Acid Strength and the Leveling Effect

Drop pure HClO₄ and pure HCl into water, and the water cannot tell them apart. Both act as the same acid — H₃O⁺ — even though HClO₄ is intrinsically far stronger. This peculiar blindness is the leveling effect. It is one of the most consequential ideas in inorganic and analytical chemistry. It explains why aqueous pH scales top out near strong-acid behavior at pH 0. It explains why "superacids" had to be invented in non-aqueous media. It also explains why titrating a mixture of strong acids in water can never distinguish them. The leveling effect matters because chemists constantly need to compare acid or base strengths. These strengths often differ by many orders of magnitude — from the mild acidity of phenol to the violent proton-donating power of HSbF₆·SO₃. Water, the default solvent, imposes a hard ceiling and floor on what strength differences it can register. Any acid stronger than H₃O⁺ is forced to become H₃O⁺ in solution. This happens by an essentially quantitative proton-transfer equilibrium. Any base stronger than OH⁻ is forced to become OH⁻. This is not a minor solvent quirk. It is a direct, calculable consequence of the solvent's own autoionization equilibrium and Brønsted–Lowry proton-transfer thermodynamics. It is built on the acid–base framework of Bronsted-Lowry Acids and Bases in Inorganic Systems. The big picture is that acid strength is always relative to a solvent's own acid–base equilibrium window. This window is bounded below by the solvent's conjugate acid (the strongest species that can persist). It is bounded above by its conjugate base (the weakest acid that still reacts completely). Water's window runs from H₃O⁺ (pKa = −1.7) down to H₂O acting as an acid toward OH⁻ (pKa = 15.7). This is a span of about 17 pKa units. Anything more acidic than H₃O⁺ is leveled to H₃O⁺. Anything less acidic than water itself is not deprotonated at all. Changing the solvent changes the window: liquid ammonia, acetic acid, and sulfuric acid each impose their own leveling ceiling and floor. This is exactly why chemists reach for non-aqueous solvents. They need this to rank superacids or superbases that water cannot discriminate. This article builds the quantitative machinery — Ka, pKa, and the proton-transfer equilibrium constant — from first principles. It derives why leveling is thermodynamically inevitable once Ka exceeds the solvent autoionization constant. It works through numerical examples, including a non-aqueous solvent case. It also maps out where the leveling effect governs real laboratory practice, from acid–base titrations to the discovery of magic acid.

Key takeaways

Acid strength in solution is quantified by Ka (or pKa = −log₁₀ Ka), the equilibrium constant for HA + solvent ⇌ conjugate base + solvated proton.
The leveling effect: any acid stronger than the solvent's own conjugate acid is fully converted to that conjugate acid, so all "strong acids" look identical in a given solvent.
Water's usable acidity window runs from H₃O⁺ (pKa ≈ −1.74) to H₂O/OH⁻ (pKa ≈ 15.7); acids/bases outside this window are leveled.
The differentiating effect is the opposite regime: a solvent weak enough as a proton acceptor (e.g., glacial acetic acid) lets strong acids retain distinguishable, unequal Ka values.
Leveling is a direct thermodynamic consequence: if Ka(HA) ≫ Ka(solvent's conjugate acid), the proton-transfer equilibrium constant K ≫ 1, and the reaction runs essentially to completion.
Key idea: a solvent cannot report an acid strength beyond its own autoionization equilibrium — the solvent's conjugate acid and conjugate base define hard limits on the observable pKa scale.

