Planning a synthetic route

Reagents, conditions and products

A chemist asked to make a new compound almost never has a bottle of it, or of anything one step away from it. What is on the shelf is cheap and made by the tonne: ethene from a cracker, benzene from crude oil, ethanol from fermentation. Between the shelf and the target lies a short series of reactions, each one turning one functional group into another, and each one already met somewhere in the course. Planning a synthesis is choosing those reactions and putting them in order.

The reactions are easiest to hold as a map, with each functional group a box and each reaction an arrow between two boxes, labelled with what makes the change happen.

Describe the drawing

A reaction map of the aliphatic functional groups drawn as two columns of boxes joined by labelled arrows, each with its reagent above and its conditions below. Halogenoalkane on the left links up to nitrile and primary amine, across to alkane and alkene, and down to the alcohols. The primary and secondary alcohols, framed together, are oxidised by acidified dichromate to aldehyde, ketone and carboxylic acid in amber arrows and regained by NaBH₄ in cyan arrows. Coral arrows tagged plus one carbon run from halogenoalkane to nitrile and from aldehyde and ketone to hydroxynitrile. At the bottom the carboxylic acid forms an ester and the acyl chloride gives the acid, the ester and the amide at room temperature.

Every step carries three pieces of information, and a route answer needs all three for every step. The reagent is the substance added: aqueous sodium hydroxide, potassium cyanide, acidified potassium dichromate(VI). The conditions are what else the reaction needs to go the right way: heating under reflux, a solvent such as ethanol, a catalyst, a temperature. The product is the compound the step makes, given by name or by structure.

Conditions matter because the same reagent can give two products. Warmed with 2-bromopropane, aqueous potassium hydroxide substitutes the bromine to give propan-2-ol; dissolved in ethanol and heated, it eliminates HBr to give propene. Acidified dichromate oxidises propan-1-ol to propanal if the aldehyde is distilled off as it forms, and on to propanoic acid if the mixture is heated under reflux with the oxidising agent in excess. Write aqueous or ethanolic, distil or reflux every time, because a scheme gives the mark for that word.

Reagents and conditions table

Every conversion in the aliphatic course, with the reagent and the conditions a scheme credits for it.

Those seventeen conversions are the whole aliphatic course, and a route needs no others. Alkenes and halogenoalkanes are the usual ways in from the hydrocarbons, and a halogenoalkane is the most useful intermediate of all: hydroxide, cyanide and ammonia turn it into an alcohol, a nitrile and an amine, and hot ethanolic hydroxide turns it into an alkene.

Synthetic route

A sequence of reactions that converts a starting material into a target compound, with the reagents and conditions for each step.

Reagent

The substance added to bring about a reaction, named in full or by a correct formula.

Conditions

The temperature, solvent, catalyst and method of heating a reaction needs to give the intended product.

Intermediate

A compound made in one step of a synthesis and used as the starting material for the next.

Reactions from benzene

Benzene has fewer reactions, and they all start in the same way. Its delocalised ring does not undergo addition under ordinary conditions, so the first step from benzene is always an electrophilic substitution: a hydrogen on the ring is swapped for a nitro group or an acyl group. What follows is ordinary chemistry of the group that has been attached.

Nitration uses concentrated nitric acid with concentrated sulfuric acid at about 50 °C; above 55 °C a second nitro group starts to go on. Tin and concentrated hydrochloric acid then reduce the nitro group to an amine. In the acid the product is the phenylammonium ion, so excess sodium hydroxide is added afterwards to release phenylamine itself. Phenylamine is the start of the azo dyes, and with ethanoyl chloride it gives N-phenylethanamide.

Friedel–Crafts acylation is the only reaction in the course that puts carbon on a ring. An acyl chloride with an aluminium chloride catalyst, heated under reflux with no water present, puts an acyl group on the ring and gives a phenyl ketone: ethanoyl chloride gives phenylethanone, C₆H₅COCH₃. From there the ketone behaves like any other ketone, and sodium tetrahydridoborate reduces it to the secondary alcohol 1-phenylethanol.

Describe the drawing

A reaction map starting from benzene, drawn as a hexagon with a circle on the left, with two arrows leaving it, bracketed together as electrophilic substitution. The upper arrow, concentrated nitric and sulfuric acids at 50 °C, gives nitrobenzene; tin and concentrated hydrochloric acid under reflux, then sodium hydroxide, reduce it to phenylamine; ethanoyl chloride at room temperature turns that into N-phenylethanamide, with its C=O drawn. The lower arrow, ethanoyl chloride with aluminium chloride, anhydrous under reflux, gives phenylethanone, which NaBH₄ in water or ethanol reduces to 1-phenylethanol.

C₆H₆(l) + HNO₃(l) → C₆H₅NO₂(l) + H₂O(l)Nitration, with concentrated H₂SO₄ as the catalyst that generates the NO₂⁺ electrophile.

