A calculation predicts that a solution is supersaturated, yet the liquid remains clear. The calculation and the observation need not contradict each other. Equilibrium describes a thermodynamic relationship; a visible precipitate also requires a formation pathway, growth and a means of detection.
This guide follows that distinction from an ion-product calculation to an invented waiting-time experiment. It explains why “no precipitate seen in five minutes” is a bounded observation, why adding a seed changes the experiment, and what a clear liquid can and cannot tell us.
eduKateSG · The Gold Standard of Chemistry
Choose your reading route
Separate equilibrium from waiting time · Chapters 1–2. Build the quantities before using a formula.
Work through supersaturation and detection · Chapters 3–5. Follow original examples, assumptions and changed conditions.
Test the evidence and continue · Chapters 6–8. Test transfer, answer common questions and follow connected reading.
Full chapter index · Practice and evidence checks · Return to Chemistry
Chapter index
Separate equilibrium from waiting time · Chapters 1–2
Work through supersaturation and detection · Chapters 3–5
Test the evidence and continue · Chapters 6–8
CHAPTER 1 OF 8 · Separate equilibrium from waiting time
1. Separate the equilibrium question from the clock
For a sparingly soluble solid MX, write the dissolution reaction MX(s) ⇌ M⁺(aq) + X⁻(aq). Its solubility product relates the equilibrium activities of the dissolved ions to that solid phase at a specified temperature.
The ion activity product Q uses the present activities. Comparing Q with Ksp indicates whether the solution is undersaturated, at saturation or supersaturated with respect to that phase. The phrase “with respect to that phase” matters: different solids can have different equilibria even when they contain the same elements.
In a dilute idealised classroom model, we approximate activities using concentrations relative to a 1 mol/L standard. Then Q is represented numerically by [M⁺][X⁻] when concentrations are entered in mol/L. In real mixtures, activity coefficients and dissolved complexes may matter.
The OpenStax treatment of precipitation and dissolution develops the equilibrium comparison. It does not make Ksp a stopwatch. Knowing that precipitation can lower free energy does not tell us that a detectable amount will form within a particular observation interval.
Ask two questions separately: “What state is favoured under these conditions?” and “How does the system reach it on the timescale we observed?” Keeping both questions visible prevents an equilibrium prediction from becoming an unsupported timing claim.
CHAPTER 2 OF 8 · Separate equilibrium from waiting time
2. Explain why a new phase may take time to appear
A small organised cluster is not just a tiny copy of a large crystal. Creating an interface costs free energy. In a simple classical picture, the favourable bulk contribution grows with cluster volume, while the interface contribution grows with surface area. A cluster can therefore face a barrier before continued growth becomes favourable.
That picture helps explain a metastable solution: it is supersaturated relative to a solid phase but can persist while formation of that phase is delayed. It is an explanatory model, not a universal account of every crystallisation pathway.
Nucleation creates a new stable growth centre. Growth adds material to existing centres. A detector may register the result only after sufficient size, number or optical contrast develops. These are related stages, but they are not interchangeable measurements.
Define what “induction time” means in the experiment
One investigator may mean the interval before the first nucleus forms; another may report the interval before an optical signal crosses a threshold. A transparent report gives the operational definition instead of assuming the term resolves the distinction.
For this guide, t_detect is time from the stated starting condition to a defined detectable signal. In a simplified single-nucleus model, t_detect = t_wait + t_growth. Mixing, thermal equilibration or other delays can complicate that relationship.
Research comparing stochastic and deterministic estimates of nucleation rates shows why assumptions about nucleation and detection affect inferred kinetics. The following examples deliberately keep those assumptions explicit.
CHAPTER 3 OF 8 · Work through supersaturation and detection
3. Calculate supersaturation before predicting an amount
Imagine equal volumes of two dilute solutions mixed without a volume change beyond addition. One initially contains 4.0 × 10⁻⁴ mol/L free M⁺ and the other 4.0 × 10⁻⁴ mol/L free X⁻. Immediately after mixing, before precipitation or complexation, each concentration is 2.0 × 10⁻⁴ mol/L.
