Acid-base theories: Lewis acids, HSAB, and Drago-Wayland parameters

Intuition Beginner

Bronsted-Lowry theory defines acids as proton donors and bases as proton acceptors. This covers many reactions but misses an entire class: reactions where no proton is transferred. G. N. Lewis proposed a broader definition. A Lewis acid is any species that accepts a pair of electrons. A Lewis base is any species that donates a pair of electrons. Under this definition, BF${}_3$ is an acid (the boron atom has an empty orbital that accepts electrons) and NH${}_3$ is a base (nitrogen has a lone pair to donate), even though no H${}^{+}$ is involved.

Not all Lewis acids and bases behave the same way. Some are hard: small, with high charge density and low polarisability. Hard acids include Na${}^{+}$, Mg${}^{2+}$, and Al${}^{3+}$. Hard bases include F${}^{-}$, OH${}^{-}$, and NH${}_3$. Others are soft: large, with low charge density and high polarisability. Soft acids include Ag${}^{+}$, Hg${}^{2+}$, and Pt${}^{2+}$. Soft bases include I${}^{-}$, SH${}^{-}$, and phosphines.

The HSAB principle (Hard and Soft Acids and Bases), proposed by Ralph Pearson in 1963, states that hard acids prefer to bind hard bases, and soft acids prefer to bind soft bases. "Like prefers like" in hardness or softness. Hard-hard interactions are electrostatic — ionic or strongly polar bonds driven by charge attraction. Soft-soft interactions are covalent — bonds driven by orbital overlap between large, polarisable electron clouds.

Visual Beginner

The diagram places hard acids and hard bases on the left, soft acids and soft bases on the right. Solid lines connect preferred pairings (hard-hard and soft-soft). Crossed lines show disfavoured pairings (hard-soft). Each acid and base is labelled with a representative example: Li${}^{+}$, Mg${}^{2+}$, Al${}^{3+}$ for hard acids; I${}^{-}$, SH${}^{-}$, CN${}^{-}$ for soft bases; and so on.

Worked example Beginner

Predict which halide ion binds more strongly to Ag${}^{+}$: F${}^{-}$ or I${}^{-}$?

Silver ion (Ag${}^{+}$) is a soft acid. It is a large cation with a low charge density and highly polarisable d-electrons.

Fluoride (F${}^{-}$) is a hard base — small, high charge density, low polarisability. Iodide (I${}^{-}$) is a soft base — large, low charge density, high polarisability.

By the HSAB principle, the soft acid Ag${}^{+}$ prefers the soft base I${}^{-}$ over the hard base F${}^{-}$.

This prediction matches experiment. Silver iodide (AgI) is far less soluble in water than silver fluoride (AgF). AgI precipitates readily from solution, while AgF is quite soluble. The strong soft-soft covalent interaction between Ag${}^{+}$ and I${}^{-}$ drives the formation of the insoluble solid.

Check your understanding Beginner

Formal definition Intermediate+

Lewis acid-base theory (G. N. Lewis, 1923) defines an acid as an electron-pair acceptor (electrophile) and a base as an electron-pair donor (nucleophile). The reaction between a Lewis acid A and a Lewis base B forms an adduct A–B in which the base donates a lone pair into an empty orbital on the acid:

$$\mathrm{A}+\mathrm{B}\to \mathrm{A}-\mathrm{B}$$

This definition subsumes Bronsted-Lowry acid-base chemistry: H${}^{+}$ is a Lewis acid (it accepts a lone pair from H${}_2$O to form H${}_3$O${}^{+}$), and any Bronsted base is also a Lewis base. But the Lewis definition extends far beyond proton transfer to include coordination complexes, solvation, and reactions with no proton involvement.

