Solubility and the Solubility Product

A kidney stone forms silently, one calcium oxalate unit at a time. It grows when the ion product in urine exceeds a stubborn thermodynamic ceiling. That ceiling has a name — the solubility product, $K_{sp}$ . It governs everything from cave stalactites to the white haze formed when you mix silver nitrate with tap water. Solubility looks like a single number you'd find on a label. In truth, it is an equilibrium constant wearing a disguise. It is subject to every rule already established for The Equilibrium Constant and Reaction Quotient. This matters to analytical chemists because precipitation is both a nuisance and a tool. Gravimetric analysis depends on driving an ion quantitatively out of solution as an insoluble solid. Selective precipitation separates one metal from another based on tiny differences in $K_{sp}$ . Unwanted precipitation — a precipitate fouling an electrode, a titrant reacting with a matrix ion — must be predicted and avoided. The same equilibrium framework that predicts whether silver chloride will precipitate from seawater also predicts whether a calcium phosphate scale will clog a heat exchanger. The big picture is that molar solubility, $s$ , is not a fixed property of a compound. It is instead a solution to an equilibrium expression. That expression depends on what else is dissolved. Adding a common ion suppresses solubility by Le Chatelier reasoning made quantitative. Removing a common ion (or adding a complexing agent) enhances it. The reaction quotient $Q_{sp}$ , compared against $K_{sp}$ , does more than one job. It tells you whether a precipitate can exist at equilibrium. It also tells you whether it will form from a given mixture right now. This is a distinction every stoichiometry problem eventually needs. This article builds the $K_{sp}$ expression from first principles. It derives molar solubility for salts of different stoichiometry. It quantifies the common-ion effect algebraically. It also establishes the $Q$ -versus- $K_{sp}$ decision rule. This comes with fully worked, asymmetric numerical examples. These include the classic trap where naive addition of concentrations gives the wrong answer.

Key takeaways

$K_{sp}$ is the equilibrium constant for a solid dissolving into its constituent ions; pure solids and liquids are omitted from the expression, exactly as in The Equilibrium Constant and Reaction Quotient.
Molar solubility $s$ (mol solute dissolved per liter) is not $K_{sp}$ itself — it must be solved from the $K_{sp}$ expression using the stoichiometric coefficients.
For a salt $M_aX_b$ , $K_{sp} = [M]^a[X]^b$ and at saturation in pure water, $[M] = as$ , $[X] = bs$ .
The common-ion effect suppresses solubility because one ion's concentration is fixed by an external source, forcing the other down algebraically (not by simple subtraction).
Precipitation is predicted by comparing $Q_{sp}$ (instantaneous ion product) to $K_{sp}$ : $Q_{sp} > K_{sp}$ means a precipitate forms; $Q_{sp} < K_{sp}$ means the solution stays clear.
Mixing two solutions dilutes both ions before you compute $Q_{sp}$ — a step students routinely skip and that flips the wrong-vs-right answer.
Key idea: solubility is an equilibrium position, computed from $K_{sp}$ and stoichiometry, not a stored constant — and whether a precipitate forms is decided by comparing the instantaneous ion product $Q_{sp}$ to that same $K_{sp}$ .

Where this lives today

Gravimetric determination of sulfate as BaSO₄, chloride as AgCl, or nickel as the dimethylglyoxime complex all rely on driving $Q_{sp}$ (or an analogous formation-constant quotient) far above the relevant solubility limit to achieve quantitative (>99.9%) precipitation, while controlling conditions to grow large, filterable crystals rather than a colloidal mess. Water treatment engineers use $K_{sp}$ calculations to predict and prevent scale formation from $\text{CaCO}_3$ , $\text{CaSO}_4$ , and $\text{Mg(OH)}_2$ in boilers, reverse-osmosis membranes, and cooling towers, often deliberately dosing antiscalants that sequester the common ion below its precipitation threshold. In biomedicine, kidney stone risk is assessed by computing the urinary saturation index — essentially $\log(Q_{sp}/K_{sp})$ — for calcium oxalate and calcium phosphate, guiding dietary and pharmacological interventions that lower urinary calcium or oxalate to suppress $Q_{sp}$ below the stone-forming threshold. Selective precipitation, exploiting differences in $K_{sp}$ among metal sulfides or hydroxides, remains a workhorse separation technique in qualitative inorganic analysis and in ore processing, where controlled pH or sulfide addition precipitates one metal while keeping a chemically similar one in solution. Pharmaceutical formulation scientists likewise engineer drug salts and polymorphs with tuned solubility products to control dissolution rate and bioavailability, sometimes intentionally exploiting the common-ion effect (e.g., chloride in gastric fluid) to predict a drug's behavior in vivo — a bridge to the ionic-strength and conditional-constant corrections that refine this basic $K_{sp}$ framework in more concentrated, biologically realistic media.

