Skip to content
LumenaryAcademy
Organic 70 min

Reaction mechanisms

Following the electrons — curly arrows, nucleophiles and electrophiles, substitution and elimination, and why some products win the race while others win the war.

Code tested · 5AI-draftedAwaiting review by a chemist

Your guides: Chloe · The Chemist, Margaret · The Analyst, Gideon · The Sceptic

MotivationH

Why chemists draw arrows

There are more than a hundred million known organic compounds and an unlimited number of possible reactions between them. No one could memorise them all, and organic chemists do not try. Instead they learn a small number of ways in which electrons move, and reason from those to reactions they have never seen. A reaction mechanism is that reasoning written down: a step-by-step account of which bonds break, which form, in what order, and through which short-lived species.

Mechanisms are hypotheses, tested against evidence: how the rate depends on concentrations (chemical kinetics), what happens to stereocentres (stereochemistry), where isotopic labels end up, and what calculations of the energy surface say. A good mechanism explains the known facts and predicts new ones. That predictive power is why a pharmaceutical chemist can plan a route to a molecule nobody has made, and why an enzymologist can explain how a protein speeds up a reaction by a factor of a billion.

This lesson builds the core toolkit: curly arrows, the vocabulary of nucleophiles and electrophiles, the four classic mechanisms of substitution and elimination at saturated carbon, and the energy profiles and rate constants that decide which pathway wins.

DefinitionsH

Curly arrows

Definition— curly arrow

A curly arrow shows the movement of a pair of electrons. Its tail starts where the electrons are, a lone pair or a bond; its head shows where they go, either to an atom (becoming a lone pair) or to a position between two atoms (becoming a new bond). A half-headed (fish-hook) arrow shows the movement of a single electron, as in radical reactions.

The notation grew out of the electronic theory of the 1920s, with Robert Robinson and Christopher Ingold among its main developers, and it is pure bookkeeping: an arrow never shows atoms moving, and the atoms follow wherever the electrons take them. Three checks keep a mechanism honest.

  1. Arrows start at electrons. A lone pair, a π bond or a σ bond, never a positive charge or a hydrogen atom.
  2. Charge is conserved. The total charge on each side of every step must be the same.
  3. Valence is respected. After each step, no carbon may have more than four bonds and no first-row atom more than eight valence electrons. If an arrow forms a new bond to an atom that already has a full octet, another arrow must break one of its bonds.

Electrons flow from electron-rich to electron-poor sites. In the language of molecular orbitals, a curly arrow is a filled orbital of one molecule (usually its highest occupied, the HOMO) overlapping with an empty orbital of another (usually its lowest unoccupied, the LUMO), and the electron pair moving from one into the bonding combination.

Definition— nucleophile and electrophile

A nucleophile ("nucleus-lover") donates an electron pair to form a new bond to an atom other than hydrogen; it is a Lewis base. An electrophile ("electron-lover") accepts an electron pair; it is a Lewis acid. A leaving group departs from a molecule carrying the electron pair of the bond it was attached by.

Typical nucleophiles have lone pairs or negative charges: HO⁻, CH₃O⁻, I⁻, HS⁻, N₃⁻, NH₃, H₂O, CN⁻. Typical electrophiles have a polarised bond to an electronegative atom or an empty orbital: the carbon of C–Br or C=O, a carbocation, H⁺, BF₃.

Nucleophilicity is not basicity. Basicity is thermodynamic: how strongly a species holds a proton at equilibrium, measured by the pKaK_\mathrm{a} of its conjugate acid. Nucleophilicity is kinetic: how fast it attacks carbon. The two often run together, but large, polarisable nucleophiles such as I⁻ and HS⁻ are excellent nucleophiles and weak bases. The solvent matters too. In water and alcohols (polar protic solvents) small anions such as F⁻ are wrapped in hydrogen-bonded solvent and react sluggishly; in polar aprotic solvents such as dimethyl sulfoxide or propanone, which cannot donate hydrogen bonds, anions are poorly solvated and can react thousands of times faster.

