From two stories to one decision
In the last four guides you met the two great ways a substitution can run. SN2 is the one-step shove: the nucleophile and the leaving group are both moving in the rate-determining step, so the rate depends on both concentrations, and the nucleophile comes in from the back to flip the carbon inside-out. SN1 is the two-step gamble: the leaving group departs first, all by itself, to make a flat carbocation, and only then does the nucleophile move in — so the rate depends only on the substrate, and the flat intermediate gets attacked from both faces. Two mechanisms, two rate laws, two stereochemical fates. This guide does not add a third pathway. It hands you the judgement to look at any real flask and call which of these two will actually win.
Four levers decide the outcome, and only four. They are the substrate (how crowded the carbon is, and how stable a carbocation it could make), the nucleophile (strong and eager, or weak and lazy), the leaving group (good or poor), and the solvent (does it cradle a free ion or not). The trick is that these levers are not equal, and they do not always pull the same way. Real life hands you mixed signals — a great leaving group but a feeble nucleophile, say — and the skill is knowing which lever overrules the others. Learn to weigh them and a wall of memorised reactions collapses into a handful of clear calls.
Lever one: the substrate decides first
If you only get one glance, spend it on the carbon bearing the leaving group — the substrate is the loudest lever by far. Count the carbon groups attached to it. A methyl or primary carbon (the leaving group sits on a carbon with one or zero other carbons) is wide open and lousy at holding positive charge: nucleophile can walk straight up its back, carbocation would be a disaster. So primary and methyl go SN2, full stop. A tertiary carbon (three carbon groups around it) is the mirror image: too hemmed in by steric hindrance for a backside attack to squeeze through, but its would-be carbocation is well propped up by hyperconjugation and induction from those three neighbours. So tertiary goes SN1, full stop. The whole tug-of-war lives in the middle, at the secondary carbon, which can do either and waits to hear from the other three levers.
leaving group on which carbon? -> preferred path CH3-LG methyl SN2 only R-CH2-LG primary SN2 only R2CH-LG SECONDARY SN1 or SN2 <-- the battleground R3C-LG tertiary SN1 only carbocation stability: 3 > 2 > 1 > methyl (SN1 wants 3 or stabilised 2) backside room: methyl > 1 > 2 > 3 (SN2 wants methyl or 1)
The other three levers, and how they pull
Once the substrate has narrowed things down (and especially when it sits on the secondary fence), read the remaining three. The nucleophile is the tie-breaker that matters most on secondary carbons. SN2 needs a partner that is moving in the rate-determining step, so a strong, eager nucleophile — high nucleophilicity, usually a negative charge like hydroxide, alkoxide, cyanide, azide — pushes hard for SN2. A weak, lazy nucleophile that is just sitting around (neutral water, an alcohol, the solvent itself) cannot drive a bimolecular step; it waits politely for the substrate to fall apart on its own, which is exactly SN1. As a slogan: strong nucleophile leans SN2, weak nucleophile leans SN1.
The leaving group is the one lever both paths agree on: both SN1 and SN2 need it to leave, so a good leaving group (a weak base content to carry the electrons off — iodide, bromide, tosylate, or water once an OH has been protonated) speeds up *both*. It rarely picks the winner between the two; it just decides whether substitution happens at all. A terrible leaving group like hydroxide (OH-) or fluoride basically refuses to go, which is why you protonate an alcohol with acid first — you turn a hopeless -OH into a splendid leaving group, neutral water. The solvent is the subtle one. A polar protic solvent (water, alcohols — anything with an O-H or N-H) wraps and stabilises the naked ions of SN1, so it cheers on SN1; it also cages small anionic nucleophiles and slows SN2. A polar aprotic solvent (acetone, DMSO, DMF — polar but no O-H to donate) leaves the nucleophile naked and furious, so it supercharges SN2.
Three worked calls
Reasoning beats rote, so let us actually run the scorecard on three representative substrates. Watch how the same four-lever scan settles each one — and notice that the stereochemistry falls out for free once you know the mechanism.
- 1-bromobutane (a primary substrate) + sodium cyanide (NaCN) in DMSO. Substrate: primary, so SN2 already. Nucleophile: cyanide is strong and charged — agrees, SN2. Solvent: DMSO is aprotic — agrees, SN2. Verdict: clean SN2. Stereochemistry: the attacked carbon here is not a stereocentre (it has two hydrogens), so there is nothing to invert; but if it were, expect inversion — the umbrella flips through with Walden inversion.