Ask the tutor: How do activity coefficients modify the $K_{sp}$ expression once the ionic strength of the solution is no longer negligible?

The physical picture

Picture a crystal of silver chloride sitting at the bottom of a beaker of water. At the crystal's surface, $\text{Ag}^+$ and $\text{Cl}^-$ ions are constantly leaving the lattice (dissolution) and constantly rejoining it (precipitation). When these two rates are equal, the system has reached dynamic equilibrium — not because dissolution has stopped, but because it is exactly balanced by the reverse process. This is no different in kind from any gas-phase or solution equilibrium: it is governed by a thermodynamic equilibrium constant, and it responds to concentration changes exactly as Le Chatelier's Principle predicts. The equilibrium is written as a heterogeneous reaction:

\text{AgCl}(s) \rightleftharpoons \text{Ag}^+(aq) + \text{Cl}^-(aq)

Because $\text{AgCl}(s)$ is a pure solid, its activity is defined as 1 and it drops out of the equilibrium expression — the same rule used for any pure condensed phase in a heterogeneous equilibrium. What remains is the solubility product:

K_{sp} = [\text{Ag}^+][\text{Cl}^-] = 1.8 \times 10^{-10} \text{ at } 25\,^\circ\text{C}

The crucial subtlety is that $K_{sp}$ is a single fixed number (at a given temperature), but the individual ion concentrations $[\text{Ag}^+]$ and $[\text{Cl}^-]$ are not fixed — they can be traded off against each other. If you dump extra $\text{Cl}^-$ into the beaker (say, from NaCl), the product $[\text{Ag}^+][\text{Cl}^-]$ must still equal $1.8\times10^{-10}$ , so $[\text{Ag}^+]$ must fall. This is the common-ion effect, and it is nothing more than algebra applied to a constant product. For salts with unequal stoichiometry, such as $\text{PbI}_2$ or $\text{Ag}_2\text{CrO}_4$ , the bookkeeping requires care: each mole of solid that dissolves releases two moles of iodide (or chromate), not one. The molar solubility $s$ and the ion concentrations are related by the stoichiometric coefficients, and forgetting the factor of 2 (or squaring it in the wrong place) is the single most common numerical error in this topic.

Figure

Dissolution equilibrium at the crystal surface: rates of dissolution and deposition balance at saturation

Figure

Stoichiometric bookkeeping for PbI₂: one mole of solid releases one Pb²⁺ and two I⁻

The core ideas

For a generic sparingly soluble salt $M_aX_b(s) \rightleftharpoons a\,M^{n+}(aq) + b\,X^{m-}(aq)$ , the solubility product is:

K_{sp} = [M^{n+}]^{a}[X^{m-}]^{b}

where:

$[M^{n+}], [X^{m-}]$ are equilibrium molar concentrations (strictly, activities — see the scope note below) of the cation and anion,
$a, b$ are the stoichiometric coefficients from the balanced dissolution equation,
the pure solid $M_aX_b(s)$ does not appear because its activity is defined as unity.