A good leaving group is a weak base: it is stable once it has left with the electron pair. So the conjugate bases of strong acids leave easily. Iodide (from HI, pKaK_\mathrm{a} about −10), bromide, chloride, water (from H₃O⁺) and sulfonates (from sulfonic acids) are good leaving groups; hydroxide, alkoxides, amide ions and hydride are very poor ones. That is why an alcohol does not undergo substitution until its OH has been turned into something better, for example by protonation to –OH₂⁺ or conversion to a sulfonate ester.

IntuitionH

Substitution: SN2 and SN1

Replacing a leaving group on a saturated carbon by a nucleophile is a nucleophilic substitution. In the 1930s Edward Hughes and Christopher Ingold at University College London showed that it happens by two different mechanisms, which they named for their rate laws.

SN2: one step, from the back

In the SN2 mechanism (substitution, nucleophilic, bimolecular), the nucleophile attacks the carbon at the same moment as the leaving group departs. It must approach from the side opposite the leaving group, because that is where the C–X antibonding orbital, the LUMO, has its large lobe.

HO−CBrHHHHOCBrHHHHOCHHH+ Br⁻δ−δ−‡reactantstransition stateproducts
The SN2 reaction of hydroxide with bromomethane. One curly arrow takes a lone pair from oxygen to the back of the carbon; a second takes the C–Br bonding pair onto bromine. In the transition state (in square brackets, marked ‡) the carbon is partly bonded to both, with its three hydrogens flattened into a plane; in the product they have flipped over like an umbrella in the wind.

Three consequences follow from the single step.

  • Rate law: rate =k[RX][Nu]= k[\mathrm{RX}][\mathrm{Nu}], first order in each, because both are in the one transition state.
  • Inversion of configuration: a stereocentre is turned inside out. Paul Walden discovered such inversions in 1896, long before anyone knew why; (R)-2-bromobutane and iodide give (S)-2-iodobutane.
  • Steric sensitivity: the nucleophile must reach the back of the carbon. SN2 is fastest at methyl, fast at primary carbons, slower at secondary ones and essentially absent at tertiary carbons, where three groups block the way.

SN1: ionise first, then capture

In the SN1 mechanism (unimolecular), the leaving group departs first, leaving a carbocation, a planar sp² carbon with an empty p orbital. In a second, fast step a nucleophile, often the solvent, attacks it.

  • Rate law: rate =k[RX]= k[\mathrm{RX}], independent of the nucleophile, because only the slow first step appears in the rate.
  • Stereochemistry: the planar cation can be attacked from either face, so a stereocentre is largely racemised (often with a slight excess of inversion, because the departing anion briefly shields one face).
  • Substrate: SN1 needs a reasonably stable carbocation: tertiary, or stabilised by a neighbouring double bond (allylic), benzene ring (benzylic) or lone pair. Polar protic solvents help, by stabilising both ions as they separate.
  • Rearrangements: a carbocation can turn into a more stable one by shifting a neighbouring hydride or alkyl group, so SN1 products sometimes have a different skeleton from the starting material.

Carbocation stability runs tertiary > secondary > primary > methyl. Alkyl groups stabilise the positive centre in two ways: by hyperconjugation, in which the electrons of adjacent C–H and C–C bonds partly spill into the empty p orbital, and by a weak inductive donation. Resonance does much more: in the allyl and benzyl cations the positive charge is shared over several carbons.

The rate laws make the difference visible. Below, an SN1 scheme with a reversible ionisation is solved exactly and compared with the steady-state approximation of Chemical kinetics, which assumes the carbocation is consumed as fast as it forms. To keep the equations linear we hold the concentrations of the nucleophile and of the leaving-group anion fixed, as they are when both are in large excess (or, for the anion, at the very start of a run, before any has been released).

python
import numpy as np

# SN1:  R-X  <=>  R+ + X-   (k1 forward, km1 back)     R+ + Nu  ->  R-Nu   (k2)
# With [Nu] and [X-] held fixed, d/dt [RX, R+] = M [RX, R+] is linear, with
# M = [[-k1, km1*X], [k1, -km1*X - k2*Nu]]. Its two eigenvalues are decay rates; the slower one
# is the pseudo-first-order rate constant we would observe for the loss of R-X.
k1, km1, k2 = 1e-3, 1e9, 1e9        # 1/s, 1/(M s), 1/(M s): carbocations are captured very fast