The HSAB principle (Pearson, 1963) classifies Lewis acids and bases on a hard-soft spectrum:

• Hard acids have small ionic radii, high positive charge, and low polarisability. Examples: H${}^{+}$, Li${}^{+}$, Na${}^{+}$, Mg${}^{2+}$, Ca${}^{2+}$, Al${}^{3+}$, Cr${}^{3+}$, BF${}_3$.
• Soft acids have large radii, low or zero charge, and high polarisability. Examples: Cu${}^{+}$, Ag${}^{+}$, Au${}^{+}$, Hg${}^{2+}$, Pt${}^{2+}$, I${}_2$, BH${}_3$.
• Hard bases have high electronegativity, low polarisability, and are difficult to oxidise. Examples: F${}^{-}$, OH${}^{-}$, H${}_2$O, NH${}_3$, Cl${}^{-}$, CO${}_3^{2-}$.
• Soft bases have low electronegativity, high polarisability, and are easily oxidised. Examples: I${}^{-}$, SH${}^{-}$, SCN${}^{-}$, CN${}^{-}$, CO, PR${}_3$.

The principle: hard acids form more stable complexes with hard bases (electrostatic, ionic bonding dominates), and soft acids form more stable complexes with soft bases (covalent, orbital-overlap bonding dominates). Borderline species (Fe${}^{2+}$, Co${}^{2+}$, Ni${}^{2+}$, Br${}^{-}$, NO${}_2^{-}$) fall between and can pair with both hard and soft partners.

The Drago-Wayland equation (1965) provides a quantitative alternative to HSAB. It correlates the enthalpy of Lewis acid-base adduct formation using two parameters for each species:

$$-\Delta H_{\mathrm{AB}}=E_{\mathrm{A}}{E}_{\mathrm{B}}+C_{\mathrm{A}}{C}_{\mathrm{B}}$$

where $E$ (electrostatic) measures susceptibility to ionic interactions and $C$ (covalent) measures susceptibility to covalent interactions. Subscript A denotes the acid, B the base. The equation predicts adduct enthalpies from tabulated $E$ and $C$ values without requiring a hardness classification. The sign convention makes $-\Delta H$ positive for exothermic adduct formation.

Key result Intermediate+

Result (Absolute hardness from HOMO-LUMO gap). Pearson and Parr showed that the absolute hardness $\eta$ of a chemical species can be defined quantitatively using the HOMO and LUMO energies:

$$\eta =\frac{I-A}{2}=\frac{-{ϵ}_{\mathrm{LUMO}}+{ϵ}_{\mathrm{HOMO}}}{2}\approx \frac{{ϵ}_{\mathrm{LUMO}}-{ϵ}_{\mathrm{HOMO}}}{2}$$

where $I$ is the ionisation energy and $A$ is the electron affinity. Hard species have large HOMO-LUMO gaps (large $\eta$), and soft species have small HOMO-LUMO gaps (small $\eta$). This definition grounds the qualitative HSAB classification in molecular orbital theory: a hard species resists changes in its electron count because its frontier orbitals are energetically remote, while a soft species accommodates charge transfer readily because its frontier orbitals are close in energy.

The absolute hardness values for selected species illustrate the classification:

Species	$I$ (eV)	$A$ (eV)	$\eta$ (eV)	Classification
F${}^{-}$	17.42	3.40	7.01	Hard
OH${}^{-}$	13.17	1.83	5.67	Hard
Cl${}^{-}$	13.01	3.61	4.70	Hard
Br${}^{-}$	11.84	3.36	4.24	Borderline
I${}^{-}$	10.45	3.06	3.70	Soft
SH${}^{-}$	10.40	2.30	4.05	Soft

The electrochemical (Mulliken) definition $\eta =(I-A)/2$ is equivalent to half the HOMO-LUMO gap in Koopmans' approximation, connecting the HSAB concept directly to the electronic structure of the species.

Exercises Intermediate+

Frontier orbital interpretation of HSAB Master

The qualitative HSAB principle gains a rigorous theoretical foundation through frontier molecular orbital (FMO) theory, as developed by Fukui and extended by Pearson and Parr. The key insight is that the hard-soft classification maps directly onto the HOMO-LUMO energy gap of the reacting species.

