Reactions somebody actually ran
Experiments from public datasets, filed by the chemistry they do rather than by the paper they came from. Each record keeps its conditions, its measured yield where one was published, and a link back to the source.
This is the half of the library that is evidence. A mechanism tells you how a class is thought to work and a transformation tells you what the engine will do with your structures; a record tells you what happened when somebody mixed those things together, under conditions they wrote down.
Classes with recorded chemistry
Buchwald–Hartwig amination
Recorded reactions, each with the conditions it was run under and a link to its source.
Suzuki–Miyaura coupling
Recorded reactions, each with the conditions it was run under and a link to its source.
Reductive amination
Recorded reactions, each with the conditions it was run under and a link to its source.
Mizoroki–Heck reaction
Recorded reactions, each with the conditions it was run under and a link to its source.
Imidazole arylation
Recorded reactions, each with the conditions it was run under and a link to its source.
Buchwald–Hartwig etherification
Recorded reactions, each with the conditions it was run under and a link to its source.
Boc deprotection
Recorded reactions, each with the conditions it was run under and a link to its source.
Buchwald–Hartwig phosphination
Recorded reactions, each with the conditions it was run under and a link to its source.
Buchwald–Hartwig thioetherification
Recorded reactions, each with the conditions it was run under and a link to its source.
Palladium-catalysed enolate arylation
Recorded reactions, each with the conditions it was run under and a link to its source.
Sonogashira coupling
Recorded reactions, each with the conditions it was run under and a link to its source.
Decarboxylative cross-coupling
Recorded reactions, each with the conditions it was run under and a link to its source.
Oxidation
Recorded reactions, each with the conditions it was run under and a link to its source.
Reduction
Recorded reactions, each with the conditions it was run under and a link to its source.
Elimination
Recorded reactions, each with the conditions it was run under and a link to its source.
Esterification
Recorded reactions, each with the conditions it was run under and a link to its source.
Deprotection
Recorded reactions, each with the conditions it was run under and a link to its source.
Addition
Recorded reactions, each with the conditions it was run under and a link to its source.
Amide formation
Recorded reactions, each with the conditions it was run under and a link to its source.
Combustion
Recorded reactions, each with the conditions it was run under and a link to its source.
Acylation
Recorded reactions, each with the conditions it was run under and a link to its source.
Olefination
Recorded reactions, each with the conditions it was run under and a link to its source.
Substitution
Recorded reactions, each with the conditions it was run under and a link to its source.
What a record can and cannot tell you
A record is a statement that this was run and this came out. It is not a statement that it will work for you, that the conditions are optimal, or that the yield is reproducible. Where the source did not record something, the page says it is not recorded rather than supplying a plausible default.
Some procedures cannot be reproduced here verbatim because of the licence on the source. Those records say the procedure is withheld and link to the original, which is the honest option: a paraphrase of a procedure is a procedure nobody checked.
Looking for a specific reaction?
Browsing works when you know the class. When you know the molecules, the Reaction Lab is faster: give it your reactants and it finds records that do the same transformation, ranked by how alike the bond changes are rather than by how alike the molecules look.
Questions people ask
Can I search by my own structures instead of browsing?
Yes, and for most questions it is the better route. The Reaction Lab takes the reactants you draw or paste and finds records that perform the same transformation, including on molecules that look nothing like yours. It needs no account.
What conditions are recorded?
Whatever the source recorded: temperature, solvent, reagents, catalyst, time and measured yield where they were published. Anything the source did not state is shown as not recorded rather than filled in with a typical value.