Where this lives today

The leveling effect is not a historical curiosity — it governs how modern chemistry actually measures and reports acid strength. Analytical chemists titrating a mixture of strong mineral acids (say, HCl and H₂SO₄ contamination in an industrial waste stream) know that a simple aqueous acid–base titration reports only total acid content, because water levels both acids to H₃O⁺; distinguishing the two requires either non-aqueous titration or a completely different analytical method (ion chromatography, for instance). Pharmaceutical formulation chemists rely on the same logic in reverse: a basic drug with an intrinsically very high pKaH (a strong conjugate acid of the protonated drug) will simply appear "fully protonated, behaves like ammonium" in aqueous gastric fluid — its true basicity is only accessible from non-aqueous measurements or computed gas-phase proton affinities. Superacid chemistry is the leveling effect's most spectacular application. George Olah's magic acid (FSO₃H–SbF₅) and related systems (HF–SbF₅) achieve acidities far beyond anything measurable in water precisely because they are used neat, with no water present to level them down to H₃O⁺. These media protonate alkanes, generate stable carbocations for direct NMR observation, and enable petrochemical isomerization catalysis — chemistry that is thermodynamically impossible in any aqueous system because water itself would instantly be protonated and buffer the acidity at the H₃O⁺ ceiling. The Hammett acidity function H₀, built explicitly to extend "pH-like" reasoning beyond water's leveling window, is the direct intellectual descendant of the differentiating-solvent logic developed here. On the basic side, organometallic and materials chemists exploit leveling constantly: reagents like NaH, LiAlH₄, Grignard reagents, and alkali metal amides (NaNH₂) are all "leveled" instantly and violently by water to hydroxide (with hydrogen gas evolution), which is precisely why these reagents must be handled under strictly anhydrous, often aprotic, conditions — any protic solvent stronger than the reagent's conjugate acid destroys it. Non-aqueous solvent chemistry for battery electrolytes (e.g., acidic ionic liquids, HF/BF₃ systems) and for superbase-mediated organic synthesis (e.g., using liquid ammonia/NaNH₂ for terminal alkyne deprotonation) both depend on carefully matching solvent basicity/acidity to the reagent so that leveling does not erase the desired reactivity window — a design principle traceable directly to the equilibrium analysis given here.

Ask the tutor: How is the Hammett acidity function H₀ constructed to extend acidity measurement beyond water's leveling ceiling, and why can't pH itself be used for superacid media?

The physical picture

Every acid–base reaction in solution is a proton-transfer competition. When an acid HA is dissolved in a solvent SH, the fundamental equilibrium is

\text{HA} + \text{SH} \rightleftharpoons \text{SH}_2^+ + \text{A}^-

Here SH₂⁺ is the solvent's conjugate acid — H₃O⁺ in water, NH₄⁺ in liquid ammonia, CH₃COOH₂⁺ in acetic acid. The extent to which this reaction proceeds is governed by the relative proton affinities of A⁻ and SH: it is itself a Brønsted acid–base reaction, and its equilibrium constant can be built from the two separate Ka values (of HA and of SH₂⁺) exactly as any two coupled acid–base equilibria combine. The crucial asymmetry is that the solvent is both reactant and reference: SH is simultaneously the base accepting the proton and the species whose own conjugate acid, SH₂⁺, sets the strongest acid that can exist unreacted in that solvent. If HA is a stronger acid than SH₂⁺ (i.e., Ka(HA) > Ka(SH₂⁺)), then thermodynamically HA must hand its proton to SH — the reaction shown above lies overwhelmingly to the right. Once that happens, whatever excess acidic strength HA possessed above and beyond SH₂⁺'s is simply thrown away: the solution now contains SH₂⁺ (plus A⁻ as a spectator anion), indistinguishable from what you'd get by dissolving any other acid stronger than SH₂⁺. This washing-out of relative strength is the leveling effect, and it is symmetric: a base stronger than the solvent's conjugate base S⁻ is leveled to S⁻ in exactly the same way.