Consider a Lewis acid with LUMO energy ${ϵ}_{\mathrm{LUMO}}^{\mathrm{A}}$ and a Lewis base with HOMO energy ${ϵ}_{\mathrm{HOMO}}^{\mathrm{B}}$. The stabilisation energy from donor-acceptor orbital interaction depends on both the energy gap $\Delta ϵ={ϵ}_{\mathrm{LUMO}}^{\mathrm{A}}-{ϵ}_{\mathrm{HOMO}}^{\mathrm{B}}$ and the overlap integral between the orbitals. Perturbation theory gives the interaction energy as:

$$\Delta E\approx -\frac{|⟨{\varphi }_{\mathrm{HOMO}}^{\mathrm{B}}|\hat{H}|{\varphi }_{\mathrm{LUMO}}^{\mathrm{A}}⟩{|}^2}{\Delta ϵ}$$

For hard-hard interactions, $\Delta ϵ$ is large (both species have wide HOMO-LUMO gaps). The orbital interaction is weak because the denominator is large, but the electrostatic interaction between the high-charge-density species is strong. Hard-hard bonding is dominated by classical Coulomb attraction, not by orbital overlap.

For soft-soft interactions, $\Delta ϵ$ is small (both species have narrow HOMO-LUMO gaps). The orbital interaction is strong because the denominator is small, and the large, diffuse orbitals of soft species overlap effectively. Soft-soft bonding is dominated by covalent charge transfer.

The HSAB principle emerges from this analysis: when both partners are hard, the electrostatic term dominates and is maximised by pairing hard with hard (maximising charge-density complementarity). When both partners are soft, the orbital term dominates and is maximised by pairing soft with soft (minimising $\Delta ϵ$). A mismatched hard-soft pairing gives neither a strong electrostatic interaction (the soft partner has low charge density) nor a strong orbital interaction (the large $\Delta ϵ$ suppresses charge transfer).

Density functional descriptors: electronegativity, hardness, and softness Master

The Parr framework in density functional theory (DFT) provides exact definitions for the qualitative concepts underlying HSAB. The total energy $E$ of a chemical species is a function of the electron number $N$ and the external potential $v(\mathbf{r})$ (the nuclear framework). The key response functions are:

Electronegativity (chemical potential, with sign reversed):

$$\chi =-\mu =-{\left(\frac{\partial E}{\partial N}\right)}_v$$

This is the same quantity introduced in the electronegativity formalism of unit 16.01.01, now applied to individual acid-base species. Electrons flow from the species with lower $\chi$ (the base, more electropositive) to the species with higher $\chi$ (the acid, more electronegative).

Absolute hardness:

$$\eta =\frac{1}{2}{\left(\frac{{\partial }^{2}E}{\partial N^2}\right)}_v$$

This is the curvature of the $E$ vs $N$ curve. A hard species has high curvature (energy changes steeply with electron count, resisting charge transfer). A soft species has low curvature (energy changes slowly, accommodating charge transfer).

Absolute softness is the reciprocal of hardness:

$$S=\frac{1}{2\eta }={\left(\frac{\partial N}{\partial \mu }\right)}_v$$

The finite-difference approximations (using integer electron changes) recover the Koopmans' theorem expressions:

$$\chi \approx \frac{I+A}{2},\eta \approx \frac{I-A}{2}$$

Pearson's maximum hardness principle (1993) states that molecular systems tend to arrange themselves so as to achieve maximum hardness. At equilibrium, a chemical system is at its hardest configuration. This principle explains why stable molecules and complexes tend to have large HOMO-LUMO gaps: a large gap means high kinetic stability and resistance to distortion or reaction.

