Retrosynthesis: Planning Organic Synthesis Backward

Retrosynthesis is the art of planning an organic synthesis backwards. Instead of asking what these starting materials will make, you start from the target molecule and repeatedly break bonds, called disconnections, to reach simpler precursors, until you arrive at cheap, available starting materials. Each backward step is a transform, the reverse of a known, reliable forward reaction. The method was formalised by E. J. Corey, who won the 1990 Nobel Prize in Chemistry for it.

Forward trial-and-error fails beyond simple molecules because the number of possible routes explodes. Retrosynthesis turns synthesis into a design process: you compare routes for step count, yield, cost and safety before touching a flask. Pharmaceutical companies use it to find the cheapest manufacturing route to every new drug, and examiners love it in JEE Advanced, GATE and CSIR-NET.

Key Concepts

• Disconnection: a backward bond-breaking step drawn with an open arrow. Good disconnections cut bonds that are easy to make forward, e.g. C-C bonds next to carbonyls (aldol) or C-O bonds of esters.
• Synthon: an idealised fragment, often charged like an enolate, produced by a disconnection; it may not be a real molecule, but it points to a real reagent.
• FGI (functional group interconversion): changing one group into another backward, e.g. an amine back to an amide, to set up a better disconnection.
• Forward equivalents: each transform must match a real named reaction with known conditions: Grignard, aldol, Wittig, Diels-Alder, Friedel-Crafts.
• Convergence: the best routes join two large fragments late rather than chaining many linear steps, because yields multiply along a chain.
• Starting materials: the analysis stops when every precursor is commercially available and inexpensive.

How It Works

1. Write the target and identify strategic bonds: polar bonds, branch points, ring bonds.
2. Disconnect one strategic bond, generating two synthons.
3. Convert each synthon into a real reagent, e.g. an enolate synthon becomes a ketone plus strong base.
4. Check the forward reaction is known, high-yielding and selective.
5. Repeat on each precursor until all are commercial starting materials.
6. Write the forward scheme, estimate overall yield and cost, and pick the shortest reliable route.

Worked Example

A planned 3-step route has step yields of 90%, 75% and 80%.

1. Overall yield = 0.90 x 0.75 x 0.80 = 0.540, i.e. 54.0%.
2. Raise the weakest step from 75% to 90%: 0.90 x 0.90 x 0.80 = 0.648, i.e. 64.8%, a gain of 10.8 percentage points.
3. Raise the best step from 90% to 95% instead: 0.95 x 0.75 x 0.80 = 0.570, i.e. 57.0%, a gain of only 3.0 points.
4. Conclusion: the lowest-yielding step dominates the route, so redesign the bottleneck first.

Common Misconceptions

• Any bond can be disconnected. A disconnection is useful only if the forward reaction exists and is reliable; breaking a bond with no known way to remake it is fantasy, not planning.
• Fewer steps always win. A 2-step route with an unknown key step loses to a 4-step route built on textbook reactions; reliability beats brevity.
• There is one correct answer. Most targets have several valid routes; the skill is comparing them on yield, cost, safety and step count.

Real-World Applications

• Pharmaceuticals: manufacturing routes for drugs like atorvastatin were designed retrosynthetically.
• Agrochemicals: at ton scale, the cheapest retrosynthetic route wins the market.
• Academia: total syntheses of natural products showcase new transforms.
• Green chemistry: route comparison finds processes with fewer toxic reagents and less waste.

Retrosynthesis turns how do I make this into small, answerable questions: which bond, which transform, which reagent. Break bonds backward from reliable reactions, compare whole routes, and always attack the lowest-yielding step first.