Figure

The acidity window: water levels everything outside H₃O⁺…OH⁻

The picture generalizes immediately to any solvent. Ammonia's conjugate acid NH₄⁺ has pKa ≈ 9.2 (in water) but the relevant leveling reference is NH₄⁺'s acidity in liquid ammonia itself; any acid stronger than NH₄⁺ in that solvent — which includes acetic acid, HCl, and even weak-in-water species like NH₃-soluble phenols — is leveled to NH₄⁺. Liquid ammonia is thus a far more powerful base solvent than water, and it correspondingly levels a huge range of acids that water differentiates perfectly well. This solvent-dependence is precisely why choosing a non-aqueous solvent is the standard trick for widening the observable acidity or basicity window — the subject taken further in Differentiating Solvents and Extending the pH Scale and, at the acidic extreme, in Superacids: Definition and Hammett Acidity.

Figure

Differentiating vs leveling solvent behavior

The core ideas

The acid dissociation (ionization) constant is the equilibrium constant for proton transfer from HA to the solvent:

K_a = \frac{[\text{SH}_2^+][\text{A}^-]}{[\text{HA}]}

where: SH₂⁺ is the solvated proton (H₃O⁺ in water), A⁻ is the conjugate base, [HA] is the equilibrium concentration of undissociated acid, and the solvent activity (taken as 1 for the dilute, nearly-pure solvent) is folded into the constant. Reading: larger Ka (smaller pKa) means the equilibrium lies further toward complete proton transfer — a stronger acid in that solvent. Because Ka can span more than 60 orders of magnitude, chemists work with

\text{p}K_a = -\log_{10} K_a

where: the negative log compresses the enormous dynamic range into a manageable linear scale; smaller (more negative) pKa = stronger acid. The solvent's own autoionization defines the ceiling and floor of the observable scale. For a solvent SH undergoing

2\,\text{SH} \rightleftharpoons \text{SH}_2^+ + \text{S}^-

the autoionization constant is

K_{\text{s}} = [\text{SH}_2^+][\text{S}^-]

where: for water this is $K_w = [\text{H}_3\text{O}^+][\text{OH}^-] = 1.0\times10^{-14}$ at 25 °C, giving $K_w=10^{-14}$ and pKw = 14. This single number sets the width of the observable acid/base strength window: the strongest acid a solvent can "host" without leveling is SH₂⁺ itself, and the weakest acid the solvent can still measure (i.e., that still shows partial, sub-unity dissociation and can act as a base itself) is SH acting on S⁻. The proton-transfer equilibrium between any two acid–base pairs combines multiplicatively. If HA is the acid of interest and SH₂⁺ is the solvent's conjugate acid, the transfer reaction

\text{HA} + \text{SH} \rightleftharpoons \text{SH}_2^+ + \text{A}^-

has equilibrium constant

K = \frac{K_a(\text{HA})}{K_a(\text{SH}_2^+)}

where: $K_a(\text{SH}_2^+)$ is the acid-dissociation constant of the solvent's conjugate acid (e.g., $K_a(\text{H}_3\text{O}^+) = 1$ by the conventional reference-state choice, since H₃O⁺ deprotonating in water regenerates H₂O + H₃O⁺, a degenerate/self-referential equilibrium — more usefully, compare Ka(HA) directly against the solvent autoionization constant, as derived below). This ratio is the master relationship: leveling occurs whenever K ≫ 1, i.e., whenever HA is a substantially stronger acid than the solvent's conjugate acid.

Derivation AWhy Ka(HA) > Ka(SH₂⁺) forces complete leveling

Goal. Show that once an acid's intrinsic Ka exceeds that of the solvent's conjugate acid by a sizeable margin, the proton-transfer equilibrium is pushed so far to the right that essentially 100% of HA is converted to SH₂⁺ + A⁻, regardless of how much stronger HA "really" is.

Step 1 — write the two half-reactions. The acid HA dissociates in solvent SH:

\text{HA} + \text{SH} \rightleftharpoons \text{SH}_2^+ + \text{A}^- \qquad K_{a,1} = K_a(\text{HA})

The solvent's conjugate acid SH₂⁺ can also, in principle, dissociate (donate a proton back to a second SH molecule — trivial/degenerate for water, but conceptually the reverse process):

\text{SH}_2^+ + \text{SH} \rightleftharpoons \text{SH}_2^+ + \text{SH} \quad(\text{degenerate})

The physically meaningful comparison is instead between HA and the solvent itself in its role as a base: the overall reaction of interest is HA transferring its proton to SH, and the question is how far this goes.