def sn1_exact(Nu, X):
    trace = -k1 - km1 * X - k2 * Nu               # sum of the two eigenvalues
    det = k1 * k2 * Nu                            # product of the two eigenvalues
    fast = (trace - np.sqrt(trace**2 - 4 * det)) / 2
    return -det / fast                            # slow = det/fast: no subtraction of near-equal numbers

def sn1_steady_state(Nu, X):
    return k1 * k2 * Nu / (km1 * X + k2 * Nu)

k_sn2 = 1e-3                                       # an SN2 substrate for comparison, 1/(M s)
print("observed rate constants for loss of R-X, in 1/s")
print("[Nu]/M  [X-]/M   SN1 exact   SN1 steady state   SN2: k[Nu]   cation lifetime")
for Nu, X in [(0.01, 0.0), (0.1, 0.0), (1.0, 0.0), (0.1, 0.1), (0.1, 1.0)]:
    lifetime = 1 / (km1 * X + k2 * Nu)
    print(f"{Nu:5.2f}  {X:5.2f}   {sn1_exact(Nu, X):.3e}   {sn1_steady_state(Nu, X):.3e}"
          f"        {k_sn2 * Nu:.1e}      {lifetime * 1e9:5.1f} ns")
Runs entirely in your browser, in an isolated worker. Nothing is sent to a server.

Increasing the nucleophile a hundredfold leaves the SN1 rate unchanged at k1k_1, while the SN2 rate rises a hundredfold: exactly the test Hughes and Ingold used. Adding the leaving-group anion slows the SN1 reaction (the common-ion effect), because the anion recaptures the cation: with equal concentrations of X⁻ and Nu and equal capture rate constants, half the cations go back, and the rate halves. An SN2 reaction would not notice. The steady-state formula agrees with the exact solution to every printed digit, because the cation lives for only 1 to 100 nanoseconds, far shorter than the time scale of the reaction.

The program finds the eigenvalues of the 2 × 2 rate matrix from its trace and determinant rather than with a general routine such as np.linalg.eigvals, for a numerical reason: the two decay rates differ by a factor of up to 101310^{13}, and a general routine, which works to about sixteen significant figures of the large one, returns 9.084×10−59.084 \times 10^{-5} instead of 9.091×10−59.091 \times 10^{-5} for the last row (try it). Rate problems with very different time scales, called stiff, need this kind of care.

IntuitionH

Elimination: E2 and E1

A base can remove a proton from the carbon next to the leaving group instead of attacking carbon, forming a double bond: an elimination. It too has two mechanisms.

In E2 (bimolecular), the base removes the proton, the C–H electrons swing in to form the π bond and the leaving group departs, all in one step: three curly arrows, rate =k[RX][B]= k[\mathrm{RX}][\mathrm{B}]. The H and the leaving group must be anti-periplanar, on opposite sides of the C–C bond in one plane, so that the C–H bonding orbital lines up with the C–X antibonding orbital. In cyclohexanes this means both must be axial (the chair conformations of Stereochemistry), which can make one ring isomer eliminate hundreds of times faster than another.

CCHBrRRRRB−123HRRBrRRcurly arrows of E2anti-periplanar: H and Br at 180°
E2 elimination. Left: the base takes the proton (arrow 1), the C–H bonding pair becomes the new π bond (arrow 2), and the C–Br bonding pair leaves with bromine (arrow 3), all at once. R stands for any group. Right: the same molecule viewed along the C–C bond (a Newman projection, front carbon at the centre, back carbon as the circle). The H on the front carbon and the Br on the back carbon point in opposite directions, a dihedral angle of 180°: anti-periplanar.

In E1 (unimolecular), the leaving group departs first to give a carbocation, as in SN1, and a weak base then removes a neighbouring proton. E1 and SN1 share their slow step, so they usually occur together, and heating shifts the balance towards elimination: elimination turns one molecule into two or three, so it has the larger entropy gain, and −TΔS-T\Delta S grows with temperature.

When a carbocation or a substrate can lose protons from different carbons, the more substituted (more stable) alkene usually dominates: the Zaitsev rule, after Alexander Zaitsev of Kazan University (1875). A bulky base such as tert-butoxide reverses this and takes the most accessible proton, giving the less substituted Hofmann product. For 2-bromo-2-methylbutane, ethoxide gives mainly 2-methylbut-2-ene, while tert-butoxide gives mainly 2-methylbut-1-ene.