Fukui functions provide a local (atom-specific) version of the global softness. The Fukui function $f(\mathbf{r})$ measures the sensitivity of the electron density at point $\mathbf{r}$ to a change in total electron number:

$$f(\mathbf{r})={\left(\frac{\partial \rho (\mathbf{r})}{\partial N}\right)}_v$$

For nucleophilic attack (the species gains electrons), the relevant Fukui function is $f^{+}(\mathbf{r})={\rho }_{N+1}(\mathbf{r})-{\rho }_N(\mathbf{r})$, approximated by the LUMO density. For electrophilic attack (the species loses electrons), $f^{-}(\mathbf{r})={\rho }_N(\mathbf{r})-{\rho }_{N-1}(\mathbf{r})$, approximated by the HOMO density. These local descriptors predict the regioselectivity of Lewis acid-base reactions — which atom in a polyatomic base will donate electrons, and which atom in a polyatomic acid will accept them.

The ECW model and quantitative acid-base parameters Master

The Drago-Wayland $E$ and $C$ parameter system, introduced at the intermediate level, was extended by Marks and Drago into the ECW model. The original equation:

$$-\Delta H=E_{\mathrm{A}}{E}_{\mathrm{B}}+C_{\mathrm{A}}{C}_{\mathrm{B}}$$

was modified to include a constant $W$ representing a baseline (solvent-corrected) enthalpy contribution:

$$-\Delta H=E_{\mathrm{A}}{E}_{\mathrm{B}}+C_{\mathrm{A}}{C}_{\mathrm{B}}+W$$

The $W$ term accounts for the fact that even in the absence of specific acid-base interaction, there is a general solvation/dispersion contribution to the measured enthalpy. In the gas phase, $W\approx 0$; in solution, $W$ depends on the solvent and must be determined empirically.

The $E$ parameter correlates with the electrostatic (ionic) component of the interaction. Acids with high $E_{\mathrm{A}}$ (e.g., phenol, $E_{\mathrm{A}}=4.33$; I${}_2$, $E_{\mathrm{A}}=1.00$) tend to interact through charge-dipole or dipole-dipole forces. The $C$ parameter correlates with the covalent (orbital-overlap) component. Acids with high $C_{\mathrm{A}}$ (e.g., I${}_2$, $C_{\mathrm{A}}=1.00$; SO${}_2$, $C_{\mathrm{A}}=0.80$) form bonds with significant electron-sharing character.

The ECW model succeeds quantitatively where HSAB is qualitative. Consider the adduct between I${}_2$ and NH${}_3$: HSAB classifies I${}_2$ as a soft acid and NH${}_3$ as a hard base, predicting a weak interaction. The Drago-Wayland calculation gives $-\Delta H=(1.00)(1.36)+(1.00)(3.46)=4.82kcal/mol$ — a modest but non-negligible enthalpy. The $C_{\mathrm{A}}{C}_{\mathrm{B}}$ term (3.46) dominates, revealing that the covalent character of NH${}_3$'s interaction with I${}_2$ is stronger than HSAB would suggest. NH${}_3$ is classified as "hard" by HSAB, but its $C_{\mathrm{B}}=3.46$ indicates substantial covalent capability.

This is the fundamental limitation of HSAB: the binary hard/soft classification collapses a two-dimensional property space ($E$ and $C$) into a single axis. The ECW model preserves both dimensions and achieves better enthalpy predictions as a result. Pearson himself acknowledged this, noting that HSAB is a qualitative guide while the $E$-$C$ framework is a quantitative tool. The two are complementary, not contradictory.

The parameter sets have been refined over decades. The original 1965 Drago-Wayland table contained about 30 acids and bases. The modern ECW database includes over 100 species with parameters determined by least-squares fitting to hundreds of measured enthalpies. The self-consistency of the parameter set (any acid-base pair's enthalpy is predicted to within ~1 kcal/mol of the experimental value when both species have well-determined parameters) demonstrates that the two-term decomposition captures the essential physics of Lewis acid-base interactions.

Lux-Flood acid-base theory and oxide chemistry Master

The Lewis framework extends naturally to high-temperature oxide and silicate chemistry through the Lux-Flood theory (Lux 1939, Flood 1947). In this formulation, an acid is an oxide ion acceptor and a base is an oxide ion donor:

$$\mathrm{Base}\rightleftharpoons \mathrm{Acid}+{\mathrm{O}}^{2-}$$

This is the acid-base theory of non-aqueous, oxide-dominated systems. In molten oxides and silicates (the domain of geochemistry and materials science), the relevant acid-base reaction is the transfer of O${}^{2-}$ rather than H${}^{+}$ or a generic electron pair.