Reading: $K_{sp}$ is fixed at a given temperature; it is the product of ion concentrations raised to their stoichiometric powers that is conserved, not the concentrations individually. This is why adding one ion forces the other down. If the salt dissolves in pure water with no other source of $M^{n+}$ or $X^{m-}$ , and $s$ is the molar solubility (mol of $M_aX_b$ dissolved per liter),

[M^{n+}] = as, \qquad [X^{m-}] = bs

K_{sp} = (as)^a(bs)^b = a^a b^b\, s^{a+b}

where:

$s$ is the single unknown, solved by taking the $(a+b)$ -th root,
$a^a b^b$ is a pure numerical factor fixed by stoichiometry (e.g., for a 1:2 salt, $a=1,b=2$ , giving $1^1\cdot2^2 = 4$ ).

Reading: this formula converts a tabulated $K_{sp}$ directly into a solubility, but the exponent $a+b$ and prefactor $a^ab^b$ depend entirely on the salt's ion ratio — there is no universal "take the square root" rule. The reaction quotient for precipitation prediction is built the same way but uses instantaneous, not necessarily equilibrium, concentrations:

Q_{sp} = [M^{n+}]_0^{a}[X^{m-}]_0^{b}

where:

the subscript 0 denotes concentrations at the moment of mixing, before any reaction/precipitation occurs,
$Q_{sp}$ has the same algebraic form as $K_{sp}$ but is evaluated off-equilibrium.

Reading: comparing $Q_{sp}$ to $K_{sp}$ tells you which direction the system must shift to reach equilibrium — exactly the logic used for any reaction quotient, specialized to a precipitation reaction:

Q_{sp} > K_{sp} \Rightarrow \text{precipitation occurs (Q must fall)}, \quad Q_{sp} < K_{sp} \Rightarrow \text{no precipitate (undersaturated)}

Derivation AMolar solubility of a 1:2 salt from

K_{sp}

, with the common-ion effect built in

Goal. Derive molar solubility $s$ for $\text{PbI}_2$ ( $K_{sp} = [\text{Pb}^{2+}][\text{I}^-]^2$ ) first in pure water, then in a solution already containing a fixed background concentration $C$ of iodide from an added salt like KI, and show algebraically why $s$ drops.

Step 1 — set up the equilibrium in pure water. Dissolution releases one $\text{Pb}^{2+}$ and two $\text{I}^-$ per formula unit:

\text{PbI}_2(s) \rightleftharpoons \text{Pb}^{2+}(aq) + 2\,\text{I}^-(aq)

Let $s$ = molar solubility. Then $[\text{Pb}^{2+}] = s$ and $[\text{I}^-] = 2s$ , because each dissolved formula unit contributes twice as much iodide as lead.

Step 2 — substitute into the $K_{sp}$ expression.

K_{sp} = [\text{Pb}^{2+}][\text{I}^-]^2 = (s)(2s)^2 = 4s^3

Solving for $s$ :

s = \left(\frac{K_{sp}}{4}\right)^{1/3}

This is the pure-water molar solubility — a single unknown solved from a single equation.

Step 3 — introduce a common ion. Now suppose the solid is added to a solution that already contains iodide at concentration $C$ (mol/L) from dissolved KI, and let $s'$ be the new, generally much smaller, molar solubility of PbI₂ in this medium. Every mole of PbI₂ that dissolves still contributes $s'$ mol/L of Pb²⁺ and $2s'$ mol/L of I⁻, but the total iodide concentration is now the sum of what came from the salt and what came from KI:

[\text{Pb}^{2+}] = s', \qquad [\text{I}^-] = C + 2s'

Step 4 — write the exact equilibrium expression and apply the common-ion approximation.

K_{sp} = s'(C + 2s')^2

If $C \gg 2s'$ (verified after solving, as in any approximation-based equilibrium calculation), this simplifies to

K_{sp} \approx s'\,C^2 \quad \Rightarrow \quad s' \approx \frac{K_{sp}}{C^2}

This is the key qualitative result: solubility in the presence of a common ion scales as $1/C^2$ for this stoichiometry — a much steeper suppression than the naive expectation of simple subtraction, and steeper than the 1:1 case (where it scales as $1/C$ ). The exponent on $C$ always equals the stoichiometric coefficient of the common ion in the $K_{sp}$ expression.

