Journal of Computer-Aided Molecular Design, 4 (1990) 1-105 1

ESCOM

MOPAC: A semiempirical molecular orbital program

James J.P. Stewart

Frank J. Seiler Research Laboratory, U.S. Air Force Academy, Colorado Springs, CO 80840, U.S.A.

1. INTRODUCTION

1.1 What is M O P A C ?

Before we start, we need a working definition for MOPAC. The following description has been
used many times to describe MOPAC: MOPAC is a general-purpose, semiempirical molecular
orbital program for the study of chemical reactions involving molecules, ions, and linear po-
lymers. It implements the semiempirical Hamiltonians MNDO, AM 1, MINDO/3, and MNDO-
PM3, and combir_es the calculations of vibrational spectra, thermodynamic quantities, isotopic
substitution effects, and force constants in a fully integrated program. Elements parameterized
at the M N D O level include H, Li, Be, B, C, N, O, F, A1, Si, P, S, C1, Ge, Br, Sn, Hg, Pb, and
I; at the PM3 level the elements H, C, N, O, F, A1, Si, P, S, C1, Br, and I are available. Within
the electronic part of the calculation, molecular and localized orbitals, excited states up to sextets,
chemical bond indices, charges, etc. are computed. Both intrinsic and dynamic reaction coordi-
nates can be calculated. A transition-state location routine and two transition-state optimizing
routines are available for studying chemical reactions.

Over the past two decades computational chemistry has undergone a considerable change.
During the 70s, there was a chaos of untested methods of decidedly questionable accuracy and
very limited applicability. This was a time in which myriad new methods, particularly in the field
of semiempirical theory, would suddenly appear. Enthusiastically advocated by their authors,
they would, as often as not, die with equal rapidity. For many experimental chemists, the lesson
learned during this era was that computational chemists - a title which few would admit to for
many years - were not to be heeded, but to be tacitly ignored.
Since then computational chemistry has become respectable. Today, three main branches enjoy
considerable popularity: molecular mechanics, semiempirical, and ab initio or Gaussian methods
[1]. Each branch has carved out a niche within which it is supreme, and from which it can recog-
nize the eminence of the other two. Thus, the conformations of macromolecules are most effective-
ly studied using the techniques of molecular mechanics, while the electronic properties of small
molecules are most accurately calculated using ab initio methods. In the middle lie the semiempir-
ical methods, of which M N D O [2], MINDO/3 [3] and AM1 [4], developed by Michael J.S.
Dewar, are the most popular. These methods are not particularly fast; in fact, they are very slow
when compared with molecular mechanics. Nor are they particularly rigorous: this is rightfully

0920-654X/$3.50 © 1990 ESCOM SciencePublishers B.V.

the province of the ab initio methods. As with Homo sapiens, a creature without specialization,
the semiempirical methods are not particularly good at any one thing. A consequence of this is
that such methods, again as with Homo sapiens, have become extremely versatile.
It is not sufficient to develop a useful computational chemistry technique; as a series of mathe-
matical formulae, new methods have no practical use. Such techniques must first have an algo-
rithmic body to inhabit in order to have any corporeal existence. Currently several such bodies
exist. In this section we will look at one of these implementations, MOPAC. This description
of MOPAC is not intended to serve as an instruction set for using the program (the Manual is
provided for that purpose), but rather as an overview of the program, along with indications of
its application.
MOPAC is a single program which incorporates many separate functions, almost all of which
were developed by individual groups prior to the writing of MOPAC. As a result, while the curr-
ent formulation was written almost entirely by the author, MOPAC really represents the cumula-
tive effort of a very large number of individuals. The individual algorithms they contributed each
represent many man-years of effort. However, by concentrating on writing a single coherent pro-
gram to implement all these different methods at the expense of advancing computational tech-
niques, the author has been able to generate a single program which is, in the words of Michael
Dewar, 'impressively easy to use' and 'contains options for nearly all the applications where our
procedures have been found 'to be useful'.*
Although MOPAC was written for use on a VAX computer, the fact that it would be trans-
ferred to other machines was recognized at the onset. Thus MOPAC was written from its incep-
tion in almost pure FORTRAN-77. This policy proved prophetic. MOPAC has since been con-
verted to run on many machines, including those made by CDC, CRAY, ETA, G O U L D , IBM,
NEC, PRIME, and ZENITH, with little difficulty. These range in size from the supercomputers,
such as the CRAY-2, to the micros, such as the IBM PC-AT and -XT and Zenith 100.
MOPAC is wholly in the public domain. The source code and documentation can be obtained
from the Quantum Chemistry Program Exchange. Over the past five years it has been developed
at the Frank J. Seller Research Laboratory of the Air Force Academy, and, as a product of gov-
ernment work, cannot be copyrighted. This is not to say that some versions of MOPAC have
not been copyrighted - versions specific to individual machines have been - but all versions pro-
duced at the Frank J. Seiler Research Laboratory are public domain and may be obtained with-
out restriction from the Quantum Chemistry Program Exchange.
In order to remain current, a new version of MOPAC has been produced at the rate of one
per year. This rate was chosen as the best compromise between keeping users current and not
asking users to continuously update their files. As far as is possible, updates have been checked
for errors and for compatibility with previous versions. There have been six versions published
so far, and these versions have been converted to run on between 15 and 20 different types of
computer. This has resulted in a large number of 'official' programs. When all the modifications
individual users have made are taken into consideration, it is obvious that there is a very high
potential for confusion. Thus users are recommended to obtain copies of MOPAC from QCPE,
and to stay as current as possible.

* F o r e w o r d to M O P A C V e r s i o n 4.00 M a n u a l .
1.2 D R A W

Concurrent with the development of MOPAC, Major Donn M. Storch of the U.S. Air Force
Academy's Department of Chemistry, wrote a graphics package to act as a general purpose data
manipulation program. This program, called DRAW [5], is very useful for generating and chec-
king the input data for MOPAC. It's main use, however, is in the graphical interpretation of the
results of a MOPAC calculation. As the author of DRAW paraphrased da Vinci, 'A picture is
worth a thousand pieces of information'. Users of MOPAC are strongly encouraged to also ac-
quire DRAW, again from QCPE.

2. SEMIEMPIRICAL METHODS

There are four distinct methods available within MOPAC: MINDO/3, MNDO, AM1, and
PM3. All are semiempirical, and have roughly the same structure. A complete knowledge of these
methods is not necessary in order to use MOPAC; however, a superficial understanding of these
methods and their relationship to ab initio methods is important for using MOPAC and particu-
larly for interpreting the results.
The four methods within MOPAC have many features in common. They are all self-consistent
field (SCF) methods, they take into account electrostatic repulsion and exchange stabilization,
and, in them, all calculated integrals are evaluated by approximate means. Further, they all use
a restricted basis set of one s orbital and three p orbitals (Px, Py, and Pz) per atom and ignore
overlap integrals in the secular equation. Thus, instead of solving

[H - ES[ = 0

the expression

IH - E I = 0

in which H is the secular determinant, S is the overlap matrix, and E is the set of eigenvalues,
is solved. These approximations considerably simplify quantum mechanical calculations on sys-
tems of chemical interest. As a result, larger systems can be studied. Computational methods are
only models, and there is no advantage in rigorously solving Schr6dinger's equation for a large
system if that system has had to be abbreviated in order to make the calculations tractable. Semi-
empirical methods are thus seen to be well balanced: they are accurate enough to have useful
predictive powers, yet fast enough to allow large systems to be studied. Of course some systems,
e.g., enzymes, are so large that only a part of the system can be included in the model.
All four semiempirical methods contain sets of parameters. For MINDO/3 atomic and diatom-
ic parameters exist, while MNDO, AM1 and PM3 use only single-atom parameters. Not all pa-
rameters are optimized for all methods; for example, in MINDO/3, MNDO and AM1 the two-
electron one-center integrals are normally taken from atomic spectra. In the list given on the next
page, parameters optimized for a given method are indicated by ,. A + indicates that the value
of the parameter was obtained from experiment (not optimized). Where neither symbol is given,
the associated parameter is not used in that method.
4

Parameters used in semiempirical methods

Parameter Description MINDO/3 MNDO AM 1 PM3

U~ and Up s and p atomic orbital one-electron one-center inte-

grals + * * *
fl~ and tip s and p atomic orbital one-electron two-center reso-
nance integral terms * * *
s atomic orbital ionization potential for two-center
resonance integral term +
lp p atomic orbital ionization potential for two-center
resonance integral term +
diatomic two-center one-electron resonance integral
multiplier *
s-type Slater atomic orbital exponent * * * *
p-type Slater atomic orbital exponent * * * *
0~A atom A core-core repulsion term * * *
~AB atoms A and B core-core repulsion term *
Gss s-s atomic orbital one-center two-electron repulsion
integral + + + *
Gsp s-p atomic orbital one-center two-electron repulsion
integral + + + *
Gpp p-p atomic orbital one-center two-electron repul-
sion integral + + + *
Gp2 p-p' atomic orbital one-center two-electron repul-
sion integral + + + *
Hsp s-p atomic orbital one-center two-electron exchange
integral + + + *
KnAor anA Gaussian multiplier for nth Gaussian of atom A * *
Ln~ or bn~ Gaussian exponent multiplier for nth Gaussian of
atom A * *
M~ or c.~ radius of center of nth Gaussian of atom A. * *

All f o u r s e m i e m p i r i c a l m e t h o d s a l s o u s e t w o e x p e r i m e n t a l l y d e t e r m i n e d c o n s t a n t s p e r a t o m : t h e
a t o m i c m a s s of t h e m o s t a b u n d a n t i s o t o p e a n d t h e h e a t o f a t o m i z a t i o n .

3. T H E D A T A F I L E

A MOPAC c a l c u l a t i o n i n v o l v e s s u p p l y i n g a d a t a file d e f i n i n g t h e s y s t e m b e i n g s t u d i e d t o
MOPAC. The MOPAC p r o g r a m r e a d s t h e file a n d g e n e r a t e s a n o u t p u t file c o n t a i n i n g t h e r e s u l t s
o f t h e c a l c u l a t i o n . T h e i n p u t file h a s a h i g h l y s t a n d a r d i z e d f o r m a t . T h i s c o n s i s t s o f t h r e e l i n e s
of text followed by a geometry specification.
 Example of datafile

Line 1 S Y M M E T R Y B O N D S L O C A L I Z E VECTORS
Line 2 Ethane
Line 3
Line 4 C
Line 5 C 1.5 1
Line 6 H 1.1 1 109 1
Line 7 H 1.1 0 109 0 1200 2 1 3
Line 8 H 1.1 0 109 0 240 0 2 1 3
Line 9 H 1.1 0 109 0 1800 1 2 3
Line 10 H 1.1 0 109 0 1200 1 26
Line 11 H 1.1 0 109 0 240 0 1 2 7
Line 12
Line 13 3 1 45678
Line 14 3 2 45678

The first line of text specifies the type of calculation. This is achieved via the use of descriptive
keywords: thus, a unipositive cation would be specified by the keyword 'CHARGE = 1'. Lines
two and three are text describing the calculation for the user's benefit only.

4. GEOMETRY SPECIFICATION

Molecular geometries are specified in one of two ways. The more common specification in-
volves internal coordinates in which each atom is related to previously specified atoms by a bond
length, an angle and a dihedral angle: i.e., a connectivity list. Sometimes it is convenient to use
reference mathematical points in the definition of the geometry: these are allowed and are called
dummy atoms. Dummy atoms are ignored during any quantum mechanical calculation. The in-
ternal coordinate definition can also be assisted by the use of what are called symmetry functions.
These allow two or more bond lengths, angles or dihedrals to be related in a simple manner, typi-
cally by forcing them to be equivalent throughout a calculation.
Although geometries are no longer commonly specified by Cartesian coordinates the option
to allow such a specification is supported. The position of each atom is defined by three Cartesian
coordinates: these are the x, y and z values of the atom's position from an arbitrary origin. Carte-
sian coordinates are distinguished from internal coordinates by the absence of a connectivity list.
Regardless of the method of specifying the geometry, MOPAC normally proceeds via internal
coordinates. The most common geometry operation is geometry optimization. Originally the Da-
vidon-Fletcher-Powell [6] method was used. However, a few systems were found for which the
D F P optimization terminated at a significant distance from a stationary point. As a result, the
D F P optimization has been replaced by the BFGS or BroydenFletcher-Goldfarb-Shanno [7]
method. This method is faster and less likely to produce flawed results. All methods use the same
general approach. Using the derivatives of the energy with respect to coordinates, the geometry
is changed so as to lower the heat of formation. When no further change can significantly lower
the heat of formation, the optimization is stopped. The geometry then corresponds to a stationary
point on the potential surface. Except for small systems, this will be one of many possible con-
formers or isomers.
5. PRECISION

Few computer programs work to the limiting precision of the host. Many operations involve
iterative approximations to an exact result; this limits the precision of the final results. Of course,
precision should not be confused with accuracy: MOPAC is only as accurate as the underlying
methods. The level of precision is determined fully by the algorithm, however, and this can be
varied over a considerable range. Within MOPAC there are two principal operations where the
level of precision is important: passing the SCF criterion and satisfying the geometry optimization
criteria.

5.1 SCF criterion

The heat of formation and the density matrices are the principal results of an SCF calculation.
The precision with which these SCF properties are calculated is determined by the SCF criterion.
The SCF criterion is very flexible. For routine use the default criterion will ensure a heat of
formation within less than 0.1 calories of the correct semiempirical answer. For more critical work
the SCF criterion can be tightened by a factor ranging from 101 to 1015: this allows SCF proper-
ties to be determined with extreme precision. Of course, increasing the precision will result in
an increase in the time necessary to complete the SCF. Even worse, many systems which achieve
an SCF as defined by the default criterion will fail to go self-consistent with a more stringent
test. In general, however, few systems which would otherwise go self-consistent will fail to achieve
an SCF if the criterion is tightened by a factor of about 10-100.