C₆H₅NO₂(l) + 6[H] → C₆H₅NH₂(l) + 2H₂O(l)Reduction, with [H] standing for the reducing agent, tin and concentrated hydrochloric acid.

Designing a route

A route starts from a comparison of the starting material with the target on two counts. The first is the carbon skeleton: count the carbons, and note whether the chain is straight or branched and whether there is a ring. The second is the functional groups: which one the starting material has, which one the target needs, and on which carbon each sits. If the skeleton is the same, the route only moves and changes functional groups. If the target has one more carbon, one step must make a carbon–carbon bond.

Then work backwards from the target. Ask what it could be made from in a single step, then what that could be made from, until the chain of answers reaches the starting material. A primary amine comes from a nitrile or a halogenoalkane. A carboxylic acid comes from a primary alcohol or an aldehyde. An ester comes from an acid and an alcohol, or from an acyl chloride and an alcohol. Two to four steps is the length of an A-level route, so the search never has to go far.

Only two reactions in the course add a carbon to a chain, and both use cyanide: substitution of a halogenoalkane gives a nitrile, and addition to an aldehyde or ketone, followed by dilute acid, gives a hydroxynitrile. Friedel–Crafts acylation adds carbon to a ring. Nothing in a planned route shortens a chain: cracking gives a mixture, and hydrolysis only splits an ester or amide back into the halves it was made from.

CH₂=CH₂(g) + HBr(g) → CH₃CH₂Br(l)Step 1, electrophilic addition.

CH₃CH₂Br(l) + CN⁻(aq) → CH₃CH₂CN(l) + Br⁻(aq)Step 2, nucleophilic substitution. The new C–C bond forms here.

CH₃CH₂CN(l) + 2H₂(g) → CH₃CH₂CH₂NH₂(l)Step 3, reduction over a nickel catalyst.

Describe the drawing

A flow chart from top to bottom of four displayed formulae joined by three arrows: ethene, bromoethane, propanenitrile and propylamine, with carbon counts 2, 2, 3 and 3. Step 1, HBr at room temperature, electrophilic addition. Step 2, KCN in ethanol and water heated under reflux, nucleophilic substitution. Step 3, hydrogen with a nickel catalyst or LiAlH₄ in dry ether, reduction. The carbon that arrives with the cyanide is drawn in red in the nitrile and again in propylamine, where it is the CH₂ bonded to nitrogen.

CH₃CH₂OH(l) + 2[O] → CH₃COOH(l) + H₂O(l)Step 2, with [O] standing for the oxidising agent, acidified potassium dichromate(VI).

CH₃COOH(l) + CH₃CH₂OH(l) ⇌ CH₃COOCH₂CH₃(l) + H₂O(l)Step 3, catalysed by concentrated H₂SO₄.

Retrosynthesis

Planning a synthesis by working backwards from the target, asking at each stage which compound could give it in a single step.

Choosing between routes

When more than one route leads to the target, the first thing to weigh is the number of steps, because no step gives all the product it could, and the losses multiply. A route with three steps that give 80%, 75% and 60% has an overall yield of 0.80 × 0.75 × 0.60 = 0.36, or 36%. A two-step route at 70% each gives 0.70 × 0.70 = 0.49, or 49%. Fewer steps usually means more product, as well as less time, solvent and waste.

The second is atom economy: how much of the mass of the reactants ends up in the product rather than in a by-product. An addition reaction puts every atom into one product and has an atom economy of 100%. Hydrating ethene with steam makes ethanol at 100%. Making ethanol from bromoethane and sodium hydroxide also produces sodium bromide, so the atom economy is 46.0 ÷ (108.9 + 40.0) × 100 = 30.9%. More than two thirds of the mass bought becomes waste.

The third is hazard. Potassium cyanide is acutely toxic, and with acid it releases hydrogen cyanide gas. LiAlH₄ reacts violently with water. Concentrated acids, bromine and acyl chlorides are corrosive. Where a safer reagent does the same job it is preferred: NaBH₄ reduces aldehydes and ketones and is stable enough to use in water, and ethanoic anhydride replaces ethanoyl chloride in making aspirin because it reacts less violently and gives ethanoic acid rather than hydrogen chloride.

Green chemistry adds a further aim: a process that needs no solvent. A solvent appears nowhere in the equation, so it never lowers the atom economy, yet it is bought, heated, separated from the product and then recovered or thrown away, which costs energy and makes waste. Most organic solvents are flammable and many are toxic. Hydrating ethene needs none, because gases pass over a solid catalyst; the cyanide step needs ethanol and water.

The fourth is whether the route gives one product or a mixture. Propylamine could be made in one step from 1-bromopropane and ammonia, but propylamine has a lone pair on its nitrogen and is itself a nucleophile. It attacks unreacted 1-bromopropane, and the flask ends up holding primary, secondary and tertiary amines and a quaternary ammonium salt. Excess ammonia reduces this but does not stop it. The nitrile route can only give the primary amine.