Go deeper · Derivation B · 1 minPredicting precipitation from mixing two solutions: the dilution-then-compare protocol

Derivation BPredicting precipitation from mixing two solutions: the dilution-then-compare protocol

Goal. Derive the correct procedure for deciding whether a precipitate forms when two solutions of soluble salts are mixed, using $\text{BaCl}_2$ and $\text{Na}_2\text{SO}_4$ as the working example, and show why omitting the dilution step gives a wrong answer.

Step 1 — identify the precipitation reaction. Ba²⁺ and SO₄²⁻ combine to form insoluble barium sulfate:

\text{Ba}^{2+}(aq) + \text{SO}_4^{2-}(aq) \rightleftharpoons \text{BaSO}_4(s) \qquad K = 1/K_{sp}

The forward direction (precipitation) is governed by the same $K_{sp}$ as the dissolution reaction, just inverted.

Step 2 — recognize that mixing dilutes both stock concentrations. If $V_1$ liters of a solution with $\text{Ba}^{2+}$ concentration $c_1$ are combined with $V_2$ liters of a solution with $\text{SO}_4^{2-}$ concentration $c_2$ , the concentrations immediately after mixing but before any reaction are found by conservation of moles in the new total volume:

[\text{Ba}^{2+}]_0 = \frac{c_1 V_1}{V_1+V_2}, \qquad [\text{SO}_4^{2-}]_0 = \frac{c_2 V_2}{V_1+V_2}

This dilution step is not optional — using the undiluted stock concentrations systematically overestimates $Q_{sp}$ and can falsely predict precipitation.

Step 3 — form the reaction quotient from the diluted concentrations.

Q_{sp} = [\text{Ba}^{2+}]_0\,[\text{SO}_4^{2-}]_0

This has exactly the same algebraic form as $K_{sp}$ for BaSO₄ (a 1:1 salt), but is evaluated at the instant of mixing.

Step 4 — apply the decision rule. Compare $Q_{sp}$ to the tabulated $K_{sp} = 1.1\times10^{-10}$ :

Q_{sp} > K_{sp} \Rightarrow \text{precipitate forms until } Q_{sp} \text{ falls back to } K_{sp}

Q_{sp} < K_{sp} \Rightarrow \text{no precipitate; ions remain fully dissolved}

If precipitation does occur, the equilibrium concentrations remaining in solution afterward are found by a separate common-ion-effect calculation (Step 3-4 of the previous derivation), not by $Q_{sp}$ itself — $Q_{sp}$ only answers the yes/no question.

Worked examples

Example 1 — Foundational: molar solubility of silver chromate from $K_{sp}$ . Silver chromate, $\text{Ag}_2\text{CrO}_4$ , has $K_{sp} = 1.12\times10^{-12}$ at 25 °C. Find its molar solubility in pure water. Write the dissolution equilibrium and identify stoichiometry:

\text{Ag}_2\text{CrO}_4(s) \rightleftharpoons 2\,\text{Ag}^+(aq) + \text{CrO}_4^{2-}(aq)

Let $s$ be the molar solubility. Each formula unit releases 2 Ag⁺ and 1 CrO₄²⁻, so

[\text{Ag}^+] = 2s, \qquad [\text{CrO}_4^{2-}] = s

Substitute into the solubility product:

K_{sp} = (2s)^2(s) = 4s^3

Solve for $s$ :

s = \left(\frac{K_{sp}}{4}\right)^{1/3} = \left(\frac{1.12\times10^{-12}}{4}\right)^{1/3} = (2.80\times10^{-13})^{1/3}

s = 6.54\times10^{-5}\ \text{mol/L}

Check the exponent by reconstructing: $[\text{Ag}^+] = 2(6.54\times10^{-5}) = 1.308\times10^{-4}$ M, $[\text{CrO}_4^{2-}] = 6.54\times10^{-5}$ M. Then