Examples:

• CaO is a Lux-Flood base: $\mathrm{CaO}\to \mathrm{C}{\mathrm{a}}^{2+}+{\mathrm{O}}^{2-}$
• SiO${}_2$ is a Lux-Flood acid: $\mathrm{Si}{\mathrm{O}}_2+{\mathrm{O}}^{2-}\to \mathrm{Si}{\mathrm{O}}_3^{2-}$
• The reaction $\mathrm{CaO}+\mathrm{Si}{\mathrm{O}}_2\to \mathrm{CaSi}{\mathrm{O}}_3$ is a Lux-Flood acid-base neutralisation.

The Lux-Flood framework explains why some oxides are acidic (P${}_2$O${}_5$, SO${}_3$, B${}_2$O${}_3$ — they accept O${}^{2-}$ to form oxyanions), some are basic (Na${}_2$O, CaO, MgO — they donate O${}^{2-}$), and some are amphoteric (Al${}_2$O${}_3$, ZnO — they can act as either acid or base depending on the other oxide present). The classification maps onto the periodic trends in electronegativity and charge density from unit 16.01.01: oxides of electropositive metals are basic, oxides of electronegative nonmetals are acidic, and the boundary falls near the metalloid region.

Amphoteric behaviour provides a bridge between Lux-Flood and Lewis theory. Al${}_2$O${}_3$ acts as a Lux-Flood acid in the presence of a strong base (CaO): $\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3+\mathrm{CaO}\to \mathrm{Ca}(\mathrm{Al}{\mathrm{O}}_2{)}_2$ (the Al${}_2$O${}_3$ accepts O${}^{2-}$). It acts as a Lux-Flood base in the presence of a strong acid (P${}_2$O${}_5$): the O${}^{2-}$ from Al${}_2$O${}_3$ is donated to P${}_2$O${}_5$. In the Lewis picture, Al${}^{3+}$ is a hard acid that can accept electron pairs from O${}^{2-}$ donors, but Al${}_2$O${}_3$ also has lone pairs on oxygen that can be donated to stronger Lewis acids. The amphotericity is a direct consequence of aluminium's intermediate position on the hard-soft scale.

Acid-base concepts in non-aqueous solvents and superacids Master

The Bronsted-Lowry definition restricts acids and bases to proton donors and acceptors in aqueous solution. The Lewis definition removes the solvent restriction entirely, but understanding the role of the solvent is essential for practical chemistry. In any solvent, the levelling effect limits the strength of acids and bases to the solvent's own acid-base properties. In water, no acid stronger than H${}_3$O${}^{+}$ can exist in solution (it is fully dissociated and "levelled" to H${}_3$O${}^{+}$), and no base stronger than OH${}^{-}$ can exist (it is levelled to OH${}^{-}$). To access stronger acids or bases, a different solvent is needed.

Superacids are defined as acids stronger than 100% sulfuric acid ($H_0\le -12$, where $H_0$ is the Hammett acidity function). The strongest known superacid is fluoroantimonic acid (HSbF${}_6$, $H_0\approx -28$), formed by combining HF with SbF${}_5$. In the Lewis picture, SbF${}_5$ is an extremely strong Lewis acid that strips F${}^{-}$ from HF, generating the naked H${}^{+}$ equivalent (actually the H${}_2$F${}^{+}$ cation) with negligible nucleophilic interference from the SbF${}_6^{-}$ anion. Superacids can protonate extraordinarily weak bases: hydrocarbons, noble gases (Xe), and even molecular hydrogen.

