Bond Dipoles and the Molecular Dipole

John Keller, Department of Chemistry & Biochemistry

     Valence electrons are distributed unevenly around a molecule whenever different atom types, i.e. elements, are bonded together. Differences in atom electronegativity cause slight charge separations between two bonded atoms, which results in formation of a local electric field.

      As described in most elementary chemistry textbooks, this field, or dipole, is represented by a vector whose length is proportional to the field strength and whose direction is from electron-poor atom to electron-rich atom, that is, partial positive to partial negative. If the molecule is symmetric, the vectors may point in opposite directions, canceling each other out and giving a zero molecular dipole. On the other hand, if there is some asymmetry in the molecule, the vectors will not cancel, giving a non-zero molecular dipole. The former molecules are "non-polar"; the latter are "polar."

    Two weaknesses in textbook approachs are that (1) partial atomic charge values actually vary depending on the method of calculation. Furthermore, there is no way to check the accuracy of such charges because "partial atomic charge" cannot be measured experimentally. (2) Dipoles associated with lone pairs are rarely taken into account.

    QM programs such as Gaussian or ORCA accurately predict dipole moments by, in effect, calculating the average position of all the molecule's electrons, and measuring the distance to the average position of all the molecule's nuclei. Lastly, the dipole equation μ = constant x er, where e is the total nuclear charge and r is the distance vector from + to - .   

    Breaking down the molecular dipole into local components can be useful, however a better approach is offered by Natural Bond Orbital (NBO) theory. NBO takes a similar approach described above, that is, it assigns a dipole to each filled, 2-electron natural bond orbital, which could be a lone pair, a sigma bond, or a π-bond.The vector sum of NBO dipoles equals the molecular dipole.

     Several examples are shown on the following web pages. These structures were optimized with Gaussian 09 with the B3LYP density functional theory method and either 6-311++G(2d,p) or aug-cc-pVTZ basis sets. Dipoles and orbital shapes were calculated within Gaussian using NBO code (nbo6.chem.wisc.edu). The dipole vectors on this pages were plotted with Jmol using the negative xyz components of the Natural Localized Molecular Orbital (NLMO) dipole, which are nearly equal to NBO dipole unless significant delocalization is involved. The NBO dipoles are obtained by manually adding the word "dipole" in the $NBO section at the end of the Gaussian input file.

Water  H2O

Water with ESP surface

Hydrogen peroxide  H2O2

Methanol CH3OH

Ammonia  NH3

Methylamine CH3NH2

Fluoroamine  NH2F

Unsymmetric molecule

Trifluoroamine  NF3

Phosphorus Trifluoride PF3

Fluoromethane CH3F

Methane CH4

Propene C3H6

Cyclopropane C3H6

Cyclobutane C4H8

Formamide CHONH2

Thioformate CHOS-

Hydroxycarbene HCOH

This site was updated 2-16-2023. Contact:   jwkeller/at/alaska.edu