Q = (1.308\times10^{-4})^2(6.54\times10^{-5}) = (1.711\times10^{-8})(6.54\times10^{-5}) = 1.119\times10^{-12}

This matches $K_{sp} = 1.12\times10^{-12}$ to three significant figures, confirming the arithmetic. The molar solubility is $6.54\times10^{-5}$ mol/L, corresponding to a mass solubility of $(6.54\times10^{-5}\text{ mol/L})(331.73\text{ g/mol}) = 2.17\times10^{-2}\text{ g/L}$ , i.e., about 22 mg per liter — consistent with $\text{Ag}_2\text{CrO}_4$ 's well-known deep red, sparingly soluble character used as the Mohr titration endpoint indicator.

Example 2 — Applied: common-ion suppression in a fluoride-fortified water supply. Calcium fluoride, $\text{CaF}_2$ , has $K_{sp} = 3.9\times10^{-11}$ at 25 °C. A municipal water supply is fluoridated to $[\text{F}^-] = 1.0\times10^{-4}$ M (about 1.9 mg/L, a typical fluoridation level). Groundwater in the same aquifer contains dissolved $\text{Ca}^{2+}$ from limestone. Find the maximum $[\text{Ca}^{2+}]$ that can remain dissolved without precipitating $\text{CaF}_2$ , and compare this to the molar solubility of $\text{CaF}_2$ in pure water. Part A — pure-water solubility, for comparison. With $\text{CaF}_2 \rightleftharpoons \text{Ca}^{2+} + 2\text{F}^-$ , $[\text{Ca}^{2+}] = s$ , $[\text{F}^-] = 2s$ :

K_{sp} = s(2s)^2 = 4s^3 \Rightarrow s = \left(\frac{3.9\times10^{-11}}{4}\right)^{1/3} = (9.75\times10^{-12})^{1/3} = 2.14\times10^{-4}\ \text{M}

In pure water, calcium fluoride would dissolve to $2.14\times10^{-4}$ M — corresponding to a natural fluoride level of $2(2.14\times10^{-4}) = 4.28\times10^{-4}$ M, well above the fluoridation target. Part B — with fixed background fluoride. Now fluoride is held near $1.0\times10^{-4}$ M by the fluoridation program (treated as fixed because it vastly exceeds any contribution from dissolving CaF₂, an assumption to verify at the end). At equilibrium:

K_{sp} = [\text{Ca}^{2+}][\text{F}^-]^2

3.9\times10^{-11} = [\text{Ca}^{2+}](1.0\times10^{-4})^2

[\text{Ca}^{2+}] = \frac{3.9\times10^{-11}}{1.00\times10^{-8}} = 3.9\times10^{-3}\ \text{M}

Check self-consistency: if $[\text{Ca}^{2+}] = 3.9\times10^{-3}$ M were entirely from dissolving CaF₂ rather than from limestone dissolution independently, it would add $2\times(3.9\times10^{-3}) = 7.8\times10^{-3}$ M of fluoride, which is not negligible compared to $1.0\times10^{-4}$ M — it's 78 times larger. This flags an important subtlety: the "fixed fluoride" assumption only holds if the fluoride truly comes from an external, well-buffered source (the fluoridation program continuously replenishing it) rather than from the CaF₂ itself. In practice, the calculation shows that any calcium above roughly $3.9\times10^{-3}$ M (about 156 mg/L, a level found in moderately hard water) risks precipitating fluorapatite-like scale or CaF₂ deposits given sufficient fluoride, which is why utilities monitor water hardness alongside fluoridation levels. The qualitative lesson — background fluoride suppresses the ceiling on tolerable calcium by four orders of magnitude relative to the pure-water case — is the operative result.