5.2 Geometry optimization criteria

Strictly speaking, the optimized geometry is the one for which no distortion of the geometry
would lead to a decrease in the heat of formation. Such a geometry is not generally realizable;
instead, geometry optimization is terminated when one of several tests is passed. Details of these
tests are given in the Manual (q.v.), but a summary can be given here. The geometry is considered
optimized if one or more of the following calculated quantities is sufficiently small: (a) the pre-
dicted change in the geometry; (b) the predicted change in heat of formation; or (c) the current
gradient norm, i.e., the scalar of the vector of derivatives of the energy with respect to coordinates.
The definition of 'sufficiently' small is given in both the MOPAC program and Manual. How-
ever, the user can specify the degree of optimization by specifying a gradient norm in the keyword
line of the data file. The calculation is stopped when the calculated gradient norm drops below
the specified limit. In order to allow users this option, when a gradient norm is specified, all other
geometry tests are suspended.

5.3 Why specify high precision?

Routine work on rigid systems such as formaldehyde ornaphthalene does not warrant high
precision. Unfortunately, many applications involve flexible molecules and for these systems in-
creased precision is necessary. Increased precision is also needed when following a reaction path
as large geometry changes can occur with very small changes in heat of formation. This point
is not trivial: computational chemistry calculations have now achieved the status of experiments
and are routinely reported in the literature. As such they must be reproducible. Since optimized
geometries, much less initial geometries, are not normally published, anyone reproducing pub-
lished work must be able to optimize geometries in the confidence that the program will give the
same results as those reported.
For many organic systems, particularly those of biological interest, high precision is necessary
in order to achieve this degree of reproducibility. For such systems the gradient norm should be
forced below about 1 kcal/mol/A, and preferably below 0.5.
Note that while calculations on simple organics and inorganics would suggest that a gradient
of 2-3 kcal/mol/A is acceptable, this is not the case for flexible organics. For the former group
the force constants are so high that minor geometric changes are sufficient to reduce the gradient
norm to zero. On the other hand flexible molecules have such low force constants, sometimes
called flat surfaces, that large geometry changes may be necessary in order to reduce the gradient
norm, in which case, quite large energy changes can result.
This is not to imply that very high precision is always wanted or even desirable. An intelligent
choice of geometric and SCF criteria should be made, one which balances computer time against
tolerable imprecision. At present there are no hard-and-fast rules for making this choice. The SCF
criterion should be sufficiently stringent to allow the geometry optimization to function effective-
ly, while the geometry criterion should allow heats of formation to be precise within a given limit,
say 0.1 keal/mol.

6. RESULTS OF SCF CALCULATIONS

MOPAC SCF calculations generate three kinds of results: (1) results that can be compared
with experiment; (2) quantum mechanical predictions which cannot be compared with experi-
ment; and (3) quantities which are used in subsequent calculations but which are not normally
made available to users.
The first set of results are the observables such as heat of formation, dipole moment, ionization
potential, and unpaired spin density distribution. When fully optimized geometries are used, these
results can be compared with experimental values. Comparison with experiment forms one criter-
ion by which any method which is advocated for its predictive power should be judged.
Some methods, such as the Gaussian ab initio methods, attempt to predict properties from
first principles. This is a much more difficult task than that addressed by the semiempirical meth-
ods. Specifically, ab initio methods must not only predict accurately but must also carry out the
calculation starting from first principles. As a result, semiempirical methods may be more accu-
rate than ab initio for systems similar to those for which they are parameterized, but ab initio
methods are likely to be more accurate for other systems, systems for which experimental evi-
dence is currently lacking. This last class includes transition states, excited states, and exotic sys-
tems not likely to be found experimentally.
SCF calculations, being quantum mechanical in nature, allow molecular orbitals, charges,
bond orders [8] and valencies [8] etc. to be calculated. These quantities are not experimentally
verifiable. Chemists may have a good idea what the values of certain quantities are, thus the C-C
bond orders in ethane, ethylene and acetylene are likely to be about 1.0, 2.0, and 3.0, respectively.
However, these are only informed guesses there is no experimental technique for determining
these quantities. A consequence of this is that different computational methods can produce dif-
ferent values for these quantum mechanical artifacts. The quality of a method should not be deter-
mined by how well it predicts these quantities.
The last set of results of an SCF calculation are quantities used internally, such as the density
matrix. These quantities are used in subsequent calculations and are not normally printed (al-
though they can be printed upon keyword request).

7. GEOMETRY OPTIMIZATION

Only very seldom are users satisfied with a single SCF calculation on a guessed geometry. In
order to allow comparison with experiment and with calculations on other systems, the geometry
must be modified from the input geometry. The most common operation is energy minimization.
This gives rise to a stationary point on the potential surface. Note that this is not necessarily
the global minimum, i.e., the most stable conformer for the most stable structural isomer for the
input collection of atoms. There is currently no algorithmic method for consistently locating glo-
bal minima for complicated molecular systems; rather, the optimized geometry will correspond
to an isomer or conformer. It is easy to see that a method which located the global minimum
would not only be unnecessary, but also undesirable.
Consider the system C2H60. If the program located the global minimum every time it was
run then it would always predict that ethanol (AHf = - 56.2 kcal/mol) was the stable isomer and
dimethyl ether (zlHf = - 44.0 kcal/mol) would not be predicted to exist. In fact, a global optimi-
zation method would, by definition, produce only one geometry and one heat of formation. The
existence of conformers would not be envisioned by such a program, much less isomers. To sum-
marize, MOPAC, in common with all other large computational chemistry packages, does not
and can not infallibly locate the global minimum, but only local minima. While this is occasional-
ly a source of unnecessary concern, it is how MOPAC was designed to operate.
Geometry optimization proceeds by calculating the resultant forces on each atom in the system
and then moving the atoms in a direction determined by these forces so as to lower the energy
of the system. When the geometry is within a preset distance or energy of the local minimum,
the optimization is stopped.

8. REACTIONS

Few calculations are complete after the geometry has been optimized. A common next step
is to model a reaction. In its simplest form this involves identifying a reaction coordinate connect-
ing two stable systems A and B and monitoring the system as it is progressively translated along
this coordinate. In this context, a reaction coordinate is a curved line representing the path be-
tween reactants and products on a 3n-6-dimensional geometry surface. Since reaction coordi-
nates are artifacts of the model, no conclusions can be drawn about any points along the coordi-
nate with the important exception of the highest energy point (call this C). The energy of
activation is the difference in energy between compounds A and C. If the highest point reached
is on a smooth surface, then C corresponds to a transition state, a second kind of stationary point
which is independent of both the choice of reaction coordinate and coordinate system used.
Refinement of the geometry of C is not as easy as optimization of A or B. To refine the geome-
try of C, the gradient norm is minimized [9], i.e., the geometry of C is changed in such a way
as to minimize the sum of the resultant forces acting on all the atoms of C.
This operation, locating transition states, is considerably more difficult than locating minima.
If a system is optimized to locate a specific minimum, it is normally sufficient for the starting
geometry to lie anywhere within the potential energy valley of that minimum in order to be sure
that the optimization procedure will converge on that minimum. With transition states, the criter-
ion for specifying the initial geometry in order to converge on a specific transition state is much
more stringent. In this case, the initial geometry must lie between the multidimensional surface
of inflection and the desired transition state. As points of inflection are located between maxima
and minima, the volume within which the initial geometry must lie in order to locate a specific
transition state is much smaller than that for simple geometry optimization, in which a specific
conformer has to be optimized.
Once a stationary point, either ground state or transition state, has been refined, it should be
characterized by calculating the force constants. A stable minimum or conformer will have no
negative force constants, while a transition state will have exactly one negative force constant;
any stationary point with two or more negative force constants is almost certainly chemically
meaningless.
In rare cases the system cannot be refined. This will occur whenever the starting geometry is
in a domain bounded by inflection surfaces and which does not contain a stationary point - a
one-dimensional example would be a shoulder on a curve. There does not appear to be any easy
way to recover from such a situation. The best course appears to be to start again and try to
avoid entering the offending domain.
More sophisticated methods of locating the domain of a saddle point exist. An example of such
a method is the 'eigenvector following' procedure [10], in which information regarding the local
gradient is used to systematically proceed up the bottom of a valley to the transition state. At
present, such methods are not available in MOPAC. It is likely that these absences will be cor-
rected in the future.

9. VIBRATIONAL FREQUENCIES

After calculating the force constants, it is a trivial matter to mass-weight the force matrix and
calculate the vibrational frequencies. By altering individual atomic masses the effect of isotopic
substitution can also be investigated.
Special routines have been written to make the task of analyzing the vibrational modes easier
[11]. Those pairs of atoms which are most strongly involved in each vibration are indicated along
with information about the relative motion. Transition dipoles, reduced masses, and a measure
of the distance all the atoms move through during half a vibrational cycle are also displayed.

10. REACTION COORDINATES

The reaction coordinate used in getting from the ground state to the transition state has no
physical meaning, because it is dependent on the particular coordinate system used. However,
there are two different paths on the potential surface leading from the transition state to reactants
and products which have limited physical meaning.
10

The first of these paths is the Intrinsic Reaction Coordinate [12], which defines the path the
atoms would follow if the system was allowed to relax, starting from the transition state. The
resultant forces on each atom are used, together with the isotopic mass of each atom, to calculate
which direction to move the atoms in order to move downhill in energy. In other words, in a
mass-weighted 3n-6-dimensional coordinate space, the atoms are moved a small distance in a
direction perpendicular to the local energy contours. By making a large number of such move-
ments, a path is traced out to either the reactants or products. This path is a fully-damped path
in that all energy acquired by the atoms in the course of the descent is lost at each point.
The other extreme is the Dynamic Reaction Coordinate (DRC), in which the energy of motion
acquired by the atoms as they descend is not shed. Instead, the acquired velocity is used in the
prediction of the next step. A system following this trajectory is unlikely to ever achieve the geo-
metries of either reactants or products. Even if such geometries were realized they would only
be specific points on a trajectory, points when the instantaneous acceleration drops to zero and
for which the velocity is a maximum. The DRC is a time-dependent calculation for the study of
reaction dynamics. For example, the time-dependent behavior (system geometry, heat of forma-
tion, etc.) for phenomena such as the Berry pseudorotation and molecular collisions can be ob-
tained. Trajectories can be followed for up to several thousand femtoseconds.
The DRC for a vibration of a ground state system can be calculated. Vibration cycle times
can be compared with those obtained from force calculations. In general, the agreement is very
good. However, in cases where atomic excursions take the system out of the parabolic domain,
anharmonic effects become important and marked differences between DRC and force constant
vibrational lifetimes do occur.

11. EXTENDED SYSTEMS

In addition to molecules, radicals, and ions, MOPAC can also efficiently calculate polymer
properties. Conventionally, polymers have been studied using solid state methods such as tight-
binding theory. In this approach, a Brillouin zone is constructed and sampled using a regular
mesh of points. This produces a number of complex density matrices, from which the real density
matrix can be derived. Such methods are quite time consuming, but are also quite rigorous and
allow calculation of the band structure of Brillouin zones.
MOPAC is intended primarily for use by chemists, so the approach to the calculation of po-
lymer structures is quite different. Instead of sampling the Brillouin zone, only a single point is
used. The use of a single point requires the unit cell or repeat distance to be large enough for
the associated Brillouin zone to be so small that all bands are essentially flat. In practical terms,
this means that unit cells need to be greater than about 10 A.
Polymer calculations require about 50% more time than calculation on molecules equivalent
in size to the unit cell used. Results are exactly as accurate as the results for equivalent molecules.

12. SUMMARY OF MOPAC CAPABILITIES

Most users use only a very small range of MOPAC's capabilities. For example, a user may
be interested only in heats of formation, and have little interest in geometries or charges, etc. This
is possibly a result of having learned how to request certain functions and how to interpret the
 11

results. Having learned to trust a certain aspect of the program, a user may be reluctant to investi-
gate other, less familiar, aspects. One consequence of this is that the user is denied information
which may be useful to his or her research effort. The following list summarizes the more impor-
tant quantities which can be calculated using MOPAC. Note that this list does not include capa-
bilities which do not actually calculate values, such as the restart function or shutdown operation.
Users are encouraged to investigate the effect of all keywords, and to familiarize themselves with
the full range of operations of MOPAC.

12.1 Calculable quantities

The main quantities calculated by MOPAC are:

Atomic charges (Q) Molecular orbitals (P)
Atomic velocities (in DRCs) (V) M.O. bonding indices (Q)
Bond orders and valencies (Q) Molecular polarizabilities (A 3)
Density matrices (Q) Moments of inertia (I)
Dipole moments (De) Mulliken populations (Q)
Effective masses (M) Optimized geometries (G)
Entropy (C/T) Partition functions (N)
Excitation energies (E) Reaction coordinates (G)
Force constants (Dy) Spin matrices (Q)
Heat capacity (C/T) State eigenvectors (P)
Heat of formation (K) Total spin states (S)
Hyperpolarizability Transition dipole moments (De)
Internal energies (K) Unit cell lengths (G)
Ionization potentials (E) Unpaired spin densities (Q)
Localized orbitals (P) Vibrational frequencies (CM)
Molecular geometries (G) Zero point energies (K)

12.2 Units

Units given in parentheses in section 12.1 are:

cubic Angstroms
C: calories per degree
CM: c m - 1
De: Debye
Dy: dynes per centimeter
E: electron volts
G: distances in ,~ngstroms, angles in degrees
I: gram-centimeter2
K: kcal/mol
M: atomic mass units
N: a ratio, therefore no units
P: normalized wave-function
Q: charge on electron
12

S: spin units (electron = ½)

T: degrees Kelvin
V: centimeters per second.

12.3 Conversionfactors

The following conversion factors may be used to convert calculated energy units into other
units.

Conversionfactorsfor energy

e\el ao eV kcal/mol kJ/mol cm- 1

°2\
a0 1.0000000
eV 27.211652 1.000000
kcal/mol 627.47237 23.05896 1.000000
kJ/mol 2625.3444 96.47868 4.184000 1.000000
cm -1 219474.63 8065.465 349.7758 83.59842 1.000000

To convert energies in units of el into units of e2, multiply by factor. The conversion factors
are based on the assumption that Avogadro's number is 6.022045 × 1023. Note that the conver-
sion factors used within M O P A C are very slightly different from those given here. This is a result
of the historical development of MOPAC. As the differences are small, and as substituting the
more recent conversion factors would lead to incompatibility with earlier versions, the original
conversion factors have been retained.