Step 2 — express the reaction quotient and completeness. Let the initial concentration of HA be $C_0$ and let $x$ be the concentration converted to SH₂⁺ + A⁻ at equilibrium. Then

K_a(\text{HA}) = \frac{x^2}{C_0 - x}

using $[\text{SH}_2^+]\approx[\text{A}^-] = x$ (ignoring background SH₂⁺ from solvent autoionization, valid when HA is not extremely dilute). Solve for the fraction dissociated, $\alpha = x/C_0$ :

\alpha = \frac{-K_a + \sqrt{K_a^2 + 4K_aC_0}}{2C_0}

This is just the standard weak-acid quadratic solution, but the key qualitative feature is in its limits.

Step 3 — take the strong-acid limit. When $K_a \gg C_0$ (i.e., Ka enormous compared to the concentration you're working at — true for any acid with pKa below about −2 at ordinary lab concentrations of 0.01–1 M), the quadratic term dominates:

\alpha \to \frac{\sqrt{K_a^2}}{2C_0}\cdot\frac{2C_0}{K_a}\Big|_{K_a\to\infty} = 1

More carefully: as $K_a/C_0 \to \infty$ , $\alpha \to 1$ to arbitrary precision. Concretely, for $C_0 = 0.10\ \text{M}$ and $K_a = 10^{7}$ (pKa = −7, like HCl), solving the quadratic gives $\alpha = 0.99999997$ — dissociation is complete to seven decimal places. This holds for any Ka exceeding roughly $10^{3}$ – $10^{4}$ at typical lab concentrations.

Step 4 — the consequence for measurement. Once $\alpha \approx 1$ for every acid with $K_a \gg C_0$ , the solution's proton activity — what a pH electrode, an indicator, or a titration endpoint actually measures — is set entirely by $C_0$ (the total acid added), not by the specific identity or intrinsic Ka of HA. Two acids with $K_a = 10^{6}$ and $K_a = 10^{20}$ , at the same formal concentration, deliver the same measured [H₃O⁺] because both are >99.999% dissociated. The solvent cannot distinguish them: it reports both as "the solvent's conjugate acid, fully formed." This is leveling, derived directly from the mass-action law rather than asserted as a rule of thumb.

Go deeper · Derivation B · 2 minThe width of the observable window: pKa(SH₂⁺) to pKa(SH) via Ks

Derivation BThe width of the observable window: pKa(SH₂⁺) to pKa(SH) via Ks

Goal. Derive the numerical span of pKa values a solvent can distinguish, using only its autoionization constant, and show this reproduces water's familiar 0-to-14 (or −1.7-to-15.7) working range.

Step 1 — define the two boundary equilibria. The upper (most acidic) boundary is set by the solvent's conjugate acid acting as an acid toward more solvent:

\text{SH}_2^+ + \text{SH} \rightleftharpoons \text{SH} + \text{SH}_2^+

which is degenerate — instead, treat the practically meaningful boundary as the point where $K_a(\text{HA}) = 1/[\text{SH}]$ , i.e., where HA is exactly as strong an acid as the solvent "typically" is toward itself. The operationally useful boundary is derived from autoionization directly.

Step 2 — relate autoionization to the two boundary pKa's. Write autoionization as the sum of two conceptual half-reactions: SH acting as an acid (toward a reference base) and SH acting as a base (toward a reference acid). By convention, in water:

K_a(\text{H}_3\text{O}^+) \equiv 1 \quad(\text{by definition, the reference acid strength})

K_a(\text{H}_2\text{O}) = \frac{K_w}{[\text{H}_2\text{O}]} = \frac{1.0\times10^{-14}}{55.5} = 1.8\times10^{-16}

so that

\text{p}K_a(\text{H}_2\text{O}) = -\log_{10}(1.8\times10^{-16}) = 15.7

This is the familiar "pKa of water" you see quoted for water acting as a Brønsted acid toward OH⁻.