The HSAB framework explains why SbF${}_5$ is such an effective Lewis acid. Sb${}^{5+}$ is a hard acid with very high charge density, and F${}^{-}$ is a hard base. The hard-hard SbF${}_5$ complex is extremely stable, and the SbF${}_6^{-}$ anion has very low basicity (the fluoride ions are tightly bound). This leaves the proton in the superacid solution essentially free — no available base can compete with SbF${}_5$ for the fluoride, so H${}^{+}$ has no counter-base to neutralise it.

Non-aqueous solvents define their own acid-base chemistry. In liquid ammonia (bp $-33{}^{\circ }\mathrm{C}$), the autoionisation is $2\mathrm{N}{\mathrm{H}}_3\rightleftharpoons \mathrm{N}{\mathrm{H}}_4^{+}+\mathrm{N}{\mathrm{H}}_2^{-}$. NH${}_4^{+}$ is the strongest acid in liquid ammonia (analogous to H${}_3$O${}^{+}$ in water), and NH${}_2^{-}$ is the strongest base (analogous to OH${}^{-}$). Amide bases like NaNH${}_2$ are far stronger bases in liquid ammonia than any base in water, and they can deprotonate extremely weak acids (such as terminal alkynes, pK${}_a\approx 25$ in water, but fully deprotonated in liquid ammonia).

The solvent also affects HSAB behaviour. Hard-hard interactions are enhanced in polar protic solvents (water, alcohols) because the solvent stabilises the ionic character of the interaction. Soft-soft interactions are enhanced in nonpolar or weakly polar solvents (benzene, CH${}_2$Cl${}_2$) where the covalent character is not competed away by solvent coordination. The same acid-base pair can show different thermodynamic preferences in different solvents — a complication that the qualitative HSAB principle handles poorly but the quantitative ECW model (with its solvent-dependent $W$ parameter) addresses directly.

Connections Master

• Main-group chemistry 16.01.02 pending provides the periodic trends in electronegativity, charge density, and polarisability that determine whether a species is hard or soft. The diagonal relationships (Li–Mg, Be–Al) and inert-pair effect (Pb preferring +2) directly influence the Lewis acidity and HSAB classification of main-group species.

• Periodic trends quantified 16.01.01 establishes the ionisation energy, electron affinity, and electronegativity framework that underpins the absolute hardness $\eta =(I-A)/2$ and the Parr DFT definitions of electronegativity and hardness.

• Coordination chemistry 16.04.01 is essentially the chemistry of Lewis acid-base interactions between metal ions (Lewis acids) and ligands (Lewis bases). HSAB theory predicts which ligands will bind preferentially to which metal centres, and the Drago-Wayland parameters quantify those preferences.

• Crystal field theory 16.03.01 connects to HSAB through the spectrochemical series: strong-field ligands (CN${}^{-}$, CO) are typically soft bases, while weak-field ligands (F${}^{-}$, H${}_2$O) are hard bases. The hard-soft distinction maps onto the $\sigma$-donor/$\pi$-acceptor character of ligands.

• Organometallic chemistry 16.05.01 depends on soft-soft interactions. Metal carbonyls, metal-phosphine complexes, and metal-alkene/pi complexes all involve soft metal centres binding to soft ligands through orbital-overlap-dominated interactions — the hallmark of soft-soft HSAB behaviour.

Historical notes Master

Gilbert N. Lewis proposed his acid-base definition in 1923, in his book Valence and the Structure of Atoms and Molecules. The same year, Bronsted and Lowry independently published their proton-transfer definition. Lewis's broader definition attracted less immediate attention because proton-transfer chemistry dominated the experimental landscape of the 1920s–1940s. The Lewis framework gained prominence in the 1950s and 1960s as coordination chemistry and organometallic chemistry developed, revealing the centrality of electron-pair donation/acceptance in metal-ligand bonding.

Ralph Pearson introduced the HSAB principle in a series of papers beginning in 1963 ("Hard and Soft Acids and Bases," J. Am. Chem. Soc. 85, 3533). Pearson's insight was empirical: he observed that the qualitative hard/soft classification, though crude, predicted trends in complex stability, solubility, and reaction selectivity across a wide range of inorganic chemistry. The 1968 Journal of Chemical Education paper ("Hard and Soft Acids and Bases," J. Chem. Educ. 45, 581) disseminated the concept to the teaching community and became one of the most-cited papers in chemical education.