13. T H E O R Y USED IN M O P A C

This appendix describes in detail some of the operations of MOPAC. Its objective is to bridge
the gap between theory, as it is usually presented in a textbook, and the actual algorithm used.
In most instances the style is simply descriptive, but in certain cases, like the overlap integral,
a brief derivation of the equations used is presented. In these latter cases, readers are encouraged
to follow the derivation step by step.

13.1 Geometry optimization

The most common use of M O P A C is for geometry optimization. This involves starting with
an approximation to the desired geometry and, by calculating the forces acting on the system,
changing the geometry so as to lower the total energy. Because force constants or second deriva-
tives are not known at this time, the initial steps in optimizing the geometry have to be conserva-
tive.
The heat of formation minimization routine in M O P A C is a modified Broyden-Fletcher-Gold-
farb-Shanno or BFGS method [7]. Minor changes were made necessary by the presence of phe-
nomena peculiar to chemical systems.
 13

In the following discussion of the geometry optimization theory, a matrix or tensor will be indi-
cated by upper case bold type, e.g. H, and a vector by lower case bold type, e.g., x. All other
symbols represent scalar quantities. Starting with a user-supplied geometry Xo, MOPAC com-
putes an estimate to the inverse Hessian Ho. The geometry optimization proceeds by

Xk+ ~ = Xk + ~dk

where

d k -- H gk

and each element of H is defined by

Hykp~+pky~H Q(PkP~)
a k + 1 ---- a k -- +
S S

where

y~ H Yk
Q=I
P~, Yk
S = p~: Yk
Pk = X k + l -- Xk ---- ~dk
Yk : gk + 1 -- gk

and

dEk
gk =
dx k

Two different methods are used to calculate the displacement of x in the direction d. During
the initial stages of geometry optimization, a line search is used. This proceeds as follows:
The geometry is displaced by (c~/4)d and the energy evaluated via an SCF calculation. If this
energy is lower than the original value, then a second step of the same size is made. If it is higher,
then a step of - (e/2)d is made. The energy is then re-evaluated. Given the three energies, a pre-
diction is made as to the value of e which will yield the minimum value of the energy in the direc-
tion d. Of course, the size of the steps is constrained so that the system would not suddenly be-
come unrealistic (e.g., break bonds, superimpose atoms, etc.). Similarly, the contingency in which
the energy versus e function is inverse parabolic is considered, as are rarely-encountered curves,
e.g., almost perfectly linear regressions. The line search is stopped when the drop in energy on
any step becomes less than 5% of the total drop or 0.5 kcal/mol, whichever is smaller.
As the geometry converges on a local minimum, the prediction of the search direction becomes
less accurate. There are many reasons for this. For example, the finite precision of the SCF calcu-
lation may lead to errors in the density matrix, or finite step sizes in the derivative calculation
14

(if analytical derivatives are not used) may result in errors in the derivatives. For whatever reason,
the gradient norm and energy minimum may not coincide. The difference is typically less than
0.00001 kcal/mol and less than 0.05 units of gradient norm.
Reduction of the gradient norm below about 1.0 is difficult using the line-minimization method
just outlined. A better procedure is to use a binary search. This is used whenever the gradient
norm drops below 1.0 and the scalar of d is less than 0.001. The steps involved are as follows:
,, the step size, is set to 1.0 if the scalar of dkgk is positive, otherwise c~is set to - 1.0. If Xk + edk
lowers the energy of the system, then Xk+ ~ is defined for this step, otherwise c~ is halved and
the energy re-evaluated. Once the energy is lower than E k + 0.00001, where E k is the heat of
formation on step k, Xk+ 1 is defined.
Normally, the initial guess to H, the inverse Hessian, is the unit matrix. However, in chemical
systems where the second derivatives are very large, use of the unit matrix would result in large
changes in the geometry. Thus a slightly elongated bond length could, in the first step, change
from, e.g., 1.6 * to - 6 . 5 A. To prevent this catastrophe, the initial geometry is perturbed by
a small amount, thus

xl = Xo + O.Ol*sign(go)

from which a trial inverse Hessian can be constructed:

H 1(i,i) = 0.01 *sign(g 0(i))/y 1(i)

Hdid)=O.O

A negative value for Hl(i,i) would lead to difficulties within the BFGS optimization. To avoid
this, Hl(i,i) is set to 0.06/abs(g(i)) whenever sign(go(i))/ydi) is negative.
As the optimization proceeds, the inverse Hessian matrix becomes more accurate. However,
as the geometry steadily changes, the inverse Hessian will contain information which does not
reflect the current point. This can lead to the predicted search direction vector making an angle
of more than 90 ° with the gradient vector. In other words, the search direction vector may point
uphill in energy. To guard against this, the inverse Hessian is re-initialized whenever the cosine
of the angle between the search direction and the gradient vector drops below 0.05.
Originally the Davidon-Fletcher-Powell [6] technique was used, but in rare instances it failed
to work satisfactorily. The BFGS formula appears to work as well as or better than the D F P
method most of the time. In the infrequent case when the D F P is more efficient, the increase in
efficiency of the D F P can usually be traced to a fortuitous choice of a search direction. Small
changes in starting conditions can destroy this accidental increased efficiency and make the BFGS
method appear more efficient. A keyword is provided to allow the D F P optimizer to be used.

13.2 Electronic section

Because this section involves a large number of symbols, it is convenient to have a common
set. Throughout the theory section, therefore, the same symbols are used to denote the same
quantities; exceptions will be indicated wherever they occur.
 15

Glossaryof symbols

Symbol Meaning

H One-electron matrix
F Two-electron or Fock matrix
q~ Atomic orbital
#, v, 2, cr Atomic orbital indices
~, Molecular orbital
i, j, k, I Molecular orbital indices
A, B Atom indices
c~ Alpha spin
fl Beta spin
Atomic orbital exponent
S Overlap matrix
P Density matrix
S, Px, Py, Pz S, Px, Py and Pz atomic orbitals
(expression) Dirac integral symbols
(pv[2a) Two-electron integral
(p] v) Overlap integral
6(2,a) Kronecker delta, 1 if2 = a,
otherwise 0
ei Eigenvalues
Z Nuclear or core charge

Combinations of these symbols are allowed, thus P ~ represents the fl density matrix element be-
tween atomic orbitals (p~ and c#~.

13.2.1 Approximations used in MNDO, MINDO/3, AMI and PM3

All four methods are similar enough to describe simultaneously. In the following discussion,
applications specific to each method will be indicated in the text. This will allow easy comparison
among the methods, a comparison which is not obvious from their names. M I N D O / 3 [3] stands
for Modified Intermediate Neglect of Differential Overlap, version 3, M N D O [2] stands for Mo-
dified Neglect of Diatomic Overlap. AM1 [4] is Austin Model 1, and M N D O - P M 3 [13], a very
new method, is the Modified Neglect of Diatomic Overlap, Parametric Method N u m b e r 3,
M N D O and AM1 being the first two ' M N D O ' - t y p e methods. In these methods all terms arising
from the overlap of two atomic orbitals which are on different centers or atoms are set to zero.
As this is not the forum for developing the ideas of H a r t r e e - F o c k theory, the derivation of the
Roothaan-Hall equations will be assumed, and our description of the methods will start with the
final Roothaan [14]-Hall [15] equations.

Basic Roothaan-Hall equations

Secular equation:

cdF -- ~iS[ci -- 0
16

Total electronic energy of system:

E -- ½P(H + F)

Fock matrix element:

2 a

Or, spin-free:

The methods all use a minimum basis set consisting of a maximum of one atomic orbital for
each angular quantum number. The normal basis set for any atom consists of one s and three
p orbitals, Px, Py and Pz.

13.2.1.1 Neglect of diatomic overlap integral

All overlap integrals arising from the overlap of two different atomic orbitals are neglected.
This reduces the overlap matrix to a unit matrix. The secular equation thus reduces to

ci]F -- ~;ilCi = 0

13.2.1.2 Neglect of three- and four-center integrals

Continuing with the neglect of differential overlap, all two-electron integrals involving charge
clouds arising from the overlap of two atomic orbitals on different centers are ignored. Since no
rotation can convert a two-center two-electron integral into a set of integrals involving three- and
four-center terms, rotational invariance is not compromised by this approximation. The effects
of this approximation on the Roothaan equations are as follows:
In the Fock matrix, if q), and ~0~are on different centers the MNDO, AM1, and PM3 matrix
element F~ reduces to

A B

white the MINDO/3 matrix element becomes

where

occ

P~ - c2i cai
 17

is summed over all occupied orbitals. Equivalent expressions exist for F ~ and P{~. Thus no Cou-
lombic terms are present in the two-center Fock matrix elements.
If rpu and ~0v are different but on the same center, then, since a minimal basis set is used, all
integrals of the type (gv]2~> are zero by the orthogonality of the atomic orbitals unless # = v
and 2 = cr or # = 2 and v = or. The off-diagonal one-center M N D O and AM1 Fock matrix ele-
ments become

F~ = H.~ + 2P~ + P<yvl#v > - P~,~((ktv[#v> + <##lvv>)

while the MINDO/3 element becomes

If ~o~is the same as ~0vthen, because of the symmetry of the two-electron integrals, the diagonal
MNDO, AM1, and PM3 Fock matrix elements reduce to

A B B

F~,. = Hu. + Z (P:+ ~< ##I vv > - P:~ ( #vl#v >) + Z Z Z P~+ p ( ##12¢ >
v B /t

and the corresponding MINDO/3 element becomes:

A B

v B ~

13.2.1.3 One-center two-electron integrals

The M N D O and AM1 one-center two-electron integrals are derived from experimental data
on isolated atoms. Most were taken from Oleari's [16] work, but a few were obtained by optimiza-
tion to fit molecular properties. The values of PM3 one-center two-electron integrals were opti-
mized to reproduce experimental molecular properties.
For each atom there are a maximum of five one-center two-electron integrals. These are
(ss]ss >, (ss]pp), (sp]sp), (pp]pp), and (pp]p'p'), where p and p' are two different p-type atomic
orbitals. In the original formulation [17] there was a sixth integral, (pp'lpp'), but it can be shown
that this integral is related to two of the other integrals by

@P'IPP' > = ½( @PlPP > - @PlP'P' >)

Proof:
If the molecular frame is rotated by 45 degrees about the z axis the atomic bases mix thus:

R(45°)px = ½(Px+ Py)

R(45°)py = ½(PI - Px)

R(45°) < PxPylP~Py> = ¼< (Px + Py)(Py- P01(P~ + Py)(Py- P~) >
= ¼((P~PxlPxPx) -t- <PyPylPyPy>-- <PxPx[PyPy>-- (PyPylPxPx>)
18

or

R(45°) ( PxPy[PxPy) = ½((PxPxlPxPx) - (PxPxIPyPy) )

For convenience these five integrals are given the following names

(PPlPP) Gpp
(sPlsP) H~p
(PPlPP) Gpp
(PPlP'P') Gp2
Using these definitions, the two-electron one-center contributions to the Fock matrix become

Fs~s: P~G~s + (Ppx+P -4-- P~

--py+~ -4-
- P~
~ p z +~]G
J ~ s p - - ( P p•x q- P p~y -t- P p z ) H s p
F~p: 2Ps~+ PH~p - Psp(Hsp + G~p)
Fpp." ~+ t Gsp - P~sHsp
Pss ~ d- PPppGpp+ (Pp,+ p + Pp,+~)Gp2 -½(Pp, + Pp,,)(Gpp ~ - Gp2 )
Fpp,.. pa+ 1
pp,~(Gpp- Gp2 ) -:epp,(Gpp + Gp2 )

These expressions are common to all methods.

13.2.1.4 MNDO, AM1, and PM3 two-electron two-center integrals

In a local diatomic frame there are 22 unique two-electron two-center integrals for each pair
of heavy (non-hydrogen) atoms. These are shown below:

Two-electrontwo-centerintegrals(LocalFrame)

1 (sslss) 12 (sP~lp=p.)
2 (sslp=p.) 13 (sP~lp¢p=)
3 (sslp~p,~) 14 (sslsP,,)
4 (p.p~tss) 15 (p~p,~]sp~)
5 (p~p~Iss) 16 (p~p~lsp~)
6 (P.P,~lP,~P~) 17 (sp.lsp~)
7 (p.p~lp'~p~) 18 (sp,,lsp,,)

11 (sp.lss) 22 (P,~P:IP,~P:)

Each integral represents the energy of an electron density distribution (electron l) arising from
the product of the first two atomic orbitals interacting with the electron density distribution (elec-
tron 2), which arises from the product of the second two atomic orbitals. Only if the first two
a.o.'s are the same and the second two a.o.'s are the same will the interaction energy have to
be positive, in which case the integral represents an electron~electron repulsion term. In all other
cases the sign of the integral value may be positive or negative.
With the exception of integral 22, all the integrals can be calculated using different techniques
without loss of rotational invariance. That is, no integral depends on the value of another inte-
gral, except for number 22. As with the Hppmonocentric integral, it is easy to show that
 19

(P~P~IP.PL) = ½((P.P. IP.P=) - (P.P. IPLP;) )

The electron density distributions are approximated by a series of point charges. There are four
possible types of distribution. These are as follows:

Types of electron density distribution

Monopole (1 charge) Unit negative charge centered on the nucleus
Dipole (2 charges) + ) charge located at position (x,y,z),
- ½ charge located at position ( - x, - y, - z)
Linear quadrupole (3 charges) + !2 charge located at the nucleus,
_ 1 charge at positions (x,y,z) and at ( - x, - y, - z)
Square quadrupole (4 charges) Four charges of magnitude + ¼, - ¼, + ¼ and - ¼ forming
a square centered on the nucleus.

These are used to represent the four types of atomic orbital products.