Step 3 — locate the lower (basic) boundary. The strongest acid that can exist unleveled in water is H₃O⁺ itself. Using the known equilibrium $\text{H}_3\text{O}^+ \rightleftharpoons \text{H}_2\text{O} + \text{H}^+$ (solv), and the standard reference value obtained from thermodynamic cycles (heat of formation/hydration of H⁺), the accepted value is

\text{p}K_a(\text{H}_3\text{O}^+) = -1.74

(equivalently $K_a \approx 55$ , reflecting that a 1 M solution of "H₃O⁺" is still 55.5 M in total water, so the effective acidity constant of the hydronium/water couple works out to this value when referenced to the standard state of pure water as solvent).

Step 4 — compute the window width. The total observable pKa range in water is

\Delta\text{p}K_a = \text{p}K_a(\text{H}_2\text{O}) - \text{p}K_a(\text{H}_3\text{O}^+) = 15.7 - (-1.74) = 17.4

Compare this to pKw = 14.0. The difference between 17.4 and 14.0 arises because the "14" figure describes the symmetric $K_w = [\mathrm{H_3O^+}][\mathrm{OH^-}]$ product referenced to unit activities of H₃O⁺ and OH⁻, whereas the 17.4-unit window additionally accounts for the fact that pure water itself (55.5 M) is the reference state for the acid H₃O⁺, shifting the practical ceiling down by about 1.7 pKa units below 0. Both numbers are correct; they simply answer slightly different questions (the autoionization equilibrium constant vs. the fully referenced two-sided acidity scale). Either way, the qualitative result stands: water cannot discriminate acid or base strengths outside roughly a 14–17 unit window, and every acid stronger than pKa ≈ −1.7 collapses onto the single value "fully ionized, appears as H₃O⁺."

Worked examples

Example 1 — Foundational: is HI leveled or differentiated in water? HI has an intrinsic aqueous pKa of about −10 (it is one of the strongest known Brønsted acids in water, stronger even than HCl). Compare this to the ceiling pKa(H₃O⁺) = −1.74.

K = \frac{K_a(\text{HI})}{K_a(\text{H}_3\text{O}^+)} = \frac{10^{10}}{10^{1.74}} \approx 10^{8.3}

Since $K \gg 1$ (by eight orders of magnitude), the proton-transfer reaction HI + H₂O → H₃O⁺ + I⁻ runs essentially to completion. Using the quadratic from the derivation above at a realistic lab concentration $C_0 = 0.50\ \text{M}$ :

\alpha = \frac{-10^{10} + \sqrt{(10^{10})^2 + 4(10^{10})(0.50)}}{2(0.50)} \approx 1 - 5\times10^{-11}

Dissociation is complete to eleven decimal places. A pH meter in 0.50 M HI(aq) reads exactly what it reads in 0.50 M HCl(aq): pH = −log₁₀(0.50) = 0.30 (ignoring activity coefficients). Both acids are leveled to H₃O⁺; water cannot tell HI from HCl by pH alone, even though their intrinsic proton-donating powers (measurable in the gas phase or in a weaker-base solvent) differ substantially.

