EDTA Titration Curves

EDTA titrimetry, despite the rise of atomic absorption and ICP-based instrumentation, remains the standard method for total water hardness reported in mg/L as CaCO$_3$ in municipal and industrial water testing, precisely because the conditional-constant framework lets a technician predict, with a pH 10 ammonia buffer and Eriochrome Black T, exactly where the endpoint will fall and how sharp it will be. Pharmaceutical quality control uses EDTA back-titrations (excess EDTA added, then back-titrated with a standard Zn$^{2+}$ or Mg$^{2+}$ solution) to assay metal content in formulations and to determine calcium or aluminum levels in antacids, exploiting exactly the same pM-vs-V mathematics but run in reverse. In clinical chemistry, EDTA itself (as the disodium or dipotassium salt) is used as an anticoagulant in blood collection tubes because it sequesters the Ca$^{2+}$ needed for the clotting cascade — a direct biological application of the same high-$K_f'$ chelation chemistry. Industrially, selective masking strategies like the fluoride-masking of Al$^{3+}$ demonstrated above are used routinely in alloy and ore analysis to titrate one metal (Zn, Cu, or Ni) in a matrix containing several others, and the same $\alpha_M$-based logic underlies the design of chelation therapy dosing for heavy-metal poisoning, where DTPA and EDTA derivatives are chosen and pH-buffered specifically to maximize $K_f'$ for the toxic metal (Pb$^{2+}$, Hg$^{2+}$) while minimizing loss of essential Ca$^{2+}$ or Zn$^{2+}$.

Ask the tutor: How would this curve change if the EDTA titrant itself were only partially standardized, introducing a systematic error in $C_Y^0$?

EDTA (ethylenediaminetetraacetic acid, $\text{H}_4\text{Y}$) is a hexadentate ligand: two amine nitrogens and four carboxylate oxygens can all coordinate a single metal ion, forming a nearly universal 1:1 complex MY with formation constant $K_f = [\text{MY}]/([\text{M}][\text{Y}^{4-}])$ regardless of the metal's usual coordination number. This 1:1 stoichiometry is what makes EDTA titrimetry so clean — no fractional equivalence points, no stepwise complex formation to track, just a single sharp reaction

The catch is that $[\text{Y}^{4-}]$ is almost never the dominant form of EDTA in solution. At low pH, EDTA is protonated to $\text{H}_2\text{Y}^{2-}$, $\text{HY}^{3-}$, or even $\text{H}_4\text{Y}$; only a small fraction $\alpha_{Y^{4-}}(\text{pH})$ exists as the fully deprotonated, metal-binding species. Simultaneously, the metal ion may be tied up by hydroxide (hydrolysis) or by an auxiliary ligand deliberately added to prevent precipitation — ammonia for Zn$^{2+}$ or Cu$^{2+}$, tartrate for some metals — so only a fraction $\alpha_M$ of total metal is "free" $\text{M}^{n+}$ available to react. Both fractions are captured by side-reaction coefficients, and folding them into $K_f$ gives the conditional formation constant:

This single number, evaluated at the working pH and auxiliary-ligand concentration, behaves exactly like an ordinary formation constant for computing the titration curve — provided we now track total (analytical) concentrations $C_M$ and $C_Y$ rather than free-ion concentrations.

The overall shape of the curve — a slow rise, a steep break near equivalence, then a slow leveling off — is generic to any 1:1 complexation titration, but the height of the break depends entirely on $K_f'$. A larger $K_f'$ pushes the pre-equivalence branch down (less free metal at a given fraction titrated) and the post-equivalence branch flatter, producing a taller vertical jump. This is exactly the same logic as a strong-acid/strong-base curve being sharper than a weak-acid/weak-base curve: $K_f'$ plays the role that $K_a K_b / K_w$ (or simply $K_a$ for the reaction with strong base) plays there.

The titration reaction is treated as if it goes to completion at each point (valid when $K_f'$ is large), and the curve is built from mass balance in three regions. Let $C_M^0, V_M$ be the initial metal molarity and volume, and $C_Y^0$ the EDTA titrant molarity, with $V$ the volume of titrant added and $V_{eq} = C_M^0 V_M / C_Y^0$ the equivalence volume.

Region 1 (before equivalence, $V < V_{eq}$). Only stoichiometric MY has formed; the excess, un-reacted metal dominates $[\text{M}]$ because $K_f'$ is large enough that essentially none of the MY re-dissociates:

where: $C_M^0 V_M$ = initial moles of metal, $C_Y^0 V$ = moles of EDTA added so far, $V_M+V$ = total solution volume.