Example 3 — Exam-level: mixing two solutions, dilution, and a two-precipitate competition. A 25.00 mL aliquot of $0.0400$ M $\text{Pb(NO}_3)_2$ is mixed with $35.00$ mL of $0.0250$ M $\text{NaCl}$ . Given $K_{sp}(\text{PbCl}_2) = 1.7\times10^{-5}$ , determine whether a precipitate of $\text{PbCl}_2$ forms, and if so, find the equilibrium $[\text{Pb}^{2+}]$ remaining in solution. Step 1 — total volume and diluted concentrations. Total volume:

V_{tot} = 25.00 + 35.00 = 60.00\ \text{mL} = 0.06000\ \text{L}

Moles of each ion before any reaction:

n(\text{Pb}^{2+}) = (0.02500\ \text{L})(0.0400\ \text{M}) = 1.000\times10^{-3}\ \text{mol}

n(\text{Cl}^-) = (0.03500\ \text{L})(0.0250\ \text{M}) = 8.75\times10^{-4}\ \text{mol}

Diluted (pre-reaction) concentrations:

[\text{Pb}^{2+}]_0 = \frac{1.000\times10^{-3}}{0.06000} = 1.667\times10^{-2}\ \text{M}

[\text{Cl}^-]_0 = \frac{8.75\times10^{-4}}{0.06000} = 1.458\times10^{-2}\ \text{M}

Step 2 — compute $Q_{sp}$ and compare to $K_{sp}$ .

Q_{sp} = [\text{Pb}^{2+}]_0[\text{Cl}^-]_0^2 = (1.667\times10^{-2})(1.458\times10^{-2})^2

= (1.667\times10^{-2})(2.126\times10^{-4}) = 3.545\times10^{-6}

Compare: $Q_{sp} = 3.55\times10^{-6}$ versus $K_{sp} = 1.7\times10^{-5}$ . Since $Q_{sp} < K_{sp}$ , no precipitate forms — the mixture is undersaturated with respect to PbCl₂, and both ions remain fully dissolved at their diluted concentrations. This is the answer: $[\text{Pb}^{2+}] = 1.67\times10^{-2}$ M, unchanged by any precipitation because none occurs. Step 3 — a variant to illustrate the case where precipitation does occur. Suppose instead the NaCl solution were $0.250$ M (ten times more concentrated), giving $n(\text{Cl}^-) = 8.75\times10^{-3}$ mol and $[\text{Cl}^-]_0 = 8.75\times10^{-3}/0.06000 = 0.1458$ M. Then

Q_{sp} = (1.667\times10^{-2})(0.1458)^2 = (1.667\times10^{-2})(2.126\times10^{-2}) = 3.545\times10^{-4}

Now $Q_{sp} = 3.5\times10^{-4} > K_{sp} = 1.7\times10^{-5}$ , so PbCl₂ does precipitate. Because chloride is in stoichiometric excess (Cl⁻ available: $8.75\times10^{-3}$ mol vs. the $2\times(1.000\times10^{-3}) = 2.00\times10^{-3}$ mol needed to consume all the lead), essentially all the Pb²⁺ is driven into the solid, leaving excess chloride as the "common ion" governing residual solubility. Let $x$ = equilibrium $[\text{Pb}^{2+}]$ , small compared to the leftover chloride. Total chloride remaining after most of the reaction: initial $8.75\times10^{-3}$ mol Cl⁻ minus $2\times(1.000\times10^{-3})$ mol consumed by essentially all the lead precipitating $= 6.75\times10^{-3}$ mol, giving $[\text{Cl}^-] \approx 6.75\times10^{-3}/0.06000 = 0.1125$ M. Then

K_{sp} = x(0.1125)^2 \Rightarrow x = \frac{1.7\times10^{-5}}{0.01266} = 1.34\times10^{-3}\ \text{M}

Checking the approximation: $x = 1.34\times10^{-3}$ M is about 12% of the assumed excess chloride's magnitude change, borderline but acceptable for a two-significant-figure answer, consistent with the tolerances used in Solving Equilibrium Problems with Approximations. Residual $[\text{Pb}^{2+}] \approx 1.3\times10^{-3}$ M in the chloride-excess variant.