The quantitative underpinning of HSAB arrived through two independent channels. Pearson and Parr's 1983 paper ("Absolute Hardness: Companion Parameter to Absolute Electronegativity," J. Am. Chem. Soc. 105, 7512) connected hardness to the HOMO-LUMO gap and the DFT framework, transforming HSAB from a qualitative heuristic into a principle with roots in electronic structure theory. Separately, Drago and Wayland's 1965 double-scale equation (J. Am. Chem. Soc. 87, 3571) provided an empirical enthalpy correlation that was quantitatively predictive without requiring a hardness classification. The tension between these two approaches — HSAB as a qualitative classification vs $E$-$C$ as a quantitative parameter set — defined much of the acid-base theoretical literature through the 1970s–1990s.

The Lux-Flood theory was proposed by Hermann Lux in 1939 and extended by Haldor Flood in 1947 to describe acid-base chemistry in molten oxides and silicates. Though less widely taught than the Bronsted-Lowry and Lewis frameworks, Lux-Flood theory is essential in geochemistry, glass science, and high-temperature materials processing.

Superacid chemistry was pioneered by George Olah, who demonstrated that superacids could generate and stabilise carbocations in solution — work that earned him the 1994 Nobel Prize. Olah's superacid research revealed that hydrocarbons, long considered inert to protonation, could be protonated and functionalised under sufficiently acidic conditions, opening new synthetic pathways.

Bibliography Master

• Lewis, G. N. Valence and the Structure of Atoms and Molecules. New York: Chemical Catalog Co., 1923. Ch. 11.

• Pearson, R. G. "Hard and Soft Acids and Bases." J. Am. Chem. Soc. 85, 3533 (1963).

• Pearson, R. G. "Hard and Soft Acids and Bases, HSAB, Part I: Fundamental Principles." J. Chem. Educ. 45, 581 (1968).

• Pearson, R. G. "Hard and Soft Acids and Bases, HSAB, Part II: Underlying Theories." J. Chem. Educ. 45, 643 (1968).

• Drago, R. S. & Wayland, B. B. "A Double-Scale Equation for Correlating Enthalpies of Lewis Acid-Base Interactions." J. Am. Chem. Soc. 87, 3571 (1965).

• Marks, A. P. & Drago, R. S. "Extension of the $E$-$C$ Equation to Include Solvent Effects." Inorg. Chem. 15, 1800 (1976).

• Parr, R. G. & Pearson, R. G. "Absolute Hardness: Companion Parameter to Absolute Electronegativity." J. Am. Chem. Soc. 105, 7512 (1983).

• Parr, R. G., Donnelly, R. A., Levy, M. & Palke, W. E. "Electronegativity: The Density Functional Viewpoint." J. Chem. Phys. 68, 3801 (1978).

• Pearson, R. G. "The Principle of Maximum Hardness." Acc. Chem. Res. 26, 250 (1993).

• Fukui, K. "Role of Frontier Orbitals in Chemical Reactions." Science 218, 747 (1982).

• Housecroft, C. E. & Sharpe, A. G. Inorganic Chemistry, 5th ed. Harlow: Pearson, 2018. Ch. 5.

• Miessler, G. L., Fischer, P. J. & Tarr, D. A. Inorganic Chemistry, 5th ed. Upper Saddle River: Pearson, 2014. Ch. 5.

• Lux, H. "Saeuren und Basen im Schmelzfluss: Die Bestimmung der Sauerstoffionen-Konzentration." Z. Elektrochem. 45, 303 (1939).

• Flood, H. & Forland, T. "The Acidic and Basic Properties of Oxides." Acta Chem. Scand. 1, 592 (1947).

• Olah, G. A. "My Search for Carbocations and Their Role in Chemistry (Nobel Lecture)." Angew. Chem. Int. Ed. 34, 1393 (1995).