Region 2 (at equivalence, $V = V_{eq}$). All metal is nominally converted to MY, but MY dissociates slightly, releasing equal amounts of free M and free (total) Y:

where: $C_{MY}$ = analytical concentration of the complex at the equivalence point (essentially all metal present), $K_f'$ = conditional formation constant at the working pH/ligand conditions.

Region 3 (after equivalence, $V > V_{eq}$). Excess EDTA is present in known analytical excess; essentially all metal is complexed, and the tiny residual free metal is obtained from the $K_f'$ expression using the known excess of total EDTA:

where: $C_{Y,\text{excess}}$ = analytical excess EDTA concentration (all forms, since $\alpha_{Y^{4-}}$ is already folded into $K_f'$), $C_{MY}$ = concentration of complex (≈ total metal, still ≈ $C_M^0V_M/(V_M+V)$ since dilution continues). Reading these three equations together: Region 1 is a dilution problem, Region 3 is a mass-action problem, and Region 2 is the delicate crossover where both effects balance — exactly the same three-region logic used for strong-acid/strong-base curves, with "excess H$^+$", "$\sqrt{K_w}$", and "excess OH$^-$" replaced by "excess M", "$\sqrt{C_{MY}/K_f'}$", and "excess Y (as $K_f'$ term)".

Example 1 — Foundational: Ca$^{2+}$ titrated with EDTA at pH 10. A 50.00 mL aliquot of 0.01000 M Ca$^{2+}$ is titrated with 0.01000 M EDTA at pH 10.00, buffered with NH$_3$/NH$_4^+$ (assume Ca$^{2+}$ has negligible ammine complexation — a good approximation, since Ca$^{2+}$ does not bind ammonia appreciably). At pH 10.00, $\alpha_{Y^{4-}} = 0.35$ (a standard tabulated value), and $\log K_f(\text{CaY}^{2-}) = 10.69$, so $K_f = 4.9\times10^{10}$. Conditional constant:

$V_{eq} = (0.01000\,\text{M})(50.00\,\text{mL})/(0.01000\,\text{M}) = 50.00$ mL. At V = 10.00 mL (before equivalence):

At V = 50.00 mL (equivalence point):

At V = 60.00 mL (after equivalence, 10.00 mL excess EDTA):

The jump from pCa = 2.18 at 10 mL before equivalence to pCa = 9.53 at 10 mL after equivalence — nearly 7.4 pCa units over a 20 mL span straddling $V_{eq}$ — is the sharp break that makes an indicator color change visible. Limiting-case check: if $\alpha_{Y^{4-}}$ were instead 1.0 (pH → 12, fully deprotonated EDTA), $K_f'$ rises to $4.9\times10^{10}$, pCa$_{eq}$ rises only slightly (to 6.50, since it depends on the square root of $K_f'$), but the post-equivalence branch drops further, since it depends on $K_f'$ linearly — confirming that raising pH sharpens the tail of the curve more than the equivalence point itself.

Example 2 — Applied: water-hardness titration with Mg$^{2+}$ interference and Eriochrome Black T at pH 10. A 100.0 mL water sample containing Ca$^{2+}$ and Mg$^{2+}$ at a combined total-hardness concentration of $3.00\times10^{-3}$ M (as total divalent cation) is titrated with 0.01000 M EDTA at pH 10.00, buffered with NH$_3$/NH$_4^+$ at total ammonia concentration such that $[\text{NH}_3]_{free} = 0.10$ M. Because both metals form 1:1 complexes of comparable strength with EDTA (log $K_f$ = 10.69 for CaY$^{2-}$, 8.79 for MgY$^{2-}$), total hardness titrates as if a single "effective" metal were present, and $V_{eq}$ is governed by total moles of M$^{2+}$.

Because Mg$^{2+}$ is weaker (lower $K_f$), the last metal to be complexed near equivalence is predominantly Mg$^{2+}$, which is exactly why Eriochrome Black T (whose Mg-indicator complex, MgIn$^-$, has a convenient intermediate stability) is used for hardness titrations rather than an indicator tuned to Ca$^{2+}$ alone. For Mg$^{2+}$, we must also account for ammine-complex side reactions since Mg$^{2+}$ weakly complexes NH$_3$; take $\alpha_{Mg} \approx 0.98$ at $[\text{NH}_3] = 0.10$ M (weak binding, small correction). With $\alpha_{Y^{4-}} = 0.35$ at pH 10.00:

At the equivalence point, treating the sample as effectively all-Mg (worst case, since Mg is titrated last and controls the sharpness of the visible endpoint region):

Compare this to the Ca-only case above (pCa$_{eq}$ = 6.27, $K_f' = 1.7\times10^{10}$): the presence of the weaker Mg$^{2+}$-EDTA equilibrium, controlling the tail behavior, gives a noticeably less deep equivalence point (pM = 5.48 vs 6.27), consistent with the general rule that mixed-metal titrations show a break governed by the weakest complex present at concentrations comparable to the strongest. This is precisely why hardness titrations at pH 10 (rather than pH 12, where Mg(OH)$_2$ would precipitate and remove Mg from the titration entirely) require careful buffer control — an issue picked up in full in EDTA Titration Techniques and Selectivity.

