Water as a Biological Solvent

Dielectric-continuum solvation models descended directly from the Born equation — generalized Born (GB) models and Poisson–Boltzmann solvers — are workhorse tools in modern computational structural biology, used to estimate solvation free energies in molecular dynamics simulations of protein folding, ligand binding, and drug design without paying the enormous computational cost of simulating every explicit water molecule. Pharmaceutical companies routinely use GB/PB-based scoring functions to rank candidate drug molecules by predicted binding free energy to a target protein, where correctly modeling how a charged or polar drug substituent trades its favorable bulk hydration for a (hopefully more favorable) interaction inside a partially desolvated, low-dielectric binding pocket is often the single largest term in the calculation. The competition between lattice energy and hydration free energy set up in this article underlies the practical chemistry of physiological electrolyte balance: intravenous saline solutions, oral rehydration therapy, and the renal handling of Na⁺, K⁺, Ca²⁺, and Mg²⁺ all depend on these ions remaining fully, favorably hydrated in aqueous body fluids rather than precipitating — a balance that shifts pathologically in conditions like kidney stone formation (calcium oxalate/phosphate crystallizing when local concentration exceeds solubility set by the hydration/lattice competition) or hypercalcemia. Inside cells and organelles, dielectric screening explains why highly charged biomolecules — the phosphate backbone of DNA, the polyanionic surface of ribosomes, the phosphorylated tails of many signaling proteins — do not simply collapse into ion pairs with every nearby cation, but instead diffuse and interact reversibly, allowing dynamic assembly and disassembly of the complexes that transcription, translation, and signal transduction depend on. The same screening reasoning explains why Mg²⁺, precisely because of its very large negative Born-type hydration free energy (Worked Example 2), is such an effective and selective cofactor for stabilizing the highly charged transition states of phosphoryl-transfer enzymes (kinases, polymerases, ATPases) — it can shed part of its hydration shell to coordinate directly to enzyme and substrate phosphate oxygens only when the enzyme's active site provides a compensating, sufficiently favorable electrostatic environment, giving these enzymes part of their exquisite substrate specificity.

Ask the tutor: How does the picture change when the "ion" is actually a highly charged, extended surface — like the phosphate backbone of DNA — rather than a simple point-like sphere?

Water's utility as a solvent traces to one geometric fact: the molecule is bent (H–O–H angle 104.5°) with oxygen far more electronegative than hydrogen, so it carries a permanent dipole moment of 1.85 debye (6.2 × 10⁻³⁰ C·m) in the gas phase, rising effectively to roughly 2.3–2.6 D in the liquid due to cooperative polarization. Because each molecule is small, roughly tetrahedral in its local hydrogen-bonding geometry, and free to rotate and translate, an enormous number of these dipoles can pack around any charged or polar object and point their negative (oxygen) or positive (hydrogen) ends directly at it. When an ion — say Na⁺ — is placed in water, the first thing that happens is reorientation, not new bond formation from scratch: nearby water molecules simply rotate so that their oxygen lone pairs face the cation, while around an anion like Cl⁻ the O–H bonds swivel to point their partial-positive hydrogens inward. This creates a primary hydration shell, a shell of water molecules in direct, quasi-fixed electrostatic contact with the ion, typically 4–6 molecules for a small cation and hydrogen-bonded arrangements for anions. Beyond this, a looser secondary shell of partially perturbed water extends the influence outward before merging into bulk, hydrogen-bonded water a few molecular diameters away. The number of water molecules in the primary shell and how tightly they are held (measured by residence times, from picoseconds for Cs⁹ to microseconds for Mg²⁺) both scale with the ion's charge density — charge divided by ionic radius. The second, complementary way to describe this same physics is macroscopic: instead of tracking individual dipole reorientations, treat water as a continuous dielectric medium with a bulk relative permittivity ε_r ≈ 78.5 at 25 °C. Any two charges immersed in this medium feel a Coulomb force reduced by a factor of ε_r relative to the same charges in vacuum, because the intervening water dipoles polarize and partially cancel the field lines running between them. This dielectric-continuum view and the molecular hydration-shell view are two levels of description of one underlying phenomenon — orientational polarization of a polar solvent — and both are needed: the continuum model gives fast, useful estimates of ionic interactions at distance, while the molecular shell picture explains structure, coordination number, and kinetics close to the solute.