Predicting which mechanism wins

Substitution and elimination compete, and the substrate and reagent decide most cases. The table gives the usual major pathway.

SubstrateGood nucleophile, weak base (I⁻, HS⁻, N₃⁻)Strong, small base (HO⁻, CH₃O⁻)Strong, bulky base (tert-butoxide)Weak nucleophile and base (H₂O, alcohols)
methyl, CH₃–XSN2SN2SN2very slow
primary, RCH₂–XSN2SN2 (some E2)E2very slow
secondary, R₂CH–XSN2mostly E2E2slow SN1 and E1
tertiary, R₃C–XSN1 and E1E2E2SN1 and E1

Two further levers: polar aprotic solvents favour SN2 (they leave the nucleophile unsolvated), polar protic solvents favour SN1 and E1 (they stabilise the ions), and higher temperatures favour elimination.

Worked examplesH

Worked examples: predicting the pathway

The method is the same every time: classify the substrate (methyl, primary, secondary, tertiary, allylic or benzylic), classify the reagent (good nucleophile or not, strong base or not, bulky or not), note the solvent and the temperature, and only then draw arrows.

Example— Iodide and 1-chlorobutane in propanone

A primary substrate, an excellent nucleophile that is a very weak base, and a polar aprotic solvent: every factor points to SN2, and the product is 1-iodobutane. This is the Finkelstein reaction (Hans Finkelstein, 1910). On its own it would be an equilibrium, since chloride can displace iodide too, but sodium iodide dissolves in propanone and sodium chloride hardly does. The chloride drops out of solution as it forms, and the equilibrium is pulled over to the right, as in Chemical equilibrium.

Example— (R)-2-bromobutane with sodium ethoxide in warm ethanol

A secondary substrate with a strong, small base: mostly E2, with some SN2. There are two kinds of β-hydrogen. Removing one from C1 gives but-1-ene; removing one from C3 gives but-2-ene, the more substituted alkene and so the major one (Zaitsev). Of the two but-2-enes, the (E) isomer dominates: in the anti-periplanar arrangement that leads to it, the two methyl groups are also anti to each other rather than crowded side by side. The minor substitution product, 2-ethoxybutane, is formed by backside attack, and since the priorities of the groups are in the same order before and after (OEt or Br first, then ethyl, methyl, H), inversion turns (R) into (S).

Example— 2-bromo-2-methylpropane in aqueous ethanol

A tertiary substrate, only weak nucleophiles (water and ethanol), a polar protic solvent and no strong base: SN1 with some E1. The cation (CH₃)₃C⁺ is captured by water to give 2-methylpropan-2-ol, by ethanol to give 2-ethoxy-2-methylpropane, or loses a proton to give 2-methylpropene. The rate is k[RBr]k[\mathrm{RBr}] and barely changes when more water is added. Add a strong base such as ethoxide instead and the mechanism switches to E2: the rate becomes second order and 2-methylpropene becomes the main product.

Example— A rearrangement: 3-methylbutan-2-ol with hydrogen bromide

Step 1: the OH oxygen takes a proton from HBr, turning the poor leaving group HO⁻ into a good one, water. Step 2: water leaves, giving a secondary cation at C2. Step 3: a hydrogen on C3 moves across with its bonding pair (a 1,2-hydride shift), giving a tertiary cation at C3. Step 4: bromide captures that cation. The major product is therefore 2-bromo-2-methylbutane, CH₃CH₂C(CH₃)₂Br, not the 2-bromo-3-methylbutane that a "swap OH for Br" reading predicts. A changed skeleton or a moved functional group is the fingerprint of a carbocation intermediate.

DefinitionsH

Energy profiles and the Hammond postulate

A reaction energy profile plots free energy along the reaction coordinate, a measure of progress from reactants to products.

Definition— transition state and intermediate

A transition state is a maximum on the energy profile: a fleeting arrangement, with partly formed and partly broken bonds, that lasts about as long as one molecular vibration and cannot be isolated. An intermediate is a local minimum between two transition states: a real, if often short-lived, species such as a carbocation. The step with the highest transition state relative to the starting materials is rate-determining.

