Volume 17: Pages 342-389, 2004
The Nature of the Chemical Bond Revisited and an Alternative Maxwellian Approach
Randell L. Mills
BlackLight Power, Inc., 493 Old Trenton Road, Cranbury, New Jersey 08512 U.S.A.
It is taught that the chemical bond exists because of a phenomenon that is unique to quantum mechanics (QM). Specifically, the nature of the chemical bond is based on a nonphysical “exchange integral“ that is a consequence of a postulated linear combination of product wave‐functions, where it is implicit that each point electron with infinite self‐electric and magnetic field energies must exist as a “probability‐wave cloud“ and be in two places at the same time (i.e., centered on two nuclei simultaneously!). A further nonphysical aspect is that the molecular solution is obtained without considering the nuclei to move under the Born‐Oppenheimer approximation, yet the molecule must have a further nonphysical perpetual‐motion‐type property of zero‐point vibration (ZPV). Additional internal inconsistencies arise. The electron clouds mutually shield the nuclear charge to provide an adjustable parameter, the “effective nuclear charge,” yet neither has any self‐shielding effect even though the clouds are mutually indistinguishable and must classically result in a self‐interaction force equivalent to half the central attractive force. Furthermore, the hydrogen molecule is electron‐spin paired. The magnitude of the corresponding force between the point electrons is equivalent to the electric force as the separation goes to zero. This term as well as the self‐interaction term is conspicuously absent from the Hamiltonian. Instead, arbitrary types of variational parameters of the wave‐functions and mixing of wave‐functions as well as other adjustable parameters, such as the effective nuclear charge, ionic character, and correlation interactions, are introduced to force the solutions of a multitude of methods, such as valence bond, valence bond plus ionic terms, molecular orbital (MO) theory, MO with configuration interaction, self‐consistent field method, SCF‐LCAO‐MO, Hartree‐Fock, Slater orbitals, ionic terms, valence‐shell electron‐pair repulsion (VSEPR), etc., to more closely approximate the experimental parameters. Yet the experimental bond energy is not calculated. Rather, a parameter De is determined from which the ZPV is subtracted and an anharmonicity term in the ZPV is added to obtain the experimentally measurable bond energy Do. ZPV has never been directly measured; it violates the second law of thermodynamics and is in conflict with direct experimental results such as the formation of solid hydrogen and Bose‐Einstein condensates of molecules. As a consequence, the bond‐energy predictions of QM have never been tested experimentally, and it is not possible to state that the methods predict the experimental bond energy at all. The many conflicting attempts suffer from the same shortcomings that plague atomic quantum theory, infinities, instability with respect to radiation according to Maxwell's equations, violation of conservation of linear and angular momentum, lack of physical relativistic invariance, etc. From a physical perspective, the implication for the basis of the chemical bond according to QM being the exchange integral and the requirement of ZPV, “strictly quantum‐mechanical phenomena,” is that the theory cannot be a correct description of reality. A proposed solution based on physical laws and fully compliant with Maxwell's equations solves the parameters of molecular ions and molecules of hydrogen isotopes in closed‐form equations with fundamental constants only. The agreement is remarkable. A physical basis for density functional theory may exist.
Keywords: nature of the chemical bond, hydrogen molecular ion, hydrogen, valence bond, exchange integral, zero‐point vibration, ellipsoidal Laplacian, Maxwellian solution
Received: January 9, 2004; Published online: December 15, 2008