Example 3 — Exam/competition level: masking and pH choice to resolve Zn$^{2+}$ in the presence of Al$^{3+}$. A solution is $5.00\times10^{-3}$ M in both Zn$^{2+}$ and Al$^{3+}$ (50.00 mL total), and the analyst wants to titrate only Zn$^{2+}$ with 0.01000 M EDTA while Al$^{3+}$ is masked as $\text{AlF}_6^{3-}$ using excess NaF (final $[\text{F}^-]_{free} = 0.050$ M). Data: $\log K_f(\text{ZnY}^{2-}) = 16.50$, $\log K_f(\text{AlY}^-) = 16.13$ — normally far too close to resolve by pH alone, since both are similar in magnitude and Al$^{3+}$ would co-titrate. Masking strategy: fluoride forms an extremely stable $\text{AlF}_6^{3-}$ complex ($\log \beta_6 \approx 19.8$ overall), which drastically lowers $\alpha_{Al}$ (the fraction of Al present as "reactive," EDTA-available species — free Al$^{3+}$ plus any partially fluorinated species not accounted for as fully masked). For a well-designed mask, essentially all Al is converted to AlF$_6^{3-}$, so treat $\alpha_{Al} \to 0$ for the purposes of this titration reaction (Al no longer competes for EDTA at all). Working pH is chosen at pH 5.00 (using an acetate buffer) — high enough to keep Zn$^{2+}$-EDTA strongly favored, low enough that Zn(OH)$_2$ does not precipitate and that the fluoride mask on Al remains intact (F$^-$ is a weaker base than needed to be protonated away at pH 5). At pH 5.00, $\alpha_{Y^{4-}} = 2.5\times10^{-7}$ (heavily protonated EDTA — a realistic tabulated value at this pH). Zinc does not appreciably complex with acetate or fluoride under these conditions, so $\alpha_{Zn} \approx 1$.

Despite the drastic drop from $K_f = 3.16\times10^{16}$ down to $K_f' = 7.9\times10^{9}$ at low pH, this conditional constant is still comfortably above the $\sim10^8$ threshold needed for a sharp break (see the scope table below), because Zn–EDTA is intrinsically so stable. Compute $V_{eq}$ for Zn$^{2+}$ alone:

At the equivalence point:

At 5.00 mL past equivalence (V = 30.00 mL):

Because Al$^{3+}$ has been almost completely masked as AlF$_6^{3-}$ ($\alpha_{Al} \approx 0$ for reaction with EDTA), it contributes essentially no competing consumption of titrant, and the curve behaves as a clean single-metal Zn$^{2+}$ titration with break pZn: 6.19 → 9.20 across the equivalence region — sharp enough for a metal-ion indicator to detect. Limiting-case check: if the fluoride mask were absent ($\alpha_{Al}\to 1$, Al$^{3+}$ competing fully), both metals would titrate together, $V_{eq}$ would double to 50.00 mL, and the "equivalence point" observed would correspond to total metal, destroying the selectivity the analyst needed — precisely the failure mode that motivates masking chemistry.

Gerold Schwarzenbach (1904–1978), a Swiss inorganic and analytical chemist working at ETH Zürich, is the figure most responsible for turning EDTA from a curious industrial chelating agent (originally developed for water softening and textile processing in the 1930s–40s) into the backbone of quantitative complexometric analysis. Schwarzenbach's systematic studies through the late 1940s and 1950s established the stability-constant framework for polyaminocarboxylate ligands, introduced the side-reaction-coefficient ($\alpha$) formalism that makes the conditional constant calculable from tabulated acid-dissociation and formation constants, and worked out — largely through his own careful potentiometric and spectrophotometric measurements — the pH dependence that determines which metals can be selectively titrated at which pH. He also pioneered the practical use of metal-ion indicators (in close collaboration with colleagues developing dyes such as Eriochrome Black T) so that the sharp equivalence points his theory predicted could actually be observed by eye in a student or industrial laboratory, without a pH meter or potentiometer.