The Coulomb interaction between two point charges separated by distance $r$ in a homogeneous medium of relative permittivity $\varepsilon_r$ is

where:

• $q_1, q_2$ are the charges (C), including sign,
• $\varepsilon_0 = 8.854\times10^{-12}\ \mathrm{C^2\,N^{-1}m^{-2}}$ is the vacuum permittivity,
• $\varepsilon_r$ is the medium's relative permittivity (dimensionless; 1 for vacuum, ≈78.5 for liquid water at 25 °C, ≈2 for hydrocarbon interiors),
• $r$ is the center-to-center separation (m).

Reading: whatever the vacuum electrostatic energy would be, immersing the pair in water divides it by ~80. This single rescaling is the quantitative essence of why polar/ionic solutes dissolve and why oppositely charged groups inside a protein or between two DNA strands feel a far weaker pull in water than the same groups would in vacuum or in a lipid interior. The free energy of transferring a single spherical ion of charge $ze$ and radius $r_{\text{ion}}$ from vacuum into a continuum dielectric is given by the Born equation:

where:

• $z$ is the ion's charge number (e.g. +1 for Na⁺, +2 for Mg²⁺),
• $e = 1.602\times10^{-19}$ C is the elementary charge,
• $r_{\text{ion}}$ is the effective (Born) radius of the ion in the solvent (m) — larger than the crystallographic radius because it must enclose the point of closest solvent dipole approach,
• $\varepsilon_r$ is the solvent's relative permittivity,
• $N_A$ converts a per-ion energy to a per-mole free energy.

Reading: hydration free energy scales as $-z^2/r_{\text{ion}}$, so doubly and triply charged ions (Mg²⁺, Ca²⁺, Al³⁺, PO₄³⁻ groups) are hydrated far more strongly than singly charged ions of similar size, and small ions are hydrated more strongly than large ones of the same charge — both trends are seen directly in real hydration free energies (Table below).

Example 1 — Foundational: Coulomb energy of an ion pair, vacuum vs. water. Take Na⁺ and Cl⁻ just touching at contact distance $r = 0.28\ \text{nm} = 2.8\times10^{-10}\ \text{m}$ (sum of their ionic radii, 0.102 nm + 0.181 nm ≈ 0.283 nm, rounded). Charges are $q_1 = +e$, $q_2 = -e$, with $e = 1.602\times10^{-19}\ \text{C}$. In vacuum ($\varepsilon_r = 1$):

Multiplying by Avogadro's number to get a molar quantity:

This is comparable in magnitude to a covalent bond — vacuum electrostatics between adjacent ions is enormous, which is exactly why the NaCl lattice is so stable in dry form. In water ($\varepsilon_r = 78.5$), the same geometry gives

This is barely larger than thermal energy $RT = 2.48\ \text{kJ/mol}$ at 298 K, so the ion pair readily separates and diffuses apart — the qualitative reason salt dissociates completely in water instead of existing as tight ion pairs. As a limiting-case check, if $\varepsilon_r \to 1$ (as if the ions were still in vacuum or deep in a nonpolar solvent), $U_{\text{water}} \to U_{\text{vac}} = -496\ \text{kJ/mol}$, recovering the un-screened result exactly, confirming the screening factor is applied correctly.