Density distributions arising from pairs of atomic orbitals

Number of
Atomic orbitals Multipole distribution charges

(ss] Monopole 1
(sPl Dipole 2
(PPl Monopole plus linear quadrupole 4
( pp'] Square quadrupole 4

Each two-electron interaction integral is the sum of all the interactions arising from the charge
distribution representing one pair of atomic orbitals with the charge distribution representing the
second pair of atomic orbitals. Thus, in the simplest case, the (sslss) interaction is represented
by the repulsion of two monopoles, while a (P~P~IP~P~), a much more complicated interaction,
is represented by 16 separate terms, arising from the four charges representing the monopole and
linear quadrupole on one center interacting with the equivalent set on the second center.
While the repulsion of two like charges is proportional to the inverse distance separating the
charges, boundary conditions preclude using this simple expression to represent the interelectron-
ic interactions. Instead, the interaction energy is approximated by

27.21
( ss[ss ) =
~/t/t/R+CA+ CB)2 1(1 1\ 2

All that remains is to specify functional forms for the terms c and A. c, the distance of a multipole
charge from its nucleus, is a simple function of the atomic orbitals; in the case of a s - p product,
this is a vector of length D1 Bohr pointing along the p axis, where
20

(2n + 1)(4~v)~n +½)

D1 = 32,_(~~+ ~p)(2n + 2)

and n is the principal quantum number. The principal quantum number is always the same
in these methods for s and p orbitals on any given atom.
The corresponding distances of the charges from the nucleus for the linear and square quadru-
poles a r e 2 D 2 and x/2D2 Bohr, respectively, where

D2=(4n2+6n+2~l/2 1

Now that the distances of the charges from the nucleus have been defined, the upper boundary
condition can be set. In the limit, when R = 0.0, the value of the two-electron integral should
equal that of the corresponding monocentric integral. Three cases can be identified:
(1) a monopole-monopole interaction, in which case the integral must converge on Gss.
(2) a dipole~lipole interaction, where the integral must converge on Hsp.
(3) the quadrupole-quadrupole interaction where the integral must converge on Hpp.
For convenience, the AA terms are given special names. These are:

Additive terms for two-electron integrals

Multipole Monocentric equivalent Name

Monopole G~s AM
Dipole H~p AD
Quadrupole Hpp = ½(Gpp- Gp2) AQ

In practice, ½(Gpp - Gp2 ) is used instead of Hpp. This eliminates any possibility of loss of rota-
tional invariance due to an incorrect value of Hpp.
In the literature, different units have been used to report the derived parameters. To date, units
of derived MNDO parameters which have been published are:

Units of derived parameters

Element Units stated Units used Ref. Element Units stated Units used Ref.
D. An Dn A~ Dn An D. An

H A A 18 P A A A A 25
Li (Nolit. Ref.) 19 S Bohr Bohr Bohr Bohr-~ 26
Be A A A A 20 C1 A A A A 27
B A A A A 21 Ge Bohr Bohr Bohr Bohr-1 28
C A A A A 18 Br Bohr Bohr Bohr Bohr-1 29
N A A A A 18 Sn None None Bohr Bohr-1 30
O A A A A 18 I Bohr Bohr Bohr Bohr-1 31
F A A A A 22 Hg None None Bohr Bohr- 1 32
AI A A A A 23 Pb None None Bohr Bohr-a 33
Si Bohr Bohr Bohr Bohr-1 24
 21

In order to maintain consistency, inverse Bohr or ao 1 are to be preferred for AM, AD and
AQ, and Bohr or a o for D1 and Dz. The other derived parameters reported may be converted
into inverse Bohr or into Bohr by one of the following two relationships:

for D~ and D2, if the units are Angstroms, divide by 0.529167.

for AM, AD, and AQ, if the units are Angstroms, take the reciprocal and multiply by 0.529167/2.

These conversion factors should work for all currently reported derived parameters, with the
exception of DQBe and DQcl, where the reported values, in A, are 0.684928 and 0.435988, respec-
tively; these should be 0.6878779 and 0.4370799.
While AM is given simply by G J 2 7 . 2 1 , AD and AQ are complicated functions of one-center
terms and the orbital exponents - recall that, in the limit, the associated charges are not all coinci-
dent. AD and AQ are solved iteratively. Given an initial estimate of AD of

I Hs p 11/3
AD = 27.21 0 2

then by iterating, an exact value of AD can be found. On iteration n the value of AD is given
by

{ Hsv an 2
)
ADn = ADn - 2 + (ADn_ 1 - ADn - z) \27"21
an - 1 -- an - 2

where

an = 2ADn - 2(4D~ + AD~- 2) -½

About 5 iterations are needed in order to get AD specified with acceptable accuracy.
I Hvv ; 1/5is made and, again, by iterating using
Similarly, for AQ an initial estimate of 27.21 (3D~)

27.21 an-2)
AQ. = AQ. _ 2 + (AQ, -1 - A Q , _ 2)
an - 1 -- an - 2

where, now,

an = 4AQn - 8(4D 2 + AQ~- 2) -½ + 4(8D 2 + AQ~- 2)_1

an exact value of AQ can be obtained. About 5 iterations are necessary.

Note that these equations are intrinsically unstable on finite-precision computers. The denomina-
tor (an - 1 - an - z) rapidly becomes vanishingly small; this is necessary in order to accurately de-
fine AD and AQ.
22

Final assembly of two-electron two-center integrals

With all the component parts defined, the two-electron two-center integrals are assembled from
the sum of all the interactions of the charges on one center with those on the other center. The
distance between the two charges must be determined - this is the vector addition of R, the inter-
atomic distance in Bohr, and the two c terms defining the location of the charges from the nucleus
- as well as the appropriate additive terms, AM, AD or AQ selected. Two examples will illustrate
this assembly:
(sslss>: This is represented by a single term. For monopoles, c = 0 and AA = AMA, A B = AMs
giving
14.399
<sslss> =
2 1F 1 1 ]2 2\1/2
0.529167 )

(ss[p=p~): p~p~ is expressed as the sum of a monopole and a linear quadrupole. This gives rise
to a total of four charges, hence four terms. However, since the interaction of the monopole with
each of the two negative charges of the dipole are the same, only three terms need to be evaluated.
In general, symmetry considerations lower the total number of terms that need to be evaluated,
so the maximum number in any integral is 8. The full integral is then represented as
14.399
<sslp=p=> -
2 1[-1 1 ]2 1672)1/2
+ 5I A + X-<.j 0.529
½ 14.399

(R2B-~- (2D~)2 + II~MA + ~Q~QBI20.5291672)1/2

!4 14.399

(R2B + I[~AMA + ~QB]20.5291672) 1/z

13.2.1.5 M I N DO /3 two-electron two-center integrals

In marked contrast to the other methods, MINDO/3 Coulomb and exchange integrals are sim-
ple. The integral is a function of the atom types and the interatomic distance only, and is of form
14.399
(AAIBB > =
(R2B + I F 14.399 + 14.39912"]1/2
2L AA /
where A A and A a are the averages of the appropriate one-center two-electron integrals. All finite
integrals over atomic orbitals on two centers are set equal. Thus, (SASAISaSa> = (SASA[papa >
= <PAPAIPBPB> = (AA[BB >.
 23

13.2.1.6 The one-center one-electron integral Hu,

This represents the energy an electron in atomic orbital q~, would have if all electrons were
removed from the system. This is approximated by adding on to the one-electron energy of the
atomic orbital in the fully ionized atom the potential due to all the other nuclei in the system.
The one-electron energy is obtained parametrically, and is given the symbol Uu,. Huu is derived
from the fundamental equation

Hu.=U..- E ZB(##]BB)
B#A

by equating the core-electron integral (##[BB) to the corresponding two-electron integral, thus

(a) M N D O and AM1 and PM3:

(##IBB) = ZB (/x~lss)
(b) MINDO/3:
<##]BB) = ZB<AAIBB >

13.2.1.7 The two-center one-electron integral Hu~

Sometimes called the resonance integral, Hu~ is approximated using the overlap integral, Su~.
Note that this violates the NDO approximation, but since resonance integrals are large, this inte-
gral is retained. This is the origin of Modified in the M N D O and MINDO/3 names. Using Slater
atomic orbitals of type

~ou = Nr n - 1 e - ~r

the overlap integral is given by

Within MNDO, AM1, and PM3, Hu~ is approximated by

1

while MINDO/3 has a very different form:

Huv = Su~flaa(Iu -[- I,)

This use of a diatomic parameter is the most distinctive difference between the MNDO/AM1/
PM3 philosophies and that of MINDO/3. Because of the difficulty in parameterizing an element
at the MINDO/3 level - the number of fl-parameters rises as the square of the number of elements
parameterized - it is unlikely that any further development of MINDO/3 will be made.

13.2.1.8 Core-core repulsion integrals

From simple electrostatics the core-core repulsion integral in eV is

14.399 ZaZ B
EN(A,B) =
RAB
24

However, the electron-electron and electron-core integrals do not collapse to the form C/RAB
(RAB in A) for distances beyond the van der Waals' radii. If the simple term given above is used,
there would be a net repulsion between two neutral atoms or molecules. To correct for this the
core-core repulsion is approximated by

EN(A,B) = ZAZB(AAIBB)

13.2.1.9 M I N D O / 3 modification to the core-core term

In addition to this term, account must be taken of the decreasing screening of the nucleus by
the electrons as the interatomic distance becomes very small. At very small distances the core-
core term should approach the classical form. To account for this, an additional term is added
to the basic core-core repulsion to give

Here ~AB is a diatomic parameter. In the case of N - H and O - H interactions only, this expres-
sion is replaced by

13.2.1.10 M N D O modification to the core-core term

The M N D O approximation to the screening effect is similar to that of MINDO/3 in practice,
but has a different functional form:

EN(A,B) = ZAZ B(SASAISBSB)[1-4- e (-~,ARA.)+ e ( -~,.a,,.)]

Again, O - H and N - H interactions are treated differently. For these interactions, use

EN(A,B)=_ZAZB(SASAISBSB)I
1+ (e( - ~"RAn)~+ e( - ~nRAn)1
13.2.1.11 AM1 and P M 3 modifications to the core-core term
These modifications are the same as that for M N D O with the addition of an extra term to
reduce the excessive core-core repulsions just outside bonding distances. The additional term
may be considered as a van der Waals' attraction term. The AMI and PM3 core-core terms are:

The extra terms define spherical Gaussian functions, the a, b, and c are adjustable parameters.
PM3 has two Gaussians per atom, while AM1 has between two and four.
 25

13.3 Heat offormation calculation

The SCF calculation produces a density, P, and Fock matrix, F. These, together with the one-
electron matrix, H, allow the total electronic energy to be calculated via

1
Eel~ t = ~-EEPu~(H.~ + V.~)
2uv

The total core-core repulsion energy is given by

E..~ = }-"}-'EN(A,B)
A<B

The addition of these two terms represents the energy released when the ionized atoms and
valence electrons combine to form a molecule. A more useful quantity is the heat of formation
of the compound from its elements in their standard state. This is obtained when the energy re-
quired to ionize the valence electrons of the atoms involved (calculated using semiempirical pa-
rameters), Eo~(A), and heat of atomization, AHf(A), are added to the electronic plus nuclear ener-
gy. This yields

AHf = Eel~o t + Enu ¢ + E E , I ( A ) + ZAHf(A)

A A

13.4 Overlap integrals

The particular technique used for the evaluation of the overlap integral depends on the atoms
involved and whether analytic derivatives are used. All four semiempirical methods use Slater-
type orbitals, STOs, although when analytic derivatives are involved [34], a Gaussian expansion
[34] of STOs is normally used.
Specific expressions for various of the overlap integrals have appeared in the literature. These
are normally used for those overlaps which involve only small principal quantum numbers, PQN,
n, and a low angular quantum number, l. For the general case, however, in which any PQN may
be encountered, the general overlap integral is used. As the final expression is rather ungainly,
a simple derivation of the overlap integral will be given.
Slater atomic orbitals are of form

(2{)~ + ~-
cp ~ = r n - le - e~Y?(0,qS)

where the Y~(0,~b) are the normalized complex spherical harmonics. Complex spherical harmon-
ics are chosen for convenience; (0,~) real orbitals have a similar behavior, but require more ma-
nipulation. The 0 dependence of spherical harmonics are the Laguerre polynomials, of form

e im¢ [-(2/+ 1)(l- m)!]l/2 sin "10 d t + m ( c o s 2 0 - 1) t

Y~(0,qS)= (2~)~[_ ~/+m-)! 2~l! (d cos 0)z+m
26

For convenience the phase factor is set to + 1; this varies according to which source is used and
the purpose for which the Laguerre polynomials are used.
Solving the differential gives

d ;+m = ~ l!(~J!!(-If +j
(d cos 0) '+m (cosB0-- 1)' J (l-~ (~-m)r (cos O)Bo-'-m

which, on rearranging to have the summation start at zero, becomes

dZ+~
--COS2 ( O - l)/ = 1/2Era
)(l- /!(2(/__j))! (.~ l)J (cosO) l - m - 2 j
(d cos O)/+m j=o j!(/-j)! ( / - m - 2 j ) !

Substituting this into the STO yields

(2~) n +~F(2I+ 1)(/- m)!11/2 Fsinm0 e im4~ "]

1/2(l-m) (2(l-j))! ( - 1)i (cos O)1-~-2j

E
j=o j! ( l - j ) ! ( / - m - 2 j ) !

At this point it is convenient to collect some of the constants together; thus,

F(I-m)!qa/2 (2(/-j))r(- 1)j

c,..j=L(~-~,j2'j!(/-j)!(l-m-2j)!
which allows us to represent the STO in a considerably simplified form:

(2~) "+~ (21+ 1)I ei~4 1/2(1-m}

g o = ~ 2~ sinmOr"-le - ¢~(2~=)~ j=o
E Clmj ( c O s O ) l - m - 2 j

The overlap integral of two STOs can then be represented as

(2~a)n"+½(2~b)"~+~ ((2/~ + 1)(2/b + 1))~

((2n.)! (2nO!)~ 2
f ~ eim4~eim4'*
sin mOa sin mob r~" - ~ r~,~- 1 e - ~e" e - rbgu__ __
2zr
0
1/2(la -m) 1/2(lb - m)
C,~< (cos 0a) ~"- m - 2j. ~ Cl~mj.(COS 0 0 'b - m - 2j. d v
j~=O Jh =0

It is impractical to solve this integral using polar coordinates. Instead, a prolate spheroidal coor-
dinate system is used. Using the identities:
 27

R(/~ + o).
ra - _ _ COS0a_ (1 + #v).
2 (`/2+ v ) '

R ( # - v). (1 - - # v )
rb - cOSOb =
2 (# - v ) '

((,/,/2 __ l ) ( 1 - - y2))½
sinO a =
+ v)

and

((/x 2 -- 1)(1 -- v2)) ½

s i n 0 b --
(`/2 - v)

R 3
this gives dv = ~ - (# + v)([2 - v) d#dvd~b.