Assume the chosen MX phase has Ksp = 1.0 × 10⁻⁸ in this idealised model. Then Q = (2.0 × 10⁻⁴)² = 4.0 × 10⁻⁸. The ratio Q/Ksp is 4.
The solution is supersaturated with respect to MX under the assumptions. This is not evidence that a precipitate was already present at the first instant.
Distinguish two supersaturation conventions
For this 1:1 solid with equal free-ion concentrations, the equilibrium concentration is 1.0 × 10⁻⁴ mol/L. The concentration ratio c/c_eq is therefore 2, while Q/Ksp is 4. Some descriptions use a stoichiometric root of the activity-product ratio. Always state which quantity a symbol such as S represents.
| Quantity | Invented value | Meaning |
|---|---|---|
| Free concentration of each ion after mixing | 2.0 × 10⁻⁴ mol/L | Initial dissolved state |
| Q/Ksp | 4 | Activity-product supersaturation ratio in the ideal model |
| c/c_eq | 2 | Equal-ion concentration ratio |
| Eventual concentration of each ion at saturation | 1.0 × 10⁻⁴ mol/L | Predicted equilibrium with the chosen solid present |
If the final mixed volume is 0.100 L, reaching that equilibrium would remove 1.0 × 10⁻⁴ mol/L of each ion, forming 1.0 × 10⁻⁵ mol of MX. This is an equilibrium material balance, assuming no competing reactions, no evaporation and the stipulated solid phase.
It predicts a possible final amount under those conditions. It does not establish when that amount appears, its particle size or whether an observer will see it.
CHAPTER 4 OF 8 · Work through supersaturation and detection
4. Work a waiting-time model without making it a law
Assume a fictional collection of identical, independent vials held at constant conditions. Before any event, each vial has a constant nucleation hazard λ = 0.20 per minute. In this Poisson model, the probability of no nucleation event by time t is exp(−λt).
At five minutes, exp(−1) ≈ 0.368. About 36.8% would still have no event in the long-run model. At ten minutes, exp(−2) ≈ 0.135. A supersaturated vial remaining clear for a while is therefore compatible with delayed nucleation, even before considering detection limits.
The mean waiting time is 1/λ = 5 minutes, but the median is ln(2)/λ ≈ 3.47 minutes. A mean is not a deadline. The probability that any particular vial nucleates exactly at five minutes is not what the mean expresses.
Now assume every first nucleus takes a further two minutes to become detectable. At five minutes, detection requires nucleation by three minutes. The model gives:
P(detected by 5 min) = 1 − exp(−0.20 × 3) ≈ 0.451.
Thus about 54.9% remain below the detection threshold at five minutes. Some have no nucleus; others have a nucleus that has not yet produced the required signal. “Not detected” combines those possibilities.
These percentages are consequences of invented assumptions, not measured probabilities for a chemical. Real nucleation rates can vary with supersaturation, temperature, surfaces and process history. Growth times can vary too. Fitting this formula without testing its assumptions would not establish a valid rate constant.
CHAPTER 5 OF 8 · Work through supersaturation and detection
5. Change the seed, the detector or the solution
Adding a suitable seed supplies an existing crystal surface. That can reduce dependence on spontaneous primary nucleation, but it does not make every other delay vanish. Growth, transport, secondary nucleation and the state of the seed still matter.
A seeded experiment and an unseeded one therefore answer related but different questions. Faster visible change after seeding supports a role for the pathway; it does not automatically prove that the solubility product changed. The seeded and unseeded crystallisation study by Cashmore and colleagues illustrates why growth and nucleation need separate examination.
Changing the detector can also change the recorded time. If one system detects small particles earlier than another, a shorter t_detect need not mean faster chemistry. Preserve the signal threshold and sampling interval when comparing runs.
Finally, apparent supersaturation may be a modelling mistake. Total metal is not necessarily free M⁺. A ligand could bind much of it. An activity correction could alter Q. The solid that appears could differ from the phase used for Ksp.
