1827 Orgunic and Biologicul Chemistry Application of the Principle of Hard and Soft Acids and Bases to Organic Chemistry Ralph G.Pearson and Jon Songstad' Contribution from the Department of Chemistry, Northwestern University, Euanston, Illinois 60201. Received November 2, 1966 Abstract: The principle of hard and soft acid and bases (HSAB principle) is applied to organic chemistry. ,Organic molecules are viewed as Lewis acid-base complexes and their relative thermodynamic stability explained in terms of two factors. One is the tendency of intrinsically strong acids to coordinate to the strongest bases. The second is the special stabilization of combinations of hard acids and bases, or soft acids and bases. The symbiotic principle is illustrated, which states that there is an extra stabilization if several soft bases (ligands) or several hard bases cluster about a single acidic atom. The same two principles are applied to rates of nucleophilic and elec- trophilic substitutionreactions in organic chemistry. ecently2a generalization was proposed which makes CHSHg+(aq) + BH+(aq) CHaHgB+(aq) + H+(aq) (1) R it possible to correlate a great many phenomena in various areas of chemistry. Use is made of the con- If the equilibrium constant for this reaction is much cept of generalized, or Lewis, acids and bases. The greater than unity, the base B is soft. If it is near unity, generalization may be called the principle of hard or less than unity, the base is hard. The proton is the and soft acids and bases (HSAB). It states that hard simplest hard acid and the methylmercury cation is one acids prefer to coordinate to hard bases and soft acids of the simplest soft acids. Table I contains a listing of prefer to coordinate to soft bases. hard and soft bases for later reference. These terms are qualitatively defined in the following If the equilibrium constants of eq 1 are used to rank ways: soft base-donor atom is of high polarizability, a series of bases, the following order of decreasing low electronegativity, easily oxidized, and associated softness is obtained. with empty, low-lying orbitals ; hard base-donor atom I- > Br- > CI- > S2- > RS- > CN- > HzO > NH3 F- > OH- is of low polarizability, high electronegativity, hard to - oxidize, and associated with empty orbitals of high It turns out that this is not a universal order since a energy and hence inaccessible; soft acid-the acceptor change in one of the reference acids will give a dif- atom is of low positive charge, large size, and has ferent series. The reason for this may be seen by con- several easily excited outer electrons; hard acid-ac- sidering the generalized acid-base exchange reaction. ceptor atom is of high positive charge, small size, and A:B' + A':B A:B + A':B' (2) does not have easily excited outer electrons Operationally, acids may be defined by following the We expect such a reaction to proceed such that the procedures of Sch~arzenbach~and Ahrland, Chatt, and strongest acid, A, is found coordinated to the strongest Davies.4 These workers divided metal ions (which are base, B. The terms hard and soft do not mean the Lewis acids) into two classes called A and B by Schwar- same as strong and weak. Thus an acid is charac- zenbach and a and b by Ahrland, Chatt, and Davies. terized by at least two properties, its strength and its Hard acids follow the same pattern as class a metal ions, hardness, or softness; the same is true for a base. and soft acids show the pattern of class b metal ions. It is well known that there is no universal order of acid For complexes with different donor atoms, the following or base strength; still we recognize that some Lewis sequences of stabilities are found. acids, such as Hf, are much stronger than other acids, such as 12, or that H- is a much stronger base than HzO. N >>P >As > Sb The HSAB principle then states that there is an extra hard {0 >>S > Se > Te F > CI > Br > I stabilization in A:B if both the acid and base are hard, or if both are soft.6 N<< P > As > Sb soft {0 << S - Se -Te We can usually recognize hardness or softness in a F < CI < Br < I qualitative way by examining an acid or base, par- Soft bases might be operationally defined by con- ticularly the donor or acceptor atoms. The situation sidering the equilibrium6 may be something like that for the terms solvent polar- (1) Chemistry Department, Bergen University, Norway. Supported (5) G. Schwarzenbach and M. Schellenberg, Helu. Chim. Acta, 48, by the Royal Norwegian Council for Scientific and Industrial Research. 