SN2SN1TSTS 1TS 2R⁺ X⁻intermediateR–XR–NuR–XR–NuΔG‡ΔG‡reaction coordinatereaction coordinatefree energy
Energy profiles for SN2 (left), a single step through one transition state, and SN1 (right), two steps with a carbocation intermediate between them. In the SN1 profile the first transition state is the highest, so ionisation determines the rate, and because that step is uphill its transition state lies close in energy, and in structure, to the carbocation.

Theorem— The Hammond postulate (an empirical principle)

If a transition state and an intermediate (or reactant, or product) next to it on the profile are close in energy, they are also close in structure. So the transition state of a strongly uphill step resembles its product, and that of a strongly downhill step resembles its reactant.

George Hammond stated this in 1955, and it is a principle rather than a theorem: it follows from the shape of smooth potential-energy surfaces, not from a proof. Its great use is for SN1 and E1. Ionisation is steeply uphill, so its transition state looks like the carbocation, and anything that stabilises the carbocation (substitution, resonance, a polar solvent) also lowers the barrier. That is why carbocation stability, a property of the intermediate, predicts the rate.

Barriers translate into rates through the Eyring equation from chemical kinetics, k=(kBT/h) e−ΔG‡/RTk = (k_\mathrm{B}T/h)\,e^{-\Delta G^\ddagger/RT}.

python
import numpy as np

R, kB, h = 8.314462618, 1.380649e-23, 6.62607015e-34
T = 298.15

def eyring(dG_kJ, T=T):
    """First-order rate constant (1/s) for a free energy of activation in kJ/mol."""
    return kB * T / h * np.exp(-dG_kJ * 1e3 / (R * T))

def readable(seconds):
    for unit, size in [("years", 3.156e7), ("days", 86400), ("hours", 3600), ("min", 60), ("s", 1)]:
        if seconds >= size:
            return f"{seconds / size:.3g} {unit}"
    return f"{seconds:.2g} s"

print("barrier   rate constant   half-life at 25 C")
for dG in [40, 60, 80, 90, 100, 120]:
    k = eyring(dG)
    print(f"{dG:4d} kJ/mol  {k:10.2e} /s   {readable(np.log(2) / k)}")

per_decade = R * T * np.log(10) / 1e3
print(f"\nevery {per_decade:.2f} kJ/mol of extra barrier makes a reaction 10 times slower at 25 C")
print(f"warming from 25 C to 35 C speeds up an 80 kJ/mol reaction {eyring(80, 308.15) / eyring(80):.1f} times")
Runs entirely in your browser, in an isolated worker. Nothing is sent to a server.
ComputationC

Kinetic and thermodynamic control

When one starting material can give two products, which one do we get? It depends on whether the products can turn back into the starting material.

Definition— kinetic and thermodynamic control

Under kinetic control the products do not interconvert on the timescale of the reaction, and their ratio is the ratio of their rates of formation, set by the difference in barrier heights. Under thermodynamic control the products equilibrate, and their ratio is set by the difference in their free energies, K=e−ΔG∘/RTK = e^{-\Delta G^\circ/RT}.

The kinetic product forms faster; the thermodynamic product is more stable. When they differ, temperature and time decide which we isolate: cold and short favours the kinetic product, warm and long the thermodynamic one. A classic case is the addition of hydrogen bromide to buta-1,3-diene, which goes through an allylic carbocation that bromide can capture at two carbons. At low temperature the 1,2-adduct dominates, because bromide attacks the carbon next to where the proton added, the closer and more positive site; when the mixture is warmed, the products equilibrate through the cation and the more substituted, more stable 1,4-adduct takes over. The same choice governs which enolate a ketone forms (carbonyl chemistry), and where a sulfonic acid group ends up on naphthalene.