Use the existing guides on finite ligand supply and free-ion solubility and ionic strength, activities and equilibrium constants to check those alternatives before blaming delayed nucleation.
CHAPTER 6 OF 8 · Test the evidence and continue
6. Turn a clear-vial observation into a useful record
| Record | Why it matters |
|---|---|
| Composition, free-species model and candidate solid | Establishes what Q/Ksp actually refers to |
| Temperature, mixing and starting-time definition | Makes elapsed times comparable |
| Container, preparation history and seed condition | Describes possible nucleation surfaces and pathways |
| Detection method, threshold and observation interval | Defines what “first appeared” means |
| All repeats, including runs without detection | Prevents selection of only the fastest events |
Suppose observation ends at ten minutes with no signal. Record “not detected within ten minutes under the stated threshold.” In time-to-event language, the record is right-censored. Do not enter ten minutes as an observed event or silently discard that vial.
Changed-context practice
In the equilibrium example, dilute the freshly mixed solution twofold before any solid forms, keeping temperature and the ideal assumptions fixed. Each ion becomes 1.0 × 10⁻⁴ mol/L, so Q = Ksp. The supersaturation driving force in that model disappears. Merely waiting longer is no longer the same experiment.
Next, retain λ = 0.20 per minute but use a detector with a fixed one-minute growth delay. Detection by five minutes has probability 1 − exp(−0.8) ≈ 0.551. This is greater than 0.451 even though the nucleation model is unchanged.
Finally, a class records only three vials that became cloudy quickly and omits seven that stayed clear. Can those three times describe the whole experiment? No. The selection depends on the outcome and hides information about the waiting-time distribution.
These are paper-and-data exercises. They build reasoning without requiring an improvised precipitation experiment or handling unfamiliar chemicals.
CHAPTER 7 OF 8 · Test the evidence and continue
7. Questions about delayed precipitation
Does a clear solution have to be unsaturated?
No. It may be unsaturated, saturated or supersaturated. Visual clarity alone also cannot exclude particles below the observation threshold.
Does Q greater than Ksp mean precipitation is impossible to delay?
No. The equilibrium comparison does not remove kinetic barriers, growth requirements or detector limits.
Is every delay caused by nucleation?
No. Mixing, temperature equilibration, reaction steps, growth and detection can contribute. Establish the relevant stages before fitting a model.
Can I compare Q values calculated from total concentrations?
Only if the species model justifies substituting those totals for the relevant free-ion activities or approximations. Complexation and acid–base speciation can invalidate the substitution.
Does seeding guarantee the same final solid?
No universal guarantee follows. Phase identity, stability and experimental conditions still need verification. A photograph of crystals is not a full identification of their structure or composition.
Does a single long delay disprove supersaturation?
It prompts a check, but does not by itself settle the issue. Reassess composition, activities, phase choice, timing and detection, then compare replicated evidence with the proposed mechanism.
CHAPTER 8 OF 8 · Test the evidence and continue
8. Follow the equilibrium and evidence routes
Build the chemistry
- The Gold Standard of Chemistry connects particles, equations, rates and practical evidence.
- Solubility, saturation and growing crystals supplies the broader first-principles explanation.
- The Gold Standard of Kinetics develops rates, mechanisms and time-dependent evidence.
See the mechanism in a different setting
- Liesegang rings connects diffusion, local depletion and precipitation patterns.
- Reusable hand warmers and sodium acetate crystallisation provides a familiar example of a triggered change.
- Why honey crystallises explores composition, nucleation and a food system with additional complexity.
Transfer the checking method
The Gold Standard of Education connects a classroom model to defensible real-world claims. How X Works offers the wider reading map.
Sources linked above were reviewed on 22 October 2026. OpenStax supports the equilibrium framework; the research articles examine the assumptions behind nucleation and detection models. All numerical chemical properties, vial probabilities and detector delays here are original teaching assumptions, not experimental findings or operating instructions.