Sorting reactions you have already watched, as you did in Sorting Five Reactions, is history. Prediction is the harder and more interesting direction: here are the reactants, nobody has run this, say what comes out β€” and then check.

That is the actual work of a chemist, and the reason Types of Chemical Reactions was worth learning. The classification is not a filing system. It is a set of instructions for what to do next.

Identify the type first

Every prediction starts by asking what shape the reactants are in.

graph TD
    A["Look at the reactants"] --> B{"How many, and what kind?"}
    B -->|"two elements"| C["Synthesis"]
    B -->|"one compound, alone"| D["Decomposition"]
    B -->|"an element and a compound"| E["Single displacement<br/>go to the activity series"]
    B -->|"two ionic compounds in solution"| F["Double displacement<br/>go to the solubility table"]
    B -->|"a fuel and oxygen"| G["Combustion"]

Notice that two of the five branches send you to a reference table. That is not a weakness in the method β€” it is the method. Nobody predicts a single displacement from first principles; you look up which element is more reactive, because somebody measured it.

Synthesis and decomposition

Synthesis joins things. A metal and a non-metal make an ionic compound, and the charges tell you the formula β€” that part is just Naming and Formulas run forwards. Two non-metals make a molecular compound, and here the ratio is genuinely not predictable from first principles, which is why sulfur and oxygen can give you either or depending on conditions.

The synthesis reactions worth memorising are the oxides with water, because they explain something you can taste and measure:

A metal oxide plus water gives a hydroxide β€” a base. A non-metal oxide plus water gives an acid. That single pair of rules is most of Acids and Bases, and it is what you were testing in Oxides and Neutralisation.

Decomposition is synthesis backwards and usually needs energy put in β€” heat, electricity, or light. Four patterns cover nearly everything you will be asked:

  • A binary compound splits into its elements:
  • A metal carbonate gives the metal oxide and carbon dioxide:
  • A metal hydroxide gives the metal oxide and water:
  • A hydrate gives the anhydrous compound and water, which is the reaction you will exploit in Molar Mass and Composition.

The carbonate one is the reaction that turns limestone into the lime used in cement, and it is a large fraction of the carbon dioxide the construction industry releases β€” the gas comes out of the rock itself, not only out of the fuel used to heat it. That is worth raising in Chemistry at Industrial Scale.

Single displacement needs the activity series

An element goes in, a different element comes out. Whether it happens at all depends on which of the two holds its electrons more loosely, and that ordering is the activity series you built by experiment in Building the Activity Series.

The rule: a more reactive element displaces a less reactive one from its compound. A less reactive element does nothing at all.

Magnesium sits above hydrogen, so it pushes hydrogen out of the acid and the tube fizzes. Copper sits below hydrogen, so copper in hydrochloric acid gives you copper sitting in hydrochloric acid. Not a slow reaction β€” no reaction.

Halogens have their own series, in plain periodic-table order: fluorine, chlorine, bromine, iodine, most reactive first. A halogen displaces any halide below it:

and bromine in sodium chloride solution does nothing. Both series are laid out in The Activity Series.

Double displacement needs the solubility table

Two ionic compounds in solution swap partners. But swapping is not by itself a reaction: the ions were already dissolved and separate, and after the swap they can be dissolved and separate again. Nothing has happened unless something leaves the solution.

There are exactly three ways for something to leave.

A precipitate forms. Check both possible products against Solubility Rules. If either one is insoluble, it falls out as a solid and the reaction goes.

A gas escapes. A carbonate with an acid is the usual case. The carbonic acid you would predict is unstable and decomposes as fast as it forms, which is why the products look like three things instead of two.

Water forms. This is neutralisation, and the water molecule locks up an and an where neither can act as an ion any longer.

Naming which of the three is happening is the difference between explaining a reaction and labelling it. The full machinery for writing what actually changed is in Precipitation and Net Ionic Equations.

When the prediction fails

Mix sodium chloride solution with potassium nitrate solution. Swap the partners: potassium chloride and sodium nitrate. Look both up β€” both soluble. No precipitate, no gas, no water. The correct answer is no reaction, written out as those two words, and it is a genuine prediction rather than an admission of defeat.

Now the more interesting failure: you predicted a reaction and the tube sat there. Three things that could mean, in the order worth checking:

  • The prediction was right and the observation is limited. β€œInsoluble” means low solubility, not zero. If both solutions were dilute, there may not be enough product to see. A faint cloudiness is a positive result.
  • The rule was applied to the wrong case. Reversed the activity series, or read the solubility table’s exceptions column too quickly. Most failed predictions are this.
  • The rule is incomplete. Rate is not part of any of this. Aluminium sits well above hydrogen in the activity series and looks unreactive in air, because it carries a tough oxide layer that has to be got through first. The prediction says what can happen, not how fast, and not what is in the way.

A wrong prediction is data

The instinct to quietly change your prediction after seeing the result is strong and worth resisting. A prediction that failed tells you something about your model that a prediction that succeeded never could β€” and in a lab report, β€œI predicted a precipitate, none formed, and here is what I now think was wrong” earns more than a page of confirmed guesses. That is the argument in What Counts as Evidence, and it is the standard The Reaction Prediction is marked against.

Practise the whole cycle in Reaction Types Practice. Then Combustion takes one of the five types and asks what happens when it runs short of oxygen.

Curriculum connection

C2.4

predict the products of different types of synthesis and decomposition reactions (e.g., synthesis reactions in which simple compounds are formed; synthesis reactions of metallic or non-metallic oxides with water; decomposition reactions, in which a chemical compound is separated into several compounds) [AI]

Link to original

C2.5

predict the products of single displacement reactions, using the metal activity series and the halogen series [AI]

Link to original

C2.6

predict the products of double displacement reactions (e.g., the formation of precipitates or gases; neutralization) [AI]

Link to original