To see the competition unfold, take a starting material A that can reversibly form B (barrier 75 kJ/mol, product 10 kJ/mol more stable than A) or C (barrier 85 kJ/mol, product 20 kJ/mol more stable). The rate equations are linear first-order ODEs, c˙=Mc\dot{\mathbf c} = M\mathbf c, solved exactly by diagonalising MM (see ordinary differential equations).

python
import numpy as np

R, kB, h = 8.314462618, 1.380649e-23, 6.62607015e-34

def eyring(dG_kJ, T):
    return kB * T / h * np.exp(-dG_kJ * 1e3 / (R * T))

# Free energies (kJ/mol) relative to the starting material A.
# B: the kinetic product (lower barrier, less stable); C: the thermodynamic product.
G_B, barrier_B = -10.0, 75.0
G_C, barrier_C = -20.0, 85.0

def fractions(T, times):
    kAB, kBA = eyring(barrier_B, T), eyring(barrier_B - G_B, T)
    kAC, kCA = eyring(barrier_C, T), eyring(barrier_C - G_C, T)
    # d[A,B,C]/dt = M [A,B,C]: a linear system, solved exactly by diagonalising M.
    M = np.array([[-kAB - kAC, kBA, kCA],
                  [kAB, -kBA, 0.0],
                  [kAC, 0.0, -kCA]])
    w, V = np.linalg.eig(M)
    c0 = np.linalg.solve(V, [1.0, 0.0, 0.0])
    return [(V @ (c0 * np.exp(w * t))).real for t in times]

times = [("1 s", 1), ("1 min", 60), ("1 hour", 3600), ("1 day", 86400), ("1 month", 2.6e6)]
for TC in [0, 60]:
    T = TC + 273.15
    print(f"at {TC} C:    A      B      C     B:C")
    for (label, t), (a, b, c) in zip(times, fractions(T, [t for _, t in times])):
        print(f"  {label:8} {a:6.3f} {b:6.3f} {c:6.3f}   {b / c:7.2f}")
    print(f"  equilibrium B:C = {np.exp((G_C - G_B) * 1e3 / (R * T)):.3f}\n")
Runs entirely in your browser, in an isolated worker. Nothing is sent to a server.

At 0 °C the kinetic product wins the race. In the first minute B and C form in the ratio of their rate constants, about 80:1, and after an hour the mixture is 96% B. But B is not trapped for ever: it slowly returns to A, and each return gives C another small chance. After a day C has reached 29%, and after a month it is 99% of the mixture. At 60 °C every barrier is crossed quickly, and within an hour the mixture has reached equilibrium, 97% C. The kinetic product wins the race; the thermodynamic product wins the war.

Further reading

How mechanisms are established, and where they matter

No one watches electrons move. Mechanisms are inferred, and modern physical organic chemistry has a well-stocked toolkit:

  • Rate laws separate unimolecular from bimolecular steps, as in the program above.
  • Stereochemistry distinguishes inversion (SN2) from racemisation (SN1), and anti from syn elimination.
  • Isotopic labelling shows which bonds break: an ¹⁸O or ¹³C label is tracked into the products by mass spectrometry or NMR.
  • Kinetic isotope effects: replacing a hydrogen by deuterium slows a step that breaks that C–H bond, by a factor of up to about seven at room temperature, because the heavier C–D bond has a lower zero-point energy. A large effect in an elimination shows that C–H breaking is in the rate-determining step, as E2 requires.
  • Linear free-energy relationships such as the Hammett equation (met in the next lesson) measure how much charge builds up in a transition state.
  • Computation: quantum-chemical methods from computational chemistry locate transition states on the potential-energy surface and estimate their energies, and are now a routine partner to experiment.

Living things use the same mechanisms. Methyltransferase enzymes move a methyl group from the cofactor S-adenosylmethionine to DNA, proteins and small molecules by an SN2 reaction, with exactly the backside attack and inversion described here; that reaction underlies the epigenetic marking of genes. Terpene synthases generate carbocations from simple precursors and steer them through rearrangements and ring closures inside a protein pocket, producing the tens of thousands of terpenes found in nature, from the menthol of mint to the antimalarial artemisinin's precursor. In industry, mechanistic understanding is how process chemists remove impurities, avoid hazardous intermediates and make reactions selective, a theme that continues in synthesis and catalysis and green chemistry.

Common misconceptionsH

Common misconceptions

Misconception

"Curly arrows show atoms moving." They show electron pairs moving. The arrow in a proton transfer goes from the base's lone pair to the hydrogen, not from the hydrogen to the base, because the electrons move towards the proton.