Example 2 — Applied: ranking acid strength using glacial acetic acid as a differentiating solvent. Glacial acetic acid (CH₃COOH, pKa of its own conjugate acid CH₃COOH₂⁺ is very low, meaning acetic acid is a much weaker base than water) is the classic differentiating solvent for strong acids. In this solvent, perchloric, sulfuric, and hydrochloric acids show genuinely different, measurable Ka values (from conductometric/potentiometric titration data, referenced to the acetic acid autoionization equilibrium):

K_a(\text{HClO}_4)_{\text{in AcOH}} \approx 10^{-4.9}, \quad K_a(\text{HCl})_{\text{in AcOH}} \approx 10^{-8.8}

(illustrative literature-consistent values; exact numbers vary slightly by source and ionic-strength convention, but the ordering HClO₄ > H₂SO₄ > HCl > HNO₃ is robust and reproducible). Take a 0.20 M solution of HClO₄ in glacial acetic acid. Using the weak-acid approximation (valid because $K_a \ll C_0$ here — this is the opposite regime from Example 1):

x^2 \approx K_a C_0 = (10^{-4.9})(0.20) = 2.5\times10^{-6}

x = [\text{CH}_3\text{COOH}_2^+] \approx 1.6\times10^{-3}\ \text{M}

\alpha = x/C_0 = 1.6\times10^{-3}/0.20 = 0.79\%

Only 0.79% of the HClO₄ is ionized — far from the >99.999% seen in water. This is the differentiating regime: because acetic acid is a much weaker proton acceptor than water, none of the strong acids reach the leveling limit, and their genuine relative strengths (HClO₄ strongest, HCl weakest, among common strong acids) become experimentally visible. This is precisely the strategy used to rank superacid strength when water's ceiling is inadequate, foreshadowing the Hammett acidity function approach used for media like fluorosulfonic acid–SbF₅ ("magic acid").

Example 3 — Exam/competition level: leveling in liquid ammonia and a mixed-solvent calculation. Liquid ammonia (bp −33 °C) autoionizes as

2\,\text{NH}_3 \rightleftharpoons \text{NH}_4^+ + \text{NH}_2^-, \qquad K_{\text{am}} \approx 10^{-33}\ (\text{at } -50\,^\circ\text{C})

Its conjugate acid NH₄⁺ has pKa (in water, as a reference point) of 9.25, but in liquid ammonia the effective ceiling for acid strength is NH₄⁺ itself — any acid stronger than the ammonium ion is leveled to NH₄⁺. Consider acetic acid (pKa 4.76 in water) dissolved in liquid ammonia. Compare its intrinsic strength to the NH₄⁺ ceiling: because acetic acid is a far stronger acid than ammonium (pKa 4.76 vs. 9.25, a difference of 4.5 orders of magnitude in Ka, and the gap is even larger once referenced against ammonia's own vastly higher basicity), acetic acid is completely leveled to NH₄⁺ in liquid ammonia:

\text{CH}_3\text{COOH} + \text{NH}_3 \longrightarrow \text{NH}_4^+ + \text{CH}_3\text{COO}^-, \qquad K \gg 1

Now ask: could you use liquid ammonia to distinguish acetic acid from, say, phenol (aqueous pKa 10.0, i.e., a much weaker acid, comparable to or slightly weaker than NH₄⁺ itself)? Since phenol's aqueous pKa (10.0) is close to NH₄⁺'s aqueous-referenced pKa (9.25), phenol is only borderline leveled — its proton-transfer equilibrium constant toward ammonia,

K = \frac{K_a(\text{PhOH})}{K_a(\text{NH}_4^+)} = \frac{10^{-10.0}}{10^{-9.25}} = 10^{-0.75} \approx 0.18

is less than 1, meaning phenol is only partially ionized in liquid ammonia and remains kinetically/thermodynamically distinguishable from fully-leveled acetic acid. This is the qualitative test examiners look for: compare Ka(HA) to Ka(solvent's conjugate acid); K ≫ 1 means full leveling, K ≈ 1 means partial/borderline behavior worth quantifying, K ≪ 1 means HA barely reacts with the solvent at all. The take-home lesson is that liquid ammonia — being a much stronger base than water — compresses the acidic end of the scale (levels many acids that water differentiates) while stretching open the basic end (it can differentiate superbases, like amide ion NaNH₂ chemistry, that water levels instantly to OH⁻).