28 (1965); G. Schwarzenbach, Chem. Eng. News, 43,92 (May 31, 1965). (2) R. G. Pearson, J. Am. Chem. SOC.,85, 3533 (1963); Science, 151, (6) Thus the equilibrium constant for the reaction A + :B $ A:B 172 (1966). might be characterized by an equation such as log K = SASB + UAUB. (3) G. 'Schwarzenbach, Experientia Suppl., 5, 162 (1956); Aduan. The factors SA and SB are strength factors for the acid and base; UA and Inorg. Chem. Radiochem., 3, 251 (1961). UB are softness factors. For a hard acid or base, u would be negative; (4) S. Ahrland, J. Chatt, and N. R. Davies, Quart. Rev. (London), 12, for a soft acid or base, u would be positive; see R. S. Drago and B. B. 265 (1958). Wayland, J. Am. Chem. Soc., 87, 3571 (1965). Pearson, Songstad ] Hard and Soft Acids and Bases ity or electronegativity. These useful concepts lack a precise definition, or rather several definitions exist to suit various kinds of data. B K.2 K, In spite of this inability to make the rules quantitative H- 1 x 1021 10-29 c at present, we hope in this paper to show that the CN- 3 x 1014 7 x 10-10 principle of hard and soft acids and bases is extremely CHzCOCH3- 2 x 10” 10-20 useful. We will take the area of organic chemistry CHaS- 2 x 109 5 x 10-11 for which a wealth of data exists for which little cor- CHI- 3 x 108 10-40 c I- 2 x 108 109.6 relation has been done in terms of hardness and softness SH- 3 x 107 1 x 10-7 concepts.’ It will turn out that much, but not all, of Br- 1 x 107 109 what we have to say has been noted before and ex- N2H3- 6 X 106 1 x 10-8 plained in various ways. We wish to show an under- c1- 3 x 106 107 lying pattern in all of these phenomena. NO1 a -104 5 x 10-4 NHz- 8 X lo3 10-33 d It should be stressed that the HSAB principle is not CeHjO- 3 x 103 1 x 10-10 a theory out is a statement about experimental facts. CH,O- 3 x 103 10-15 Accordingly an explanation of some observation in F- N10-3 1.4 x 10-3 ~~ ~~ terms of hard and soft behavior does not invalidate some a Bonding to 0 in each case. CHIN02 is about 1.5 kcal more other, theoretical explanation. In fact, the various stable than CH30N0. * Values of K,, are from ref 9, except for theories which have been put forward2 to explain the B=NH2-, N2H3-, NO2-, CHI-, and F-. See ref 13, 16, and 31 for principles of hard and soft acids and bases in general data; also Technical Notes 270-1 and 270-2, National Bureau of Standards, 1965-1966. c Estimated. See F. Basolo and R. G. usually include the previous explanations for the par- Pearson, “Mechanisms of Inorganic Reactions,” John Wiley and ticular cases to be discussed in this paper. Sons, Inc., New York, N. Y., 1958, p 344. d Estimated from KB of aniline, phenol, and water. See F. G. Bordwell, “Organic Thermodynamic Examples Chemistry,” The Macmillan Co., New York, N. Y., 1963, p 867. We will first show how the thermodynamic stabilities of many kinds of organic molecules can be rationalized different’ from those in the gas phase by more than a by thc hard-soft concept. This leads lo a better under- factor of 25 or so for reaction 4. Also the value of standing of many well-known facts and to a prediction AHo is not different from the value of AGO, in the gas, of some results that are probably not well known to by more than a kilocalorie or two. Since we will be most organic chemists. discussing large differences, it will be possible to get The meihod that is used is to mentally break down an data reasonably comparable to that in Table I1 by just organic species into a Lewis acid fragment such as a knowing gas phase or aqueous heats of formation. carbonium ion or acylium ion, and a base fragment It can be seen that the equilibrium constant for reac- such as a carbanion, a hydride ion, or a halide ion. tion 4 is very large for bases such as CH3-, I-, CH3S-, The stability of the molecule is then considered in terms and H- which are listed as soft bases in Table I. For of the acid-base interaction hard bases such as CH3O- and C6H50-, the equilibrium A+ :B+A:B (3) constant is much smaller, and, for F-, the constant is When the acceptor atom of A is carbon, we are talking less than unity. The immediate conclusion is that the about what Parker has called carbon basicity.* An methyl carbonium ion is a softer acid than is the proton important paper by Hine and Weimarg has recently which is a hard acid.