Misconception

"A transition state is an intermediate that lives very briefly." A transition state is an energy maximum and cannot be isolated or stored at all. An intermediate sits in an energy well, however shallow, and can sometimes be observed, trapped or even crystallised. George Olah received the 1994 Nobel Prize in Chemistry for making carbocations long-lived enough, in extremely acidic and non-nucleophilic solutions, to record their NMR spectra directly.

Misconception

"Strong bases are always good nucleophiles." Basicity is about equilibrium with a proton, nucleophilicity about rate of attack on carbon. tert-Butoxide is a strong base and a poor nucleophile; iodide is a weak base and an excellent nucleophile.

Misconception

"The more stable product always forms." Only under thermodynamic control. Many reactions are under kinetic control, and some of the most useful products in synthesis are kinetic products that would not survive equilibration.

Exercises

Exercises

1. Reaction mechanisms

Warm-upHuman

Which pair of curly arrows correctly shows the proton transfer HO⁻ + H–Cl → H₂O + Cl⁻?

Choose one

2. Reaction mechanisms

CoreHuman

(S)-2-Bromopentane is treated with sodium iodide in propanone. What is the main product?

Choose one

3. Reaction mechanisms

CoreHuman

Which of these reactions go mainly by SN1 (with some E1)? Select all that apply.

Choose all that apply

4. Reaction mechanisms

StretchHuman

In 1935 Edward Hughes and his colleagues in London dissolved optically pure 2-iodooctane in propanone with sodium iodide containing a radioactive isotope of iodine, and measured two rates: how fast the labelled iodine entered the molecules, and how fast the optical rotation fell. If every exchange is an SN2 reaction with inversion, how many times faster does the optical rotation fall than the label is taken up (the ratio of the two rate constants)?

5. Reaction mechanisms

CoreHuman

3,3-Dimethylbutan-2-ol, CH₃CH(OH)C(CH₃)₃, reacts with hydrogen bromide through a carbocation. What is the major product?

Choose one

6. Reaction mechanisms

CoreHuman

A reaction run at −78 °C (195.15 K) under kinetic control gives two products in the ratio 99 : 1. What is the difference between their free energies of activation, ΔΔG‡\Delta\Delta G^\ddagger, in kJ/mol? Give the nearest whole number. (R=8.314R = 8.314 J mol⁻¹ K⁻¹.)

7. Reaction mechanisms

CoreComputational

A primary kinetic isotope effect arises because a C–D bond has a lower zero-point energy than a C–H bond. In the simplest model the stretching vibration of the bond being broken is lost at the transition state, so the deuterated molecule faces a barrier higher by the difference in zero-point energies, 12hc(ν~CH−ν~CD)\tfrac12 hc(\tilde\nu_\text{CH} - \tilde\nu_\text{CD}), and

kHkD=exp⁡ ⁣(hc(ν~CH−ν~CD)2kBT).\frac{k_\text{H}}{k_\text{D}} = \exp\!\left(\frac{hc(\tilde\nu_\text{CH} - \tilde\nu_\text{CD})}{2k_\text{B}T}\right).

Treat each bond as a harmonic oscillator, so ν~∝1/μ\tilde\nu \propto 1/\sqrt{\mu} with reduced mass μ=mCmX/(mC+mX)\mu = m_\text{C}m_\text{X}/(m_\text{C} + m_\text{X}). Write kie(wavenumber_CH, T) returning kH/kDk_\text{H}/k_\text{D} for a C–H stretching wavenumber in cm⁻¹ and a temperature in kelvin.

Runs entirely in your browser, in an isolated worker. Nothing is sent to a server.

8. Reaction mechanisms

CoreComputational

The Hughes–Ingold test in code. Given initial rates measured at several nucleophile concentrations (substrate concentration fixed), the order in nucleophile is the slope of ln⁡(rate)\ln(\text{rate}) against ln⁡[Nu]\ln[\text{Nu}]. Write order_in(conc, rates) returning that slope (use a least-squares straight line, for example np.polyfit), and classify(conc_nu, rates) returning 'SN1' if the order is below 0.5 and 'SN2' otherwise.

Runs entirely in your browser, in an isolated worker. Nothing is sent to a server.

Further reading and resources