Example 2 — Applied: Born hydration free energy of Mg²⁺ vs. Na⁺. Use effective Born radii $r_{\text{Mg}} = 0.086\ \text{nm}$ and $r_{\text{Na}} = 0.116\ \text{nm}$ (these Born/hydration radii are somewhat larger than bare crystallographic radii, reflecting the closest approach of a solvent dipole's center rather than its point charge; values chosen here to match commonly tabulated single-ion Born radii). Take $\varepsilon_r = 78.5$. For Na⁺ ($z=1$):

Evaluating the prefactor first:

With $(1-1/78.5) = 0.9873$:

This overshoots the experimental Na⁺ hydration free energy (≈ –375 to –411 kJ/mol depending on convention) because the simple Born model ignores the discreteness of the first hydration shell and dielectric saturation right next to the ion — a known, expected limitation flagged explicitly below — but it correctly captures the trend. For Mg²⁺ ($z=2$, $r=0.086\ \text{nm}$), note that $z^2$ quadruples the numerator while $r$ shrinks, both pushing the energy more negative:

Scaling from the Na⁺ prefactor (which used $r=1.16\times10^{-10}$ m) by the radius ratio and the charge-squared factor:

Mg²⁺ is hydrated over five times more strongly than Na⁺ despite being a smaller ion — the $z^2$ dependence dominates. This is exactly why Mg²⁺ in biology (e.g., bound in ATP-Mg²⁺ complexes, or coordinating phosphate backbones in ribozymes) retains a tightly bound, slowly exchanging hydration shell, while Na⁺ ions exchange with bulk water on the sub-nanosecond timescale.

Example 3 — Exam-level: dielectric screening inside a protein interior vs. bulk water. A buried salt bridge in a protein — say a Lys⁺ side chain and an Asp⁻ side chain — sit with charge centers 0.45 nm apart in a local environment approximated as having effective dielectric constant $\varepsilon_{\text{eff}} = 4$ (typical of a protein interior, intermediate between vacuum and bulk water because of partial solvent penetration and the polarizability of the protein matrix), compared to the same pair fully exposed at the protein surface where $\varepsilon_{\text{eff}} \approx 40$ (partially screened bulk water, reduced from 78.5 by the nearby low-dielectric protein surface). Buried case:

First compute the vacuum-equivalent prefactor at this distance:

Dividing by $\varepsilon_r = 4$ (using the attractive sign, one positive one negative charge):

Surface case, same distance but $\varepsilon_r = 40$:

The buried salt bridge is about ten times stronger than the surface-exposed one purely because of dielectric screening, even though the distance and charges are identical. This is precisely why buried ion pairs can contribute substantially to protein stability (tens of kJ/mol) while surface salt bridges typically contribute only a few kJ/mol — an effect well documented in protein engineering and mutagenesis studies of thermostability. As a limiting-case check, if the "buried" environment were assigned bulk water's $\varepsilon_r = 78.5$ instead of 4, $U$ would fall to $-3.9$ kJ/mol, near thermal energy, showing why fully solvent-exposed salt bridges are only marginally stabilizing — consistent with experiment.

The modern quantitative treatment of ion solvation traces to Max Born, who in 1920 published the continuum electrostatic model — the Born equation derived above — while working at the University of Göttingen in a period of intense activity in theoretical physics and physical chemistry. Born is far better known to most students for his foundational contributions to quantum mechanics (the Born rule for probability amplitudes, and work with Heisenberg and Jordan on matrix mechanics), for which he shared the 1954 Nobel Prize in Physics. His 1920 ionic solvation paper was, by comparison, a smaller and more applied piece of work, but it supplied biochemistry and physical chemistry with the first workable bridge between microscopic charge and macroscopic dielectric response — a tool still taught, cited, and extended (via generalized Born and Poisson–Boltzmann continuum solvent models) in every modern computational chemistry and structural biology curriculum. Born's broader career illustrates how deeply intertwined early 20th-century atomic physics and physical chemistry were: the same electrostatic reasoning that let physicists model electron distributions and ionic crystals (his collaboration with Alfred Landé and later with Max von Laue and others on lattice energies directly informs the "lattice energy" side of the dissolution cycle in this article) also gave chemists a first-principles handle on why salts dissolve. The lattice-energy concept itself descends from the Born–Landé equation, a close cousin of the solvation formula, both built from the same electrostatic self-energy calculation applied to different geometries (a periodic lattice vs. an isolated sphere in a dielectric).