Substituting these identities into the previous expression we get:

0 -1 1

(([22 __ 1 ) ( 1 - - V2)) m R "~-1 Rnb- 1 e - R~a(# + v)/2 e- R~b('U v)/2

- -

(#-q-v)m([2--y) m 2 n~-I (#+V) h a - 2rib_

l - 1 ([2 - - V) n b - 1
2n

(la -- m)/2 (/b -- m)/2

C l mj ClbrnJb (1 + [ 2 . ¢ ) / a - m - 2 j a (l~_~v)/b--mm2jb R 3 (`/2 + v)(# - v)d/Mvdq5
2
j. = 0
2
Jb = 0 • a (`/2 @ y ) l . - m - 2j. (`/22 __ v)lb--m-- 2ju 8

which, on integrating over ~b and rearranging, gives:

1 i l 1
<~a~b> =

-1 1

1 / 2 ( l ~ - m ) 1/2(lb--m)
C 2 y2)m la + 2ja
2
Ja = 0
2
Jb = 0
C/amja lbmjb (`/2 - - l ) m (1 - - (12 + Y) n ~ -

(`/2 - - y)nb -- lo + 2Jb(1 + #V) t" - m - 2j.(1 -- #V) tb - m- 2jbe - ½R¢.CU+ v)e - ½RCb(U- V)d[
2 dv

This is a p r o d u c t of six simple expressions of type (a + b) n. E x p a n d i n g each term as a binomial

generates six s u m m a t i o n s :
28

f n.+~nu+Z F ( 2 / ~ + 1)(2lb+ 1) a/2

-1 1
2 k l Rn.+nu+l

1/2(/~ - m) 1/2(/b - m) - lb + 2jb

E E C l ma jsC l b m J b ~ , naa-l+2Eajnb E /a- m-
2 2ja /b--m--
E ~,2j
j =0 jb=O ka= 0 kb= 0 Pa=O Pb=0 qa=0 qb=O

(lb -- m -- 2jb)! (la - m - 2j.)! ( n b - - l b + 2jb)!

(/b -- m - 2jb - qb)!qb! (/a - - m - 2j~ - qa)!qa] (nb -- 1b + 2jb -- Pb)!Pb!

(n a -- Ia + 2ja)! m! 2
( - - 1) k~ + k'° + m +Pb + qb
(n a -- 1a + 2ja -- Pa)!Pa! (m - kb)! kb!(m -- k.)! ka!

#2k~+e.+Pb+q~+qu v2kb+na--l~+2j~+m,--lb+2jb--p~--Pb+q~+qb e-½RG(u+v) e-½R~b(U--v) d/zdv

Using integration by parts, and m a k i n g use of the following integrals

i n+1 n! - An(a)
xne-"Xdx=e-a~,=l ~ aU(n-kt+l)!
1
1
n+ 1 n! ea
xne-axdx = -e-a #=IE a U ( n - # + 1)! u=l a G - - ~ - l ) ! - B"(a)
-1

the overlap integral b e c o m e s

( q ? a ~ b ) -- ~n~+½ ~pgb+~ U(21.+ 1)(21b + 1 ) ] 1/2 1/2(/.-m) 1/2(lb-m)

2 (2nna~ L Rn~+nb+l E
j.=O
Cl.mj~ E
jb=O
ClbmJ b
ka= 0

~, na-/aE + 2ja rib-/t,+

£ 2jb /a--m--
E 2ja /b--m--
2 2jb ( l b - - m -- 2jb)! (l. - m - 2j~)!
kb=O pa=O Pb=O qa=O qb=O (lb - - m - - 2 j b - - q b ) ! q b ! (/a -- m -- 2ja -- qa)!q.!

(rib -- lb + 2jb)! (n~ -- l. + 2ja)! m! 2 ( - - 1)k~+kb+m+Pb+qb

(rib -- lb + 2jb -- Pb)!Pu! (n~ - l~ + 2ja -- P~)!Pal (m -- kb)! kb!(m -- k~)! kfl

B 2 k b + na -- la + 2Ja + nb -- lb + 2Jb -- Pa -- Pb + qa + qb A2k +P +Pb+q +qb

in which the coefficients C~mj have the n u m e r i c a l values

 29

1 m j elm j l m j Clm j
0 0 0 1.0 3 0 0 5/2
1 0 0 1.0 3 1 0 (225/48) ½
1 1 0 (1/2) ½ 3 2 0 (15/8)½
2 0 0 3/2 3 3 0 (5/16) ½
2 1 0 (3/2) ½ 3 0 1 -3/2
2 2 0 (3/8) ~- 3 1 1 -(3/16fi
2 0 1 - 1/2

which is the most convenient form for algorithmic use.

13.5 Configuration interaction

For some systems a single determinant is insufficient to describe the electronic wave function.
For example, square cyclobutadiene and twisted ethylene require at least two configurations to
describe their ground states. More than one configuration is also needed if an excited state is
required - the RHF SCF converges on a ground state or, if half-electron methods are used, on
a mixture of states, while the excited state involves a different configuration. Radicals also present
a difficulty at the R H F level in that the SCF wavefunction corresponds to an equal mixture of
the two doublets, with a corresponding error in the total energy. This error can most easily be
corrected by performing a MECI calculation. To allow several configurations to be used, a Multi-
Electron Configuration Interaction technique has been developed [36].
MECI is a completely general C.I. The resulting states are space- and spin-quantized, there
is no restriction on total spin, the starting wavefunction can be closed or open shell, and even
or odd electron systems are allowed, although for simplicity the starting configuration is assumed
to be closed-shell in the description of the MECI. As MECI requires the space parts of the c~
and fi molecular orbitals to be identical, only RHF wavefunctions are used.
Each configuration which can be generated in a molecule may be represented by a single Slater
determinant; this is called a microstate. The final states will be linear combinations of these micro-
states. In general, microstates will not be eigenfunctions of the total spin operator, but will be
mixtures of different spin states.
The initial configuration used to generate the SCF is arbitrary; for half-electron systems it will
not even correspond to a microstate, each M.O. having a fractional electron occupancy. Even
if the starting wavefunction is a closed shell it would still correspond to only one of a large number
of microstates to be used in the MECI. As a result, before the MECI is started all electronic terms
arising from the electrons in the initial configuration, which will be used by MECI, are removed.
The starting wavefunction will thus consist of a low-lying doubly occupied set of M.O.s and a
high-lying empty set of M.O.s, neither of which will be involved in the MECI, and in between
a small set of M.O.s from which the electrons have been removed. This set of M.O.s will be in-
volved in the MECI.
Microstates are normally written as an antisymmetrized product of p c~and q fi electrons,

~g = [(p + q)!]-½ ~ ( - 1 ) P P[gq(1) c~(1)q/2

p
(2) (p)
o~(2)...Op a(p)gq( p + 1)fl(p+ 1)..,O{qp+q)fl(p+q)]
30

where [(p + q)!]-½ is the normalization constant, P is an operator which permutes the electron
coordinates, and (-1) P assumes the values -1 or + 1 for odd and even permutations, respectively.
Rather than having all the e electrons appearing first in a microstate it is more convenient to
order the one-electron wavefunctions in order of their indices as they occur in the full set of M.O.
If both e and fl M.O.s of the same index occur, then e precedes fl, thus:

g*g= [(p + q)!]-} ~ ( - 1)Pp[~O(t*)a(1)lp~2)fl(2)0~a)cff3)O(2*)fl(4)..4Olp+ q)fl(p + q)]

P

Only those M.O.s involved in the MECI are of interest, thus from the full set of M.O.s, filled
and empty

[01(1)~(1)
01(2)fl(2)
2(3)~(3)...]
O2(4)fl(4)...J

the ground-state configuration (assumed to be closed shell for simplicity) can be represented by

111 111000
00010

where a i represents a spin molecular orbital occupied by one electron and 0 represents an empty
M.O. Denoting the lower bound of the M.O.s involved in the C.I. by B and the upper bound
by A,

1 1 1...'1 1 100...000'...00 7
I 1 1 1..." 1 1 1 0 0 . . . 0 0 0 " . . . 0 0
B A
J
all M.O.s below (to the left of) B can be considered as part of the core while those above (to
the right of) A are empty and can likewise be ignored. We can thus focus our attention on the
M.O.s in the range B-A.
For convenience, microstates will be expressed as a sum of molecular orbital occupancies, so
that

A
= E +
i=B

For example, if the ground state configuration g*gis closed shell, then the occupancy of the M.O.s
would be

O " ~ = O e g = O g=[1 .... ,1,0 ..... 01

The electronic energy, Er, of any microstate ~ur is the sum on the one- and two-electron energies:
P q P q P q
Er = ~ Hii + Z n i i -+- 1 Z (Jij -- Kij) "Jr-½ Z (Jij-- KU) + Z Z Ju
i i ij ij i j
 31

where Hii is the one-electron energy of M.O. 0i

Jii is the two-electron Coulomb integral (~bi ~hil~hj ~kj>
and Kii is the two-electron exchange integral (¢i O~l~hl0j>
In this section it is more convenient to express it in terms of molecular orbital occupancies:
A
Er ---- E Hii(Oi:zr + Of r) -t- ~2 (l(JiJ -- Kij)(Oi~r Ojar -4- (~.flrO
_, ~ ,.~r, ~ -t'-j..('~.ar(~
-,,--, - - .flr~
j,
i=B ij

Similarly, the orbital energies can be written

p q
si~i~= Hii + ~ (Jij - K0) + ~ Jij
J J

or, in terms of orbital occupancies

A A
8~ir = H i i "4- E (Jij - K i j ) O ~ ~ + Z J i j O j pr
j =B j =B

As stated above, all electronic terms arising from the electrons of the M.O.s involved in the start-
ing configuration must be removed. This lowers the orbital energies thus:

A
8ii+ -----8ii m E ( 2 J i j - - K i ~ ) O ~
j=B

The total energy of the system after removal of these electrons is given by

A A i-1
Eg+ = Eg - ~ 2~i+ O~ -I- Jii(og) 2 + Z ~ 2(2J~j - K~j) O1~O~
i=B i=Bj=B

By redefining the system so that those filled M.O.s which are not used in the MECI are consid-
ered part of an unpotarizable core, the new energy levels s + can be identified with the one-electron
energies Hii and the total electronic energy Er of any microstate is set equal to the sum of the
energy of the electrons considered in the microstate plus Eg+ .

13.5.1 Construction of secular determinant

Microstates can be generated by permuting available electrons among the available levels. Ele-
ments of the C.I. matrix are then defined by

Evaluation of these matrix elements is difficult. Each microstate is a Slater determinant, and
the Hamiltonian operator involves all electrons in the system. Fortunately, most matrix elements
are zero because of the orthogonality of the M.O.s. Only the non-zero elements need be evaluat-
ed; three types of interaction are possible:
32

(a) ~, = ~b" Since the two wavefunctions are the same, this corresponds to the energy of a
microstate. As the electronic energy of the closed shell is common to all configurations considered
in the C.I., it is sufficient to add on to E~ the energy terms which are specific to the microstate,
thus

(~r*plHIt/*p) = Eg+ + E ( J i j - Kij)Oj Oi + (Jij - Kij)Oj#p Of pq-

i=B i=B
A A
-ij ~i --j
i=Bj=B

(b) Except for 0i in g~ and ~,j in g~b; ~//a ---- ~'/b" Assuming ~ and Oj to be e-spin the interaction
energy is

< ~alFI ~'~> = ( - 1)w(eij+. + E ( ( i j l k k ) - <ik~k))O{a + (ijlkk)O{ a

k

This presents a problem. Unlike ai+, which has already been defined, there is no easy way to calcu-
+
late aij. Rather than undertake this calculation, use can be made of the fact that, for the starting
configuration

= ( = Hij + E ( ( i j l k k ) - (ik~k))O~, g + E(ijlkk ) O~g

k k

or
A A
gij ~'~ /;i~' -1- E ((ijlkk)- (ik~k))O~g+ E (ijlkk)O~g
k=B k=B

e~ corresponds to an off-diagonal term in the Fock matrix, which at self-consistency is, by defini-
tion, zero. Therefore

A A
gi~' = - E [(ijlkk)-(ik~k)lO~g- E (ijIkk)O~g
k=B k=B

which can be substituted directly into the expression for ( gJalF]~b) to give

( ~IFI g'~) = ( - 1)wy'[(ijlkk ) - (ik~k)](0{ ~- O{g) + E(ij{kk ) (OPka-- O~kg)

k k

All that remains is to determine the phase factor. One of the microstates is permuted until the
two unmatched M.O.s occupy the same position. The number of permutations needed to do this
when the two M.O.s are of e spin is

j-1
w= E ( o ; . + o~.)
k=i+l
 33

assuming j > i; otherwise

j-1
w= oF + g ( 0 7 + o~.)
k=i+l

(c) Except for @i and ~,j in gsa and @k and ~l in ~b; ~a = ~b" Two situations exist: (i) when
all four M.O.s are of the same spin; and (ii) when two are of each spin. Thus,
(i) All four M.O.s are of the same spin. The interaction energy is

( ~tr/alVl~/Jb> = ( -- 1)w[(ikbl> - (il~k >1

in which the phase factor is

j-1 1-1
W= (Om
~a"~- Om~a) -~- E ( O ~ + vN#a$
. , , - v..1..i -('~fla
- - k-1- Nfl a
m=i+l m=k+l

if the four M.O.s are of e spin; otherwise,

j-1 1-1
W= E (Ore~ta -~- Omfla) "-~ E fla "~ OJeta "t- o~a
(O~ma-~- Ore)
m=i+l m=k+l

Two M.O.s are of each spin. In this case there is no exchange integral, therefore the interaction
energy is

< ~alFl~b> = ( - 1)w(ik~1)

and the phase factor is

i 1
w = Y (°~a+ o~9 + E (o;2 + o~9
m=k m=j

if i > k, then w = w + O [ ~ + O~ ~,
i f j > l, then w = w + O [ b + O~ b,
finally i f i > k and j > l or i < k and j < l, then w = w + l.
All other matrix elements are zero. The completed secular determinant is then diagonalized.
This yields the state vectors and state energies, relative to the starting configuration. In turn, the
state vectors can be used to generate unpaired spin density (at the RHF level) for pure spin states.
If the density matrix for the state is of interest, such as in the calculation of transition dipoles
for vibrational modes of excited or open shell systems, or for other use, the perturbed density
matrix is reconstructed.

