Abstract: A new model for atomic structure is proposed in which there is an integral relation between the nuclear geometry and covalent molecular geometry. The model explains the known geometry of a number of covalent molecules. A new type of double bond with tetrahedral carbon atoms and a new triple bond with the double bond geometry are allowed on the new model. The implications of the model are discussed.

Introduction: Covalent geometry and proton sites in the nucleus.

The currently accepted bicameral model for atomic structure postulates that the atom is composed of two relatively independent parts, the electrons and the nucleus. The nucleus plays no role, except as a point charge, in the properties of the electronic part {ref.1}. It is not generally recognized that this model faces severe trials in chemistry {ref.3}. As for the structure of the nucleus {ref.4b}, we seem to have reached a dead end. Apart from the idea that it may consist of protons and neutrons we have no clue about the nature of the nucleus. A simple question like the location or the sites of the protons or neutrons in the nucleus cannot be answered with any degree of certainty {ref.4}.

The electronic part is dealt with in the aufbau model. Electrons are “added” to nuclei with progressively increasing positive charge and take their place in energy levels governed by Pauli’s exclusion principle. The model requires ad hoc assumptions at every stage to explain experimental observations in chemistry. Thus to reconcile the model with the equality of the four valences of carbon, the electron energy levels have to undergo hybridization or form molecular orbitals.

The problem lies in equating inequalities. The two are not inter-convertible. They are mutually exclusive.

Ad hoc attempts to fit theory to observation are total anathema to our concept of theory. The process of hybridization changes, again in ad hoc fashion, as the bonding pattern progresses from single to double and to triple bonds. The most telling example of the ad hoc assumptions is the attempt to introduce a new bonding pattern called “σ donation and π back-donation” bond to explain the structure of the platinum complex Zeisse’s salt. In the case of propellane {ref.9} and the covalent chemistry of boron, including the structure of boron hydrides and crystalline boron, the model does not have any scope even for ad hocism. It just fails.

There is more. Quantum mechanical treatment is not compatible with the well-known chirality of the asymmetric carbon atom {ref.2a}. This view had been largely ignored {ref.2b} in the twenty five years after it was published. Its disruptive content had been papered over and not confronted. The reason for this significant failure arises from the fact that the QM depends on probability ...top

considerations, whilst chemistry deals with the individuality of the elements. Probability considerations cannot be applied to cases where individuality is the determinant. This important limitation had not been given due consideration.

The assumption that the 4s orbital is of lower energy than the d orbital is also arbitrary. This assumption is necessary to account for the break of the period between Argon and Potassium.

What happens if the two parts of the atom are mutually interdependent? Since protons carry the charge of the nucleus, covalent structural chemistry should reflect the location of protons in the nuclear structure. This is confirmed by our examination of the covalent chemistry of carbon.

In what follows, we have drawn a correlation between nuclear geometry and molecular geometry without making any assumption on the nature of the chemical bond. It is clear that the tetrahedral geometry of the carbon atom survives in all covalent bonding varieties, namely, in the single bonds (tetrahedral, square and inverted), and in van’t Hoff and tetrahedral double bonds, and in van’t Hoff and benzyne triple bonds. The correlation leads to the implicit deduction that the bonding sites in the atom are invariant and may correspond to proton sites in the nucleus. The model has been found valid for silicon (dealt with in the next paper) and explains the structural chemistry of boron without any ad hoc assumptions or changes in the basic postulates. (The article on boron is considered Part III of the series. It is submitted alongside for the reviewer’s opinion. It is not meant for publication.)

Assumptions:

1. Every nucleus has a definite geometrical structure {ref.4} composed of nucleons. No distinction is made between protons and neutrons inside the nucleus. The component particles are considered to be of uniform size and of identical properties.


2. Nuclear structure determines and dictates atomic structure. In the following study nuclear models are treated as atomic models.

(There is no experimental evidence for this assumption. Equally there is no evidence for the assumption hat he nucleus has no influence on the electronic part.)

The model deals only with the molecular geometry and has nothing to say on the nature of the chemical bond. Bonding and other interactions take place at corners or projections on the atomic surfaces. The atomic surface areas are divided into:

1. Primary bonding sites called b-surfaces and colored red in the models


2. Secondary bonding sites called the quiescent or the q-surfaces, colored yellow.