- 2-bromo-2-methylpropane, that is tert-butyl bromide (a tertiary substrate), warmed in plain water/ethanol with no added strong nucleophile. Substrate: tertiary, so SN2 is physically blocked — SN1 territory. Nucleophile: just the neutral solvent, weak — agrees, SN1. Solvent: water/ethanol is protic, cradles the cation — agrees, SN1. Verdict: textbook SN1 (in fact a solvolysis, the solvent doing the attacking). Stereochemistry: the flat carbocation is attacked from both faces, so a single-handed start would give a racemic, roughly 50:50, mix — it racemizes.
- 2-bromobutane (a secondary substrate) — the genuine fork in the road. With a strong charged nucleophile like sodium methoxide in DMSO, the other levers swing it to SN2 (inversion at the stereocentre). Swap to warm methanol alone — now the nucleophile is weak and neutral, the solvent is protic, and the same secondary substrate drifts toward SN1 (racemization). Same carbon, opposite outcome — proof that on a secondary substrate the conditions, not the substrate, cast the deciding vote.
Notice the pattern that makes the stereochemistry effortless: you never memorise "this product is inverted" — you read the *mechanism* and the stereochemistry follows automatically. SN2's one-step backside attack mechanically forces inversion at the stereocentre (every molecule flips), while SN1's flat-cation intermediate mechanically permits attack from either side and so scrambles handedness into a racemic mixture. This is also the most-tested trap in the whole topic, so say it out loud once more: SN2 inverts, SN1 racemizes — and both follow from the shape of the step, not from a rule you have to recall.
Honest edges — where the clean rules blur
These four levers are a genuinely powerful predictor, but they are a model, not a law of nature, and a careful learner should hold them honestly. First, real SN1 is rarely perfectly racemic. Because the leaving group has only just departed, it sometimes still shields the face it left from, so the nucleophile leans slightly toward the opposite face — you often see a little excess of inversion rather than a flawless 50:50. "SN1 racemizes" is the dominant truth and the right answer for a first course, but "mostly racemic, with a slight inversion bias" is the more exact picture. Second, the boundaries are fuzzy: a secondary substrate under in-between conditions can run partly by each path at once, giving a mix of inverted and racemized product. Nature does not read our two-column table; it just follows the lowest-energy route available, and sometimes two routes are close in energy.
Two more honest reminders worth carrying. Temperature has a quiet hand on the wheel: heat generally favours the path with more particles forming and the more disordered outcome, which is why warming a flask tends to wake up the elimination competitor more than substitution — keep that in your back pocket for the next guide. And do not over-trust "primary is always SN2": an allylic or benzylic primary carbon, whose carbocation would be stabilised by an adjacent pi system, can sneak through SN1 too. The levers are guides for thinking, not commandments. Their real value is that they force you to *reason from mechanism* — and reasoning beats memorising every single time a question changes one variable on you.
Setting up the showdown: elimination is next
Here is the twist that the whole next rung turns on, and the reason your nucleophile-versus-base instinct from the acid-base guides will pay off. A nucleophile and a base are often the same species wearing two hats. When that species attacks the carbon bearing the leaving group, you get substitution — the story you just mastered. But when it instead reaches past that carbon and rips off a hydrogen from the neighbouring carbon (a beta-hydrogen), the leaving group still departs, but now a new pi bond forms and you make an alkene. That is elimination. Substitution and elimination are run by the very same reagents, on the very same substrates, and they are always quietly competing for the same molecule.
Beautifully, the same four-lever thinking carries straight over, with the same SN2/SN1 split mirrored as a fast E2 / slow E1 split. The instinct you most need is the one already foreshadowed: a strong but *bulky* base — tert-butoxide is the poster child — is too fat to squeeze in and attack the carbon, so it gives up on substitution and grabs a proton instead, tipping the balance toward elimination. Heat does the same. So as you turn the page, carry these four levers and one extra question with you: not only "SN1 or SN2?" but "substitution or elimination at all?" The next rung is the full grudge match — and you are walking into it already holding the framework that decides it.