Limiting cases, scope & common traps

Situation	What happens	Governing check
Pure water, no common ion	$s$ from $K_{sp} = a^ab^b s^{a+b}$ directly	Use full stoichiometric exponents
Common ion present, $C \gg 2s'$ (or $bs'$ )	$s'$ suppressed, scales as $K_{sp}/C^{\,b}$	Verify $C \gg$ contribution from dissolution afterward
Two solutions mixed	Must dilute to total volume before computing $Q_{sp}$	Compare diluted $Q_{sp}$ to tabulated $K_{sp}$
$Q_{sp} = K_{sp}$ exactly	Solution is exactly saturated; no net change	Borderline case, rarely realized exactly in practice
High ionic strength (concentrated salt solutions)	Activity coefficients deviate from 1; concentration-based $K_{sp}$ becomes approximate	Full treatment needs $K_{sp} = a_M^a a_X^b$ (foreshadowed in ionic-strength corrections)
Complexing or pH-dependent anions (e.g., CO₃²⁻, S²⁻, F⁻)	Side reactions change the effective free-ion concentration	Requires conditional/competing-equilibrium treatment

Common trapMisconception: "common-ion effect subtracts concentrations"

A frequent error is assuming that adding, say, 0.10 M Cl⁻ to a AgCl-saturated solution simply reduces $[\text{Ag}^+]$ by 0.10 M. This confuses stoichiometric subtraction (valid for a reaction going essentially to completion) with an equilibrium constraint. The correct approach recognizes that $[\text{Ag}^+][\text{Cl}^-] = K_{sp}$ must hold exactly at equilibrium; you solve for $[\text{Ag}^+]$ as $K_{sp}/[\text{Cl}^-]$ , not by subtraction. The two methods can differ by orders of magnitude when $K_{sp}$ is very small, because subtraction ignores that essentially none of the added common ion is consumed by dissolution — it stays almost entirely as-added.

The chemist behind it

The solubility product concept emerged from the broader mass-action framework developed by Cato Guldberg and Peter Waage in the 1860s, but its systematic application to sparingly soluble salts and precipitation analysis is most associated with Wilhelm Ostwald and the German school of physical chemistry in the 1890s, who formalized the treatment of heterogeneous equilibria using activities and concentrations. Ostwald's dilution law and his broader program of quantifying ionic equilibria — building on Arrhenius's theory of electrolytic dissociation — gave analytical chemists the tool to predict, rather than merely observe, when a precipitate would form. Gravimetric analysis, already a mature 19th-century technique built on empirical precipitation recipes, was transformed by this quantitative footing: chemists could now calculate detection limits, design selective separations based on differing $K_{sp}$ values, and understand why "digestion" of a precipitate (aging it in contact with its mother liquor) improved crystal quality by allowing the system to approach true equilibrium rather than a kinetically trapped, more soluble metastable form.

Origins · 1 minWilhelm Ostwald and how this was discovered

Wilhelm Ostwald

1853–1932 · Riga-born physical chemist, professor at Leipzig, architect of the ionic-equilibrium framework

Ostwald synthesized Arrhenius's dissociation theory and van't Hoff's thermodynamic treatment of solutions into a coherent, quantitative discipline — physical chemistry — that gave meaning to constants like $K_{sp}$ , $K_a$ , and $K_w$ as members of a single conceptual family. He shared the 1909 Nobel Prize in Chemistry for his work on catalysis, chemical equilibria, and reaction rates, but his textbooks and his founding of the journal Zeitschrift für physikalische Chemie did as much to establish the field as any single discovery.

History

- 1864 — Guldberg and Waage propose the law of mass action, the conceptual ancestor of all equilibrium constants including

K_{sp}

- 1884 — van't Hoff publishes his studies on chemical dynamics, formalizing the thermodynamic basis for equilibrium constants.

- 1887 — Arrhenius proposes the theory of electrolytic dissociation, explaining why salts produce independently mobile ions in solution.

- 1889 — Ostwald's dilution law connects degree of dissociation to concentration for weak electrolytes, extending mass-action reasoning to ionic systems.

- 1923 — Debye and Hückel provide the first rigorous theory of ionic activity coefficients, later refining

K_{sp}

calculations at nonzero ionic strength.