Chirality is the fifth. Adding cyanide to an aldehyde other than methanal creates a chiral centre, and attack on either face of the planar carbonyl is equally likely, so the product is a racemic mixture. If the target is a drug that must be a single enantiomer, the route needs a separation or a chiral catalyst, and both add cost.

Overall yield

The yield of a route of several steps, found by multiplying the fractional yields of the individual steps together.

Atom economy

The mass of the desired product divided by the total mass of the reactants, multiplied by 100, from the balanced equation.

Hazard

A property of a substance or procedure that could cause harm, such as toxicity, flammability or corrosiveness.

Risk

The chance that a hazard causes harm as the substance is actually used, which precautions such as a fume cupboard reduce.

Racemic mixture

An equimolar mixture of the two enantiomers of a chiral compound, which has no overall effect on plane-polarised light.

How a route question is marked

A question that asks for a synthesis is marked step by step. A scheme for a three-step route typically has a mark for each intermediate and a mark for the reagents and conditions of each step, six in all. An intermediate can be given by name or by a structure, as long as either is unambiguous: propanenitrile or CH₃CH₂CN, not the nitrile. A mechanism is not part of the answer unless the question asks for one.

Reagents have to be named well enough to be the right substance. Dichromate alone is not acidified potassium dichromate(VI); KCN is not HCN; [O] and [H] are shorthand in equations, never reagents. Conditions are wanted wherever they decide the product: aqueous or ethanolic hydroxide, distil or reflux, excess ammonia, 50 °C for nitration, no water for Friedel–Crafts.

A route also has to be chemically possible in the order given. Reducing a ketone and then trying to add cyanide to the alcohol that results fails, because cyanide adds to C=O, not to C–OH. Reading the route forwards, checking that each product is what the next reagent acts on, catches that kind of error.

Other boards: more reactions and Grignard reagents

OCR, Edexcel and Cambridge each add reactions to the course, and a route set on their papers can use them. The one that matters most is the hydrolysis of a nitrile, which all three examine. Heated under reflux with dilute hydrochloric or sulfuric acid, the C≡N becomes COOH and the nitrogen leaves as ammonium ions. After the cyanide step it gives an acid with one more carbon, so bromoethane becomes propanoic acid in two steps.

CH₃CH₂CN(l) + 2H₂O(l) + H⁺(aq) → CH₃CH₂COOH(aq) + NH₄⁺(aq)Acid hydrolysis of propanenitrile, heated under reflux.

The same three boards turn a carboxylic acid into an acyl chloride: with thionyl chloride, SOCl₂, for OCR and Cambridge, and with phosphorus(V) chloride, PCl₅, for Edexcel and Cambridge. Thionyl chloride is the cleaner choice, because both of its by-products are gases that leave the flask. The acyl chloride reacts with an alcohol at room temperature and goes to completion, so ethyl ethanoate made from ethanoyl chloride escapes the equilibrium that caps an esterification.

CH₃COOH(l) + SOCl₂(l) → CH₃COCl(l) + SO₂(g) + HCl(g)Ethanoyl chloride from ethanoic acid with thionyl chloride.

All three also take benzene further, with two more electrophilic substitutions that need a halogen carrier and no water. Bromine with iron(III) bromide gives bromobenzene. A halogenoalkane with aluminium chloride gives an alkylbenzene: this is Friedel–Crafts alkylation, which adds carbon to a ring as acylation does but tends to add a second and third alkyl group, because the alkylbenzene reacts faster than benzene.

Edexcel adds a third way to make a carbon–carbon bond, the Grignard reagent, an organomagnesium halide RMgX. It is made by heating a halogenoalkane under reflux with magnesium turnings in dry ether (ethoxyethane). Bromoethane gives ethylmagnesium bromide, CH₃CH₂MgBr.

CH₃CH₂Br(l) + Mg(s) → CH₃CH₂MgBrIn dry ether under reflux. The Grignard reagent stays dissolved in the ether, so it is written without a state symbol.

Magnesium is far less electronegative than carbon, so the carbon bonded to it is δ− and acts as a nucleophile: it attacks the δ+ carbon of a C=O group, or of carbon dioxide, and forms a new C–C bond, and dilute acid then turns the O–MgBr into O–H. Everything is kept dry, because water protonates that carbon and turns the reagent into an alkane. With solid carbon dioxide, then dilute acid, the product is a carboxylic acid with one more carbon; with an aldehyde or ketone it is an alcohol whose skeleton is the two pieces joined at the old carbonyl carbon, primary from methanal, secondary from another aldehyde, tertiary from a ketone.

Grignard reagent

An organomagnesium halide, RMgX, made from a halogenoalkane and magnesium in dry ether, whose carbon bonded to magnesium acts as a nucleophile and forms a new C–C bond.