13.6 Spin angular momentum

State functions are eigenvalues of the Sz and S z operators. The derivation of the expectation
value of the S 2 operator is given in this section.
34

The fundamental spin operators have the following effects:

Sx~ = 1/2fl Sxfl = 1/2~

Sy~ = i/2fl Syfl = - i/2~
Sza = 1/2~ Szfl = - 1/2fl

Using these expressions, various useful identities can be established:

s ~ = s~ + Sy~ + s~
I + = (Sx +iSy) I+fi=~
I - = (Sx - iSy) I - 0~ = fl
Sx2 + Sy2 = (I + I -) + i(SxSy - SySO
= (I- I +) + i(SyS x -- SxSy )
=½(I+I - + I - I +)

and finally i ( S y S x - SxSy ) = S z

For any microstate, gJ, the expectation value of the S 2 operator is given by

( s ~ ) = ( ~(Sz: + Sy~ + Sx~) ~ )

The first part of this expression is obvious, viz.:

( ~S~I ~ ) = ¼(N= - N ~)

However, the effect of Sy2 + Sx2 is not so simple. By making use of the fact that the operators in-
volve two electrons, a large number of integrals resulting from the expansion of the Slater deter-
minants can be readily eliminated. The only integrals which are not zero due to the orthogonality
of the eigenvectors, i.e., those which may be finite due to the spin operators, are

( gt[Sy2 + $ 2 [ ~ ) = 2 E [(~Oi~Oi[S2[~flYJ) - (~Oi~Oj[sal~o~qYJ)]

i<j

Using the relationships already defined, this expression simplifies as follows:

$1S2 = SlzS2z + ½(I~-I2 + I~-I~)

( ~(S 2) ~ ) = 2 y, (¼(28(msimsj) - 1 - ½(1 - 8(msimsj))(gqg'j )2]

i<j

or,
pq
3(p+q) p(p--1) q(q-1) (p+q)(p+q--1)
(~($2)~) - 4 t- ~ t 2 4 E (~iq6) 2
ij
pq
( ~(S 2) ~ ) = ½(p + q) + ¼(p _ q)2 _ y, ( ~/i~/j) 2
ij
 35

For the general case, in which the state function is a linear combination of microstates, the
expectation value of S is more complicated:

ij
As with the construction of the C.I. matrix, the elements of this expression can be divided into
a small number of different types:
(a) ~a = SUb: Since the two wavefunctions are the same, this corresponds to the expectation
value of a microstate, and has already been derived.
(b) Except for ffi in ~a and ~j in ~b; ~ , = ~b: Assuming ~i and ffj to have alpha-spin the ex-
pectation value is

( ~a[ S2 "}-Sx2[~'/b) = ( -- I'~)E ((ijlkk) - ( i k ~ k ) ) OZ" + (ijlkk) Oflka

k

The effect of the spin operator is to change the spin of the electrons but leave the space
part unchanged. All integrals vanish identically due to one or more of the following identi-
ties:

(~iq/j) = 0 ( m i m j ) = 6(i,j)
(~iq/k) = 60,k) (~j~bk) = 60,k)
Therefore, ( S 2 ) = 0
(c) Except for ~/i and ~j in ~ and g'k and g'l in ~ub; ~a = ~ub: Two situations exist: (i) when
all four M.O.s are of the same spin; and (ii) when two are of each spin.
When all four M.O.s have the same spin, the effect of the spin operator is to reverse the
spin of two M.O.s in the ket half of the integral By spin orthogonatity this results in an
integral value of zero.
In the case where two M.O.s are of ~ spin and two are of fl spin the matrix elements,
after elimination of those terms which are zero due to space orthogonality, are

( 5 2 ) = ( -- 1)w((l~ti~/k[Z2[~j~ti) -- (~ti~/i[Z2l~/j~lk))

The effect of S 2 on ~bk and @1is to reverse the spin of these functions; this gives

where ~k' has the opposite spin to that of ~k.

Thus, only if ¢i and ~j are spatially identical with ~0k and ~gl will (S 2) be non-zero. The
phase-factor w is such that if i = k and j = 1 then w = 0, and if i = 1 and j = k then w = 1,
for all other cases the matrix element is zero, so the phase of w is irrelevant. For these two
cases, the matrix element is
(S 2 ) = 1 if (I + + I-)(~bi + ~j) = (~bk+ qq)

otherwise ( S 2 ) = 0.
(d) If more than two differences exist, ( S 2 ) = 0.
36

14. AN SCF CALCULATION

Semiempirical calculations can be run on computers using readily available programs. It is pos-
sible to use such programs for research without having any knowledge of their workings. This
does not imply any failing on the part of the researcher; after all, it is possible to write extensive
computer programs while having little knowledge of how a computer works. However, in order
to more efficiently use such programs, a more than casual knowledge of the theory involved is
desirable. In the following section the details of a very simple calculation will be described. This
calculation can be carried out as a 'black box' calculation, but the following exercise may help
to satisfy the intellectual curiosity of users of semiempirical methods regarding the mechanics of
carrying out an SCF calculation.
So far we have looked at the theory used in semiempirical methods. In order to understand
how an SCF calculation works, we will now carry out a very simple example. We will use the
M N D O method because it currently is the most popular of the 'NDO' methods. The CNDO/2
method is very similar, and the example we will look at will emphasize the similarity. The system
to be examined is a regular hexagon of hydrogen atoms in which the H - H distance is 0.98314
A. Of course, a regular hexagon of hydrogen atoms is not a stable system; the only reason we
are using it here is to demonstrate the working of an SCF calculation. The optimized geometry
was obtained by defining all bond lengths to be equal, constraining all bond angles to be 120
degrees and defining the system as being planar.
We will need various reference data in order to follow the calculation. By reference to the
MOPAC source file B L O C K . F O R we find that:

Gss(H) = < o,q sll/r{ o, os> = 12.848 eV

U~s(H) = (qo~lH]q~) = - 11.906276 eV
~ = 1.331967 Bohr
/~s = - 6.989064 eV
Eisol = -- 11.906276 eV
E.tom = 52.102 kcal/mol

This exercise is designed to allow the reader to reproduce each step. All that is needed in order
to follow the working is a hand calculator.

Interatomic distance matrix (A)

Atom 1 2 3 4 5 6

1 0.0000
2 0.9831 0.0000
3 1.7029 0.9831 0.0000
4 1.9663 1.7029 0.9831 0.0000
5 1.7029 1.9663 1.7029 0.9831 0.0000
6 0.9831 1.7029 1.9663 1.7029 0.9831 0.0000
 37

The overlap integral of two Slater orbitals between two hydrogen atoms is particularly simple:

(O[O)=(e-")(a2/3+a+l)

where a = ~R/ao
At the optimum H - H distance of 0.983143 A, this yields an overlap integral of 0.4643. The
nearest-neighbor one-electron integral is thus

H(1,2) = $1,2(fl~ + ~p)/2 = - 3.2457 eV

In general, overlap integrals are more complicated and also involve angular components, but the
principles involved are the same. You may want to check other off-diagonal terms in the one-
electron matrix, or you may accept the results given here.

One-electron matrix (eV)

Atom 1 2 3 4 5 6

1 - 51.7112
2 - 3.2457 - 51.7112
3 -1.0970 -3.2457 --51.7112
4 -0.6992 --1.0970 --3.2457 --51.7112
5 - 1.0970 -0.6992 -1.0970 - 3.2457 -51.7112
6 - 3.2457 - 1.0970 - 0.6992 - 1.0970 --3.2457 --51.7112

On-diagonal one-electron integrals are more complicated than the off-diagonal terms. The one-
electron energy of an electron in an atomic orbital is the sum of its kinetic energy and stabilization
due to the positive nucleus of its own atom, Uss or Upp, plus the stabilization due to all the other
nuclei in the system.
Each electron on a hydrogen atom experiences a stabilization due to the five other unipositive
nuclei in the system. Within semiempirical theory the electron-nuclear interaction is related to
the electron-electron interaction via

Ee. n = -Zn(~Os(Pslq)sq)s)

Given the two-electron two-center integral matrix the calculation of the diagonal terms of the
one-electron matrix is straightforward:

H,, n = - 11.9063 - 2(9.6583) - 2(7.0632) - 6.3620

For interactions between an atomic orbital and a non-hydrogen atom there will be 10 terms; these
arise from all permutations of the basis set, s, Px, Py, Pz with the atomic orbital under the neglect
of differential overlap approximation. The 10 integrals are (iilss), (ii]spx), (iilpxPx), (iilsPy),
(iilpxpy), (iilpypy), (ii[sp~), (iilp~p~), (iilpypz), and (iilp~p~).
38

Two-electronintegrals(elO

Atom 1 2 3 4 5 6

1 12.8480
2 9.6583 12.8480
3 7.0632 9.6583 12.8480
4 6.3620 7.0732 9.6583 12.8480
5 7.0632 6.3620 7.0732 9.6583 12.8480
6 9.6583 7.0632 6.3620 7.0732 9.6583 12.8480

14.1 Starting density matrix

The density matrix is necessary in order to calculate the Fock matrix, but, in turn, the Fock
matrix is necessary in order to calculate the density matrix. To break this impasse, a guessed den-
sity matrix is used. The guess is very crude: all off-diagonal matrix elements are set to zero, and
all on-diagonal terms on any atom are set equal to the core charge of that atom divided by the
number of atomic orbitals. Our starting guess for H6 consists of a unit matrix.
Each iteration of the SCF calculation consists of assembling a Fock matrix from the one-elec-
tron matrix, the two-electron integrals, and the density matrix, diagonalizing it to obtain the ei-
genvectors, and finally reassembling the density matrix. At some point the change in density ma-
trix drops below a preset limit. When this happens we say that the field is self-consistent. We
will now carry out these steps for the H 6 system.

14.2 Assembly of the starting Fock matrix

In the first iteration this is particularly simple, as there are no off-diagonal terms in the density
matrix. Only the on-diagonal terms are affected. Each on-diagonal term in the Fock matrix Faa
is modified by the electrostatic field of all the electrons in the system except the electron or fraction
of an electron in the atomic orbital q~a. Consider F(1,1). The total initial population of qh is 1.0,
composed of equal amounts of a and fl electron density. An electron in q)1 would therefore experi-
ence the electrostatic repulsion of half an electron. An electron cannot repel itself, however it will
be repelled by its partner electron of opposite spin.
In addition, each electron will be affected, normally repelled, by the electrostatic field of all
the electrons on all the other atoms. Each atom has one electron, so the total energy of an elec-
tron, i.e., the diagonal Fock matrix element, is given by

F(1,1) = 51.7112 + ½(12.848) + 2(9.6583 ÷ 7.0632) ÷ 6.3620

The Fock matrix is obtained by adding the two-electron terms to the one-electron matrix. The
elements of the Fock matrix represent the sum of the one- and two-electron interactions. For
the system of six hydrogen atoms, this has the following form:
 39

Initial Fockmatrix (e1/)

Atom 1 2 3 4 5 6

1 - 5.4823
2 - 3.2457 - 5.4823
3 - 1.0970 - 3.2457 - 5.4823
4 - 0.6992 -1.0970 - 3.2457 -5.4823
5 - 1.0970 -0.6992 -1.0970 - 3.2457 - 5.4823
6 - 3.2457 - 1.0970 - 0.6992 - 1.0970 - 3.2457 - 5.4823

14.3 Diagonalization of the Fock matrix

T h e F o c k m a t r i x is t h e n d i a g o n a l i z e d t o yield t h e f o l l o w i n g set o f e i g e n v a l u e s a n d e i g e n v e c t o r s :

Energy level Molecular orbitalcoefficients

1 2 3 4 5 6

6 - 0.4857 0.4082 -0.4082 0.4082 - 0.4082 0.4082 -0.4082

5 - 1.8388 0.5774 -0.2887 - 0.2887 0.5774 - 0.2887 - 0.2887
4 - 1.8388 0.0000 0.5000 - 0.5000 0.0000 0.5000 -0.5000
3 - 6.9317 0.5774 0.2887 - 0.2887 -0.5774 -0.2887 0.2887
2 - 6.9317 0.0000 0.5000 0.5000 0.0000 -0.5000 -0.5000
1 - 14.8669 0.4082 0.4082 0.4082 0.4082 0.4082 0.4082

14.4 Exercises invoh, ing eigen vectors

In the following exercises 'verify' means using a hand calculator. They are intended to confirm
understanding of the theory involved. Work through one or more examples to confirm the validi-
ty of t h e s t a t e m e n t t h a t f o l l o w s .
1 Verify that the eigenvectors are normalized.
2 Verify that the eigenvectors are orthogonal to each other.
3 Verify that the eigenvalues are correct.
4 Verify that the eigenvectors diagonalize the Fock matrix.
5 V e r i f y t h a t t h e d i a g o n a l s u m r u l e is o b e y e d , i.e., t h a t t h e s u m o f t h e e i g e n v a l u e s is e q u a l t o
the sum of the diagonal matrix elements (the trace) of the Fock matrix.

14.5 Iterating density matrix

T h e d e n s i t y m a t r i x is t h e n r e f o r m e d u s i n g t h e o c c u p i e d set o f e i g e n v e c t o r s , i.e., t h e l o w e s t t h r e e
levels. T h i s yields:
40

Density matrix

1 2 3 4 5 6

1 1.0000
2 0.6667 1.0000
3 0.0000 0.6667 1.0000
4 -0.3333 0.0000 0.6667 1.0000
5 0.0000 -0.3333 0.0000 0.6667 1.0000
6 0.6667 0.0000 - 0.3333 0.0000 0.6667 1.0000

Verify that the density matrix is correct.

14.6 Iterating Fock matrix

The second Fock matrix can then be constructed using this density matrix. The on-diagonal
terms are identical to those in the first Fock matrix, since the atomic orbital electron densities
are unchanged, but the off-diagonal terms are now changed. The off-diagonal terms are modified
to allow for exchange interactions. (Note that not all exchange terms are stabilizing.)
Let us evaluate the matrix element F(1,2):

F(1,2) = - 3.2457 - 1(0.6667)(9.6583) eV

The second Fock matrix is given below.

Second Fock matrix (elO

Atom 1 2 3 4 5 6

1 - 5.4823
2 -6.4651 -5.4823
3 -1.0970 -6.4651 - 5.4823
4 +0.3611 -1.0970 -6.4651 - 5.4823
5 - 1.0970 + 0.3611 - 1.0970 - 6.4651 - 5.4823
6 - 6.4651 -1.0970 + 0.3611 - 1.0970 --6.4651 --5.4823

Diagonalization of this matrix yields the same set of eigenvectors as we had initially. In general,
several iterations are necessary in order to obtain an SCF; however, a few systems exist for which
symmetry restrictions on the form of the eigenvectors allow them to achieve an SCF in one itera-
tion. Hexagonal H 6 is one such system. Although the eigenvectors are the same, the eigenvalues
obviously have to be different.
Exercise: Verify that the SCF energy levels of H 6 a r e - - 2 0 . 2 4 5 5 , - 11.2115, - 11.2115, 2.4410,
2.4410, and 4.8928 eV.
Once an SCF is achieved we need to calculate the heat of formation.
 4l

14.7 Calculation of heat offormation

The heat of formation is defined as

AHf = Eelec t ÷ Enu c ÷ Eisol q'- Eatom

where Eelec t is the electronic energy, Enu c is the nuclear-nuclear repulsion energy, Eiso~ is the energy
required to strip all the valence electrons off all the atoms in the system, and Eatom is the total
heat of atomization of all the atoms in the system.
E e is calculated from 1 p ( H + F), or

1 6 6

Ee = 2-~=~
Z ~=1
Z P~,~(H~,,+ F~,,)

Using the data we have already derived, we can calculate Eelect as

E~e¢ t = 3( + 1 . 0 0 0 0 ) ( - 51.7112 + - 5.4823)

+ 6( + 0.6667)( - 3.2457 + - 6.4651)
+ 3( - 0 . 3 3 3 3 ) ( - 0.6992 + - 0.3611)

or

Eel~t = -- 210.0858 eV

Enu ~ is a relatively straightforward calculation, and is equal to 130.2864 eV. The total energy of
the system is thus - 79.7990 eV.
We are now ready to calculate the AHf. As the total energy and Etsol are in eV, we must first
convert them into kcal/mol.

AHf = 23.061( - 79.7990 + 71.4377) + 6(52.1020) kcal/mol

or

AHf = 119.792 kcal/mol

It is convenient to combine Eiso~ and Eatom together so as to simplify this calculation. In order
to convert any total energy (Ee~ect + Enuc) into a AHf, the following operation must be performed:

in which the index 'i' is over all atoms in the system.

Users of M O P A C m a y wish to verify this calculation for a system of their own choice. T o facili-
tate this, the following table m a y prove useful.
42

Values of Eisol +atom

Element Eisol+ atom(eV/atom)

MINDO/3 MNDO AM1 PM3

Hydrogen - 14.764312 - 14.165588 - 13.655739 - 15.332633

Lithium - 6.793583

Beryllium - 27.541992

Boron - 67.584394 - 70.200344 - 69.601659

Carbon - 126.880346 - 127.910952 - 128.226140 - 118.640263

Nitrogen - 192.410048 -207.466249 - 207.307791 - 162.513823

Oxygen -309.652672 -320,451178 -318.682192 -291.924879

Fluorine -475.817831 -477.502913 -483.109715 -438.336301

Aluminum - 47.931017 - 50.311708

Silicon - 95.576505 - 87.539565 - 83.701885 - 72.488357

Phosphorus - 154.270388 - 156.236921 - 124.436836 - 121.236135

Sulfur -231.996798 -228.891710 - 186.333060

Chlorine - 347.185366 - 354.374768 - 373.455532 - 316.452049

Chromium - 138.938301

Zinc - 31.231065

Germanium - 80.129955

Bromine - 347.840783 - 353.473742 - 353.699430

Tin - 95.454929

Iodine - 341.704860 - 347.970786 - 289.422586

Mercury - 29.456154

Lead - 107.856099

These numbers may be used in conjunction with the semiempirical electronic and nuclear ener-
gies to calculate the heat of formation.

15. ACCURACY OF SEMIEMPIRICAL CALCULATIONS

The following tables illustrate the relative accuracy of three semiempirical methods: MNDO,
AM 1, and MNDO-PM3. MNDO-PM3 is a recent [13] re-parameterization of the M N D O meth-
od, in which the AM 1 form of the core--core interaction is used. In cases where two sets of M N D O
parameters have been published, the more recent set will be used. Only 11 of the 12 elements
surveyed here have been parameterized at the AM1 level: these are H [4], C [4], N [4], O [4], F
[37], Si [38], P [39], S [40], C1 [37], Br [37], and I [37]. In addition, boron has been parameterized
at the AM1 level [41].
As MNDO-PM3 is very new, a word or two about it is in order. In parameterizing M N D O
and AM 1, only a very few molecules could be used. This was a natural constraint imposed by
the software and equipment available at the time these methods were being developed. MNDO-
PM3 was parameterized using a radically different optimization procedure, and used about 400-
500 experimental reference data in order to define the values of the parameters.
Average AHf errors are given in Table 2 and geometry errors are summarized in Table 9. Refer-
ence geometry data are taken from many sources, mainly from X-ray and microwave structures.
 43

No differentiation is made between these various sources, although microwave structures are pre-
ferred to X-ray. Further, no differentiation is made between various definitions of bond lengths.
Thus the bond length corresponding to the bottom of the harmonic well, r0, is not distinguished
from the equilibrium bond length, or from the bond length corresponding to the system in which
the zero point energy is included. For semiempirical methods, the average error in bond lengths
is large relative to the differences between the bond lengths obtained using the various definitions.
Average dipole errors are 0.45D (MNDO), 0.35 (AM1), and 0.38 (PM3), while average ioniza-
tion errors are 0.78 (MNDO), 0.61 (AM1), and 0.57 (PM3).
Users of MNDO-PM3 are cautioned that, at the time of going to press, little information is
available as to the accuracy of PM3 in predicting activation barriers, higher ionization potentials,
or vibrational frequencies. In consequence, results of such calculations should be treated with
caution, until a suitable survey can be done.

15.1 limitations of MNDO

MNDO is the oldest of the three methods surveyed here, and as a direct result, is the least
accurate. MNDO has many advantages over earlier semiempirical methods, but as MNDO is
such a large improvement over those methods, enumeration of MNDO's good points would be
invidious to the other methods. Instead, only the limitations likely to be encountered by users
will be mentioned. The principal drawbacks to MNDO are:
1 Sterically crowded molecules are too unstable
Example: neopentane
2 Four-membered rings are too stable
Example: cubane
3 The hydrogen bond is virtually non-existent
Example: water dimer
4 Hypervalent compounds are too unstable
Example: sulfuric acid
5 Activation barriers are generally too high
6 Non-classical structures are predicted to be unstable relative to the classical structure
Example: ethyl radical
7 Oxygenated substituents on aromatic rings are out-of-plane
Example: nitrobenzene
8 The peroxide bond is systematically too short by about 0.17 A
9 The C-O-C angle in ethers is too large by about 9°.

15.2 Accuracy of AM1

AM 1 is a distinct improvement over MNDO, in that the overall accuracy is considerably im-
proved. Specific improvements are:
1 The strength of the hydrogen bond in the water dimer is 5.5 kcal/mol, in accordance with
experiment.
2 Activation barriers for reaction are markedly better than those of MNDO.
3 Hypervalent phosphorus compounds are considerably improved relative to MNDO.
44

4 In general, errors in z/He obtained using AM1 are about 40% less than those given by
MNDO.
Unfortunately, with this improvement a few deficiencies were introduced. The most important
of these are:
1 AM1 phosphorus has a spurious and very sharp potential barrier at 3.0 ~. The effect of
this is to distort otherwise symmetric geometries and to introduce spurious activation barri-
ers. A vivid example is given by P406, in which the notionally equal P-P bonds are predicted
by AM1 to differ by 0.4 A. This is by far the most severe limitation of AM1.
2 Alkyl groups have a systematic error due to the heat of formation of the CH 2 fragment being
too negative by about 2 kcal/mol.
3 Nitro compounds, although considerably improved relative to MNDO, are still systemati-
cally too positive in energy.
4 The peroxide bond is still systematically too short by about 0.17 A.

15.3 Accuracy of PM3

The author believes that PM3 is a distinct improvement over AM 1. Evidence of this is:
1 Hypervalent compounds are predicted with considerably improved accuracy.
2 Overall errors in AHf are reduced by about 40% relative to AM1.
Although PM3 is more accurate in predicting the AHe of many systems than either AM1 or
MNDO, it does have some limitations. The more important of these are:
1 There is currently a paucity of information regarding the limitations of PM3! This should
be corrected naturally as results of PM3 calculations are reported.
2 The barrier to rotation in formamide is practically non-existent. In part, this can be cor-
rected by the use of the MMOK option (described in the MOPAC Manual).
Users of PM3 should be aware that PM3 has other defects. The only reason they are not enu-
merated here is that PM3 is a very recently developed method. As a result, these defects have
not yet been identified.
The tables given here have been carefully checked for accuracy. In many instances the calculat-
ed results differ from those of earlier publications. In most cases, the difference has been investi-
gated and found to be due to errors in the earlier work. This does not, of course, mean that the
tables are fully accurate, only that care has been taken to minimize error.

ACKNOWLEDGEMENT

This work would not have been possible without the help of my wife, Anna, whose considerable
literary skills allowed her to convert the original manuscript into something readable. I thank
her for the many hours she devoted to improving the coherence of the volume. Of course, any
errors in fact or logic are purely my responsibility.
 45

REFERENCES

1 For an introduction to these programs, see Clark, T., A Handbook of Computational Chemistry, John Wiley and
Sons, New York, NY, 1985, ISBN: 0-471-88211-9.
2 Dewar, M. J. S and Thiel, W., J. Am. Chem. Soc., 99 (1977) 4899.
3 Bingham, R. C., Dewar, M. J. S. and Lo, D. H., J. Am. Chem. Soc., 97 (1975) 1294.
4 Dewar, M. J. S., Zoebisch, E. G., Healy, E. F. and Stewart, J. J. P., J. Am. Chem. Soc., 107 (1985) 3902.
5 'DRAW', Donn M. Storch, Quantum Chemistry Program Exchange, No. 493, 1985.
6 Fletcher, R. and Powell, M. J. D., Comput. J., 6 (1966) 163; Fletcher, R., Comput. J., 8 (1965) 33; Davidon, W.
C., Comp. J., 10 (1968) 406.
7 Broyden, C. G., Journal Inst. Mathem. Appl., 6 (1970) 222; Fletcher, R., Comput. J., 13 (1970) 317; Goldfarb, D.,
Mathem. Comput., 24 (1970) 23; Shanno, D. F., Mathem. Comput., 24 (1970) 647; See also summary in: Shanno,
D. F., J. Optim. Theory Appl., 46 (1985) 87.
8 Armstrong, D. R., Perkins, P. G. and Stewart, J. J. P., J. Chem. Soc., Dalton, 838 (1973).
9 Two transition state location routines are provided: (a) Komornicki, A. K. and McIver, J. W., Chem. Phys. Lett.,
10 (1971) 303; Komornicki, A. K. and McIver, J. W., J. Am. Chem. Soc., 94 (1971) 2625; (b) Bartels, R. H., Universi-
ty of Texas, Center for Numerical Analysis, Report CNA-44, Austin, TX, 1972.
10 Simons, J., Jorgensen, P., Taylor, H. and Ozment, J., J. Phys. Chem., 87 (1983) 2745.
11 MOPAC, Version 4.00 Manual, available from QCPE, University of Indiana, Bloomington, IN 47405.
12 For a review, see Truhlar, D. G., Steckler, R. and Gordon, M. S., Chem. Rev. 87 (1987) 217.
13 Stewart, J. J. P., J. Comp. Chem., 10 (1989) 209; Stewart, J. J. P., J. Am. Chem. Soc., 10 (1989) 221.
14 Roothaan, C. C. J., Rev. Mod. Phys., 26 (1951) 89.
15 Hail, G. C., Proc. R. Soc. London, Ser. A., 205 (1951) 541.
16 Oleari, L., DiSipio, L. and DeMichells, G., Mol. Phys., 10 (1966) 97.
17 Dewar, M. J. S. and Thiel, W., Theor. Chim. Acta, 46 (1977) 89.
18 Dewar, M. J. S. and ThM, W., J. Am. Chem. Soc., 99 (1977) 4907.
19 Parameters taken from MNDOC program, written by Walter Thiel, Quantum Chem. Prog. Exch., No. 438, 2 (1982),
63.
20 Dewar, M. J. S. and Rzepa, H. S., J. Am. Chem. Soc., 100 (1978) 777.
21 Dewar, M. J. S. and McKee, M. L., J. Am. Chem. Soc., 99 (1977) 523I.
22 Dewar, M. J. S. and McKee, M. L., J. Comp. Chem., 4 (1983) 84.
23 Dewar, M. J. S. and Rzepa, H. S., J. Am. Chem. Soc., 100 (1978) 58.
24 Davis, L. P., Guidry, R. M., Williams, J. R., Dewar, M. J. S. and Rzepa, H. S., J. Comp. Chem., 2 (198l) 433.
25 Dewar, M. J. S., Freidheim, J., Grady, G., Healy, E. F. and Stewart, J. J. P., Organometallics, 5 (1986) 375.
26 Dewar, M. J. S., McKee, M. L. and Rzepa, H. S., J. Am. Chem. Soc., 100 (1978) 3607.
27 Dewar, M. J. S. and Reynolds, C. H., J. Comp. Chem., 7 (t986) 140.
28 Dewar, M. J. S. and Rzepa, H. S., J. Comp. Chem., 4 (1983) I58.
29 Dewar, M. J. S. and Healy, E. F., J. Comp. Chem., 4 (1983) 542.
30 Dewar, M. J. S., Healy, E. F. and Stewart, J. J. P., J. Comp. Chem., 5 (1984) 358.
31 Dewar, M. J. S., Grady, G. L. and Stewart, J. J. P., J. Am. Chem. Soc., 106 (1984) 6771.
32 Dewar, M. J. S., Grady, G. L., Merz, K. and Stewart, J. J. P., Organometallics, 4 (1985) 1964.
33 Dewar, M. J. S., Holloway, M., Grady, G. L. and Stewart, J. J. P., Organometallics, 4 (1985) 1973.
34 Dewar, M. J. S. and Yamaguchi, Y., Computers and Chemistry, 2 (1978) 25.
35 Stewart, R. F., J. Chem. Phys., 52 (1970) 431.
36 Armstrong, D. R., Fortune, R., Perkins, P. G. and Stewart, J. J. P., J. Chem. Soc. Faraday II, 68 (1972) ,839.
37 Dewar, M. J. S. and Zoebisch, E. G., Theochem, 180 (1988) 1.
38 Dewar, M. J. S. and Jie, C., Organometallics, 6 (1987) 1486.
39 Dewar, M. J. S. and Jie, C., Theochem, 187 (1989) 1.
40 Dewar, M. J. S., Private communication.
41 Dewar, M. J. S., Jie, C. and Zoebisch, E. G., Organometallics, 7 (1988) 513.
Appendix of Tables
1 Comparison of experimental and calculated heats of
formation for organic molecules 49
2 Comparison of experimental and calculated heats of
formation for organic radicals 58
3 Comparison of experimental and calculated heats of
formation for organic ions 60
4 Comparison of experimental and calculated heats of
formation for normal valent inorganic molecules 62
5 Comparison of experimental and calculated heats of
formation for hypervalent molecules 67
6 Comparison of experimental and calculated heats of
formation for inorganic ions 70
7 Statistical analysis of differences between experimen-
tal and calculated heats of formation 72
8 Barriers to rotation and inversion 73
9 Intermolecular interaction energies (kcal/mol) 74
10 Comparison of experimental and calculated
molecular geometries for organic molecules 75
11 Comparison of experimental and calculated
molecular geometries for inorganic molecules 84
12 Unsigned average errors in bond lengths (A) 90
13 Signed average errors in bond lengths (A) 91
14 Average errors in calculated geometries 92
15 Comparison of experimental and calculated dipole
moments 93
16 Comparison of experimental and calculated
ionization potentials 97
 49

TABLE 1
COMPARISON OF EXPERIMENTAL AND CALCULATED HEATS OF FORMATION FOR ORGANIC MOLE-
CULES

Empirical Chemical H e a t of f o r m a t i o n Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

CH 4 Methane - 17.9 - 13.0 4.9 5.9 9.1 a

CzHz Acetylene 54.2 50.7 - 3.5 3.7 0.6 b
CzH~ Ethylene I2.4 16.6 4.2 2.9 4.0 a
CzH6 Ethane -20.2 - 18.1 2.1 0.5 2.8 a
C3H 4 Allene 45.6 47.1 1.5 - 1.7 0.5 a
C3H4 Cyclopropene 66.2 68.2 2,0 2.1 8.6 c
C3H4 Propyne 44.4 40.2 - 4.2 - 3.0 - 1.0 a

C3H 6 Cyclopropane 12.7 16.3 3.5 -- 1.5 5.0 a

C3H 6 Propene 4.9 6.4 1.5 0.1 1.7 a
C3H 8 Propane -24.8 -23.6 1.2 -0.1 0.5 a
C~H2 Diacetylene 113.0 102.5 - 10.5 - 9.8 -6.9 d
C4H4 Vinylacetylene 72.8 66.4 - 6.4 - 7.2 -4.9 d
C4H6 I - M e t h y l c y c l o p r o p - I-erie 58.2 57.4 - 0.8 - 4.5 6.5 a
C4H 6 Bicyclobutane 51.9 69.2 17.3 12.2 26.2 a
C4H 6 1,2-Butadiene 38.8 38.0 - 0.8 - 5.3 -1.7 a
C4H 6 l-Butyne 39.5 35.7 - 3.8 - 3.3 -2.0 a
C4H 6 2-Butyne 34.7 29.8 - 4.9 - 9.8 -2.7 a
C,,H 6 Cyclobutene 37.5 37.7 0.2 - 6.5 8.3 c
C,,H 6 Methylenecyclopropane 47.9 44.5 - 3.4 - I0.0 -0.2 a
C4H 6 1,3-Butadiene 26.0 31.0 5.0 3.0 3.9 a
C4H 8 1-Butene - 0.2 1.8 2.0 0.6 0.6 a
C4H 8 cis-2-Butene - 1.9 - 2.5 - 0.6 - 2.4 0.3 a
C4H8 Cyclobutane 6.8 -3,8 - 10,6 - 18.7 -7.8 a
C4H 8 Isobutene - 4.3 - 3.3 1,0 2.3 3.1 a
C4H 8 trans-2-Butene - 3.0 - 3.8 - 0.8 - 2.1 -0.3 a
C4Hlo n-Butane - 30.4 - 29.1 1.3 0.7 -0.8 a
C4Hlo Isobutane - 32.4 - 29.5 2.9 5.6 3.0 a
CsH 6 Cyclopentadiene 32.1 31.8 - 0.3 0.0 5.0 c
CsH 8 1,2-Dimethylcyclopropene 46.4 46.7 0.3 - 7.2 8.2 e
CsH 8 Methylene cyclobutane 29.1 19.7 - 9.4 - 18.2 -4.0 f
CsH 8 1,cis-3-Pentadiene 19. I 21. I 2.0 0.6 1.8 a
C5H8 Cyclopentene 8.3 3.0 - 5.3 - 8.6 -5.3 c
C5H8 Bicyclo(2.1.0)-pentane 37.3 37.8 0.5 - 7.1 8.8 e
CsH 8 1,4-Pentadiene 25.3 26.6 1.3 - 0.7 -0.5 a
CsH 8 Spiropentane 44.3 43.1 - 1.2 - 10.6 6.2 a
CsH 8 1,trans-3-Pentadiene 18.1 21.3 3.2 1.0 2.5 a
CsHto 1-Pentene - 5.3 - 4.0 t.3 0.3 - 1.4 a

CsHlo 2-Methyl-l-butene - 8.6 - 7.9 0.7 2.1 1.8 a

C5Hlo 2-Methyl-2-butene - 10.1 - 12.2 -2.1 -0.4 0.1 a
C5Hlo 3-Methyl-I-butene - 6.6 - 3.9 2.7 4.3 2.2 a
CsH~o cis-2-Pentene - 7.0 - 7.7 - 0.7 - 2.3 - 1.9 a

CsH~o cis-Dimethylcyclopropane 1.3 1.4 0.1 - 3.4 3.5 e

CsHlo Cyclopentane - 18.3 -23.9 - 5.6 -- 12.2 - 10.5 c

CsH~o trans-2-Pentene - 7.9 - 8.7 - 0.8 - 2.4 -1.8 a

50

TABLE 1 (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp, Calc. PM3 MNDO AM I

C5H12 2-Methylbutane - 36.8 - 34.4 2,4 6.8 1.4 a

CsH12 Neopentane -40.3 -35.8 4,5 15.7 7.5 a
CsH12 n-Pentane -35.1 -34.5 0.6 0.7 -2.9 a
C6H6 Benzene 19.8 23.5 3.6 1.5 2.2 a
C6H6 Fulvene 47.5 56.2 8.7 6.2 15.2 a
C6H8 1,3-Cyclohexadiene 25.4 20.4 -5.0 -10.9 -7.9 a
C6Hlo 2,3-Dimethyl- 1,3-butadiene 10.8 14.0 3.2 4.3 6.6 a
C6H1o Cyclohexene - 1.1 -4.9 -3,8 -8.8 -9,0 a
C~H10 1,5-Hexadiene 20.1 21.1 1.0 -0.5 -2.3 a
C6Hlo 1,2-Dimethylcyclobutene 19.8 16.2 -3.6 -13.4 7.2 e
C6Hlo Bicyclopropyl 30.9 36.1 5.2 -2.2 8.6 a
C6H12 Cyclohexane -29.5 -31.0 -1.5 -5.3 -9.0 c
C6H>, n-Hexane -39.9 -39.9 0.0 0.8 -4.9 a
C7H8 Cycloheptatriene 43.2 42.5 -0.7 -9.4 -4.9 c
C7H s Norbornadiene 59.7 58.8 --0.9 3.2 8.0 e
C7H8 Toluene 12.0 14.1 2.1 1.6 2,4 a
C7H12 Norbornane - 12.4 - 13.7 -1.3 2.0 -2.0 g
C7H16 n-Heptane - 44.9 - 45.3 -0.5 1.0 -6.8 a
C8H8 Cubane 148.7 113.8 -34.9 -49.6 2.5 a
C8H8 Styrene 35.3 39.2 3.9 2.3 3.4 a
CsHlo Ethylbenzene 7.2 9.5 2,3 1.6 1.5 a
Call14 Bicyclo(2.2.2)-oct ane -24.1 -27.8 -3.7 -2.2 -11.9 g
CsHls n-Octane - 49.9 - 50.8 -0.8 1.4 -8.6 a
C9H2o n-Nonane - 54.7 - 56.2 -1.5 1.5 -10.7 a
CloHs Azulene 73.5 81.3 7.8 -1.4 10.9 a
CloHs Naphthalene 36.1 40.7 4.6 2.2 4.5 a
CloH16 Adamantane -31.9 -34.6 -2.7 5.5 -11.3 h
C14Hlo Anthracene 55.2 6I .7 6.5 3.6 7.7 a
C14Hlo Phenanthrene 49.5 55.0 5.5 6.2 7.9 a
CO Carbon monoxide - 26.4 - 19.7 6.7 20.5 20.7 b
CH20 Formaldehyde -26.0 -34.1 -8.1 -6.9 -5.5 a
CH40 Methanol -48.1 -51.9 -3.8 -9.3 -8.9 a
C2H20 Ketene -11.4 -9.2 2.2 4.6 5.7 a
C2H~O Acetaldehyde - 39.7 - 44.2 --4,5 -2.5 -1.8 a
CzH40 Ethylene oxide -12.6 -8.1 4.5 -2.9 3.6 a
C2H60 Ethanol - 56.2 - 56.8 -0.6 -6.8 -6,5 a
C2H6O Dimethyl ether -44.0 -48.3 -4.3 -7.2 -9.2 a
C3H60 Acetone - 51.9 - 53.3 -1.4 2.5 2.7 a
C3H60 Propanal -45.5 -49.3 -3.8 -2.5 -2.8 a
C3H60 Trimethylene oxide - 19.3 -26.7 --7.5 -17.9 -6.3 a
CsHsO Isopropanol -65.1 -64.0 1.1 -0.3 -2.9 a
C3HsO Propanol - 61.2 - 63.6 -2.4 -6.3 -9,4 a
C4H40 Furan -8.3 -4.0 4.3 -0.3 11.3 a
C4H60 2-Butenal -24.0 -27.1 -3.1 -3.4 -1.6 a
C4H60 Divinyl ether - 3.3 - 0.7 2.5 0.5 4.1 a
C4H80 2-Butanone - 57.0 - 57.4 -0.4 2.9 1.9 i
 51

TABLE l (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

C4H80 Butanal - 48.9 - 54.7 -5.8 -3.9 -6.2 a

C4HsO Tetrahydrofuran -44.0 -51.3 -7.3 -15.3 -14.4 a
C4HloO Diethyl ether - 60.3 - 59.6 0.7 0.1 -4.7 a
C4HloO t-Butanol - 74.7 - 71.3 3.4 10.4 3.1 a
CsHsO Cyclopentanone - 46.0 - 37.2 8.8 9.6 9.9 a
C5HloO Tetrahydropyran - 53.4 - 57.4 -4.0 -8.6 -13.7 a
C5H120 3-Pentanol -75.2 -73.8 1.4 1.7 -5.5 a
C6H60 Phenol -23.0 -21.7 1.4 -3.6 0.8 a
C6HloO Cyclohexanone - 54.0 - 60.2 -6.1 -6.1 -9.3 a
C7H60 Benzaldehyde - 8.8 - 10.6 -1.8 -0.8 -0.1 a
C7H80 Anisole - 17.3 - 14.6 2.7 -0.4 1.4 a
CloHsO 1-Naphthol -5.1 -4.1 1.0 -1.7 2.9 a
CIoHsO 2-Naphthol -10.1 -4.6 5.5 0.4 6.3 a
COz Carbon dioxide -94.1 -85.0 9.0 19.0 14.2 b
CH202 Formic acid - 90.5 - 94.4 -3.9 -2.1 -6.9 J
C2HzO2 trans-Glyoxal - 50.7 - 64.3 -13.6 -10.7 -8.0 a
C2H402 Acetic acid - 103.3 - 102.0 1.3 2.2 0.3 a
C2H402 Methyl formate -83.6 -87.0 -3.4 -1.9 -7.5 d
C2H602 Dimethyl peroxide -30.1 -34.1 -4.0 1.8 3.1 a
C2H602 Ethylene glycol -93.9 -95.2