, was that conformation is not a minor structural footnote but a controlling factor in chemical reactivity — a claim that only makes sense once one accepts that steric strain differences of under 1 kcal/mol per gauche interaction can accumulate into decisive, multi-kilocalorie preferences that dictate which face of a molecule reacts.