K_{sp}

the same thing as "solubility"?

No. $K_{sp}$ is a dimensionless (or concentration-squared-type) equilibrium constant, fixed at a given temperature, while molar solubility $s$ is a concentration (mol/L) that must be calculated from $K_{sp}$ using the salt's stoichiometry. Two salts can have very different molar solubilities despite similar $K_{sp}$ values if their ion ratios differ, because the algebraic relationship $K_{sp} = a^ab^b s^{a+b}$ depends on the exponents $a$ and $b$ . Always convert $K_{sp}$ to $s$ (or vice versa) explicitly rather than treating them as interchangeable numbers.

Why does the common-ion effect suppress solubility rather than just proportionally reducing it?

Because the constraint is multiplicative, not additive: $K_{sp}$ fixes the product of ion concentrations raised to their stoichiometric powers. When one ion is held near a large fixed value $C$ by an external source, the other must fall to compensate — and if that other ion appears with exponent $b>1$ in $K_{sp}$ , the suppression is even steeper, scaling as $1/C^b$ . This is why even modest common-ion additions can suppress solubility by several orders of magnitude, especially for salts like PbI₂ or Ca(OH)₂ where an ion appears squared.

When should I use

Q_{sp}

instead of

K_{sp}

Use $Q_{sp}$ whenever you need to decide whether a system will move toward precipitation from a given, possibly non-equilibrium, starting mixture — for instance, right after mixing two solutions. $Q_{sp}$ has the identical algebraic form as $K_{sp}$ but is evaluated at the concentrations present at that instant. Once you've established that precipitation occurs, you switch back to $K_{sp}$ -based equilibrium calculations (often with common-ion reasoning) to find the concentrations remaining in solution afterward.

Does the common-ion effect ever fail to apply?

Yes — if the "common" ion also participates in a competing equilibrium (protonation of a weak-acid anion, complexation by excess ligand, or oxidation state changes), simple $K_{sp}$ reasoning breaks down and you need a conditional or effective solubility product that folds in the side reactions. For example, adding excess ammonia to a AgCl suspension increases solubility rather than suppressing it, because Ag⁺ is pulled out of the $K_{sp}$ equilibrium by forming the soluble complex $\text{Ag(NH}_3)_2^+$ — a competing-equilibria effect layered on top of, not replacing, the underlying $K_{sp}$ .

How accurate is the concentration-based

K_{sp}

in real, non-dilute solutions?

Tabulated $K_{sp}$ values are thermodynamic constants defined in terms of activities, not concentrations, so the simple concentration-based expressions used here are excellent approximations only in dilute solution (typically ionic strength below about 0.01 M). At higher ionic strengths, activity coefficients deviate meaningfully from 1, and the true thermodynamic solubility product requires activity corrections — this refinement is the natural next step once the basic concentration-based framework here is secure.

Why do some sources report

K_{sp}

values that differ by 10–50% from each other for the same salt?

Because $K_{sp}$ is experimentally difficult to measure with high precision for extremely insoluble salts (concentrations in the $10^{-8}$ – $10^{-12}$ M range strain most detection methods), and because it is sensitive to temperature, ionic strength of the measurement medium, and sometimes trace complexation effects that different research groups may or may not have controlled for. When precision matters, always cite the specific $K_{sp}$ value and its source/temperature rather than treating it as a universal constant.

What's the difference between "solubility" and "molar solubility"?

Solubility, reported in g/L or g per 100 mL, is the mass-based quantity often seen on solubility charts and used in practical lab work (how many grams dissolve). Molar solubility $s$ , in mol/L, is the quantity that plugs directly into equilibrium expressions like $K_{sp} = a^ab^b s^{a+b}$ . Converting between them just requires the molar mass: mass solubility = $s \times M$ , where $M$ is molar mass in g/mol. Always solve equilibrium problems in molar solubility first, then convert to mass units at the end if needed.

Solubility and the Solubility Product

Where this lives today

The physical picture

The core ideas

Worked examples

Limiting cases, scope & common traps

The chemist behind it

Before this