Limiting cases, scope & common traps

Regime	Condition	What happens	Example
Strong leveling	$K_a(\text{HA}) \gg K_a(\text{SH}_2^+)$	HA fully converted to SH₂⁺; strengths indistinguishable	HClO₄, HCl, HI, H₂SO₄ in water
Differentiating	$K_a(\text{HA}) \ll K_a(\text{SH}_2^+)$ and $K_a(\text{HA})$ small relative to solvent basicity	HA only partially ionized; relative Ka values remain meaningful	Acetic, formic, benzoic acids in water; strong acids in glacial AcOH
Borderline	$K_a(\text{HA}) \approx K_a(\text{SH}_2^+)$	Partial leveling; both species present at comparable concentration	Phenol in liquid ammonia (worked above)
Base leveling	$K_b(\text{B}) \gg K_b(\text{S}^-)$	B fully converted to OH⁻ (in water); strong bases (NaOH, KOH, NaNH₂ dissolved) all appear as OH⁻	NaNH₂, NaH, organolithiums in water (violently)
No reaction (too weak)	$K_a(\text{HA})$ below solvent's own acidity as an acid	HA not deprotonated at all; solvent is "invisible" to it	CH₄, NH₃ in water (need superbase solvents/media, e.g., liquid NH₃ + NaNH₂)

The chemist behind it

The formal articulation of the leveling effect is usually credited to the broader Brønsted–Lowry acid–base framework developed in 1923, but the practical, quantitative recognition that solvents impose a strength ceiling is deeply associated with the work of Johannes Nicolaus Brønsted on relative acid strengths in different solvents, and with the American physical chemist who systematized non-aqueous acid–base scales for superacid chemistry, James Bryant Conant, whose 1920s–1930s work on "superacids" (acids stronger than sulfuric acid) in non-aqueous and mixed media first demonstrated experimentally that solvent choice reveals or conceals acid-strength differences. Conant showed that acids too strong to differentiate in water — and even in glacial acetic acid — could be ranked using indicators and non-aqueous titration, laying groundwork later extended dramatically by George Olah's superacid chemistry.

Origins · 1 minJames Bryant Conant and how this was discovered

James Bryant Conant

1893–1978 · Harvard physical organic chemist; later Harvard president and U.S. science administrator

Conant's early research (1920s) into the acid strength of very strong acids in non-aqueous solvents — establishing that perchloric acid behaves as a substantially stronger acid than sulfuric or hydrochloric acid once a differentiating (weak-base) solvent is used — provided the first quantitative experimental demonstration of what would later be called the leveling and differentiating effects. He coined the term "superacid" for acids capable of protonating very weak bases such as ketones, which water and even acetic acid cannot register as acidic media at all.

History

- 1884 — Svante Arrhenius proposes ionic dissociation, the first quantitative acid–base framework and precursor to Ka.

- 1909 — Sørensen introduces the pH scale, motivating a logarithmic treatment of proton concentration.

- 1923 — Brønsted and (independently) Lowry define acids and bases via proton transfer, providing the language for "conjugate acid of the solvent."

- 1927 — Conant and collaborators demonstrate quantitative differentiation of strong-acid strengths using glacial acetic acid as solvent.

- 1966 onward — George Olah develops magic acid (FSO₃H–SbF₅) and the Hammett acidity function framework, extending the differentiating-solvent idea far beyond water's leveling ceiling.

What exactly is being "leveled" — the acid or the solution's measured acidity?

It is the observed, solution-phase acidity that gets leveled, not the acid's intrinsic thermodynamic strength. HA's gas-phase proton affinity or its Ka in a weaker-base solvent still reflects its true strength. What water erases is the ability to detect that strength difference in aqueous solution, because the proton-transfer equilibrium HA + H₂O ⇌ H₃O⁺ + A⁻ runs to completion for any sufficiently strong acid. The solution then simply contains H₃O⁺ at a concentration set by the total acid added, with no memory of which acid supplied the proton.

Why is water's leveling window quoted sometimes as 14 units and sometimes as about 17.4 units?

The number 14 comes directly from pKw = −log₁₀(Kw) = −log₁₀(10⁻¹⁴), the autoionization constant referenced to unit activities of H₃O⁺ and OH⁻. The 17.4-unit figure additionally accounts for water's own standard-state concentration (55.5 M) entering the reference state for H₃O⁺ as an acid, shifting the practical ceiling to pKa(H₃O⁺) ≈ −1.74 rather than 0. Both numbers are self-consistent within their own conventions; the 17.4-unit window is the more chemically meaningful one for comparing arbitrary acids and bases side by side on a single pKa axis.

Does the leveling effect apply to weak acids too, or only to "strong" acids?

It applies to any acid whose Ka exceeds the solvent's conjugate-acid reference by enough to push the proton-transfer equilibrium constant well above 1. In water this threshold sits near pKa ≈ −1.7; acids with pKa above that (acetic acid, HF, phenol, HCN) are not leveled and show genuinely different, measurable Ka values — this is precisely why weak-acid chemistry (buffers, titration curves with real inflection shapes) works at all in water. Leveling is specifically a strong-acid (or strong-base) phenomenon, bounded by the solvent's own equilibrium window.

Can a solvent level acids without itself undergoing autoionization, e.g., an aprotic solvent?

Aprotic solvents (acetonitrile, DMSO, THF) do not have their own SH ⇌ SH₂⁺ + S⁻ autoionization in the same protic sense, but they still have a "leveling" reference point: the conjugate acid of any adventitious proton source, or more precisely, the strongest acid the solvent itself can stabilize without further reaction. In practice, aprotic solvents are excellent differentiating media for strong acids and superbases precisely because they lack a strongly competing proton-accepting or proton-donating equilibrium — this is why pKa scales in DMSO and acetonitrile span far more than water's ~17 units and are the standard media for measuring superacid and superbase strengths quantitatively.

How does the leveling effect connect to the choice of indicator in a titration?

An acid–base indicator is itself a weak acid/base pair with its own pKa, and it can only report distinctions that lie within the solvent's observable window. Titrating a leveled strong acid in water with any indicator (phenolphthalein, methyl orange) gives an endpoint determined by total moles of H₃O⁺ present, not by which strong acid contributed it — because the analyte itself has already been leveled before the indicator ever "sees" it. This is why aqueous acid–base titration can measure total acidity of a mixture of strong acids but never resolve their individual identities or concentrations without an independent method.

Is the leveling effect relevant to biological chemistry, given that cells are aqueous?

Yes, profoundly. Physiological pH (~7.4) sits well inside water's differentiating window, which is exactly why biology can exploit a huge range of finely tuned pKa values (from strongly acidic aspartate/glutamate side chains, pKa ≈ 4, to strongly basic arginine, pKa ≈ 12.5) for enzyme catalysis, proton pumps, and buffering. If water leveled everything the way it levels HCl, biochemistry's exquisite pH-dependent machinery (histidine's pKa ≈ 6 acting as a near-perfect physiological proton shuttle) simply could not exist — biological acid–base chemistry depends on operating deep inside, not at the edges of, the aqueous window.

Why can't we just use a very dilute solution to avoid leveling and see the "true" Ka of a strong acid?

Diluting shifts the equilibrium position (via the $\alpha$ vs. $C_0$ relationship derived above) but does not change $K_a$ itself, and for acids with $K_a$ many orders of magnitude above the solvent's reference ceiling, even extreme dilution (down to $10^{-6}$ M or below, where activity coefficient and autoionization corrections become significant anyway) still gives $\alpha$ close to 1. Practically, by the time dilution is extreme enough to show partial dissociation, the solution's own background H₃O⁺ from water autoionization dominates the equilibrium, so no clean measurement of the acid's intrinsic Ka is recoverable this way — non-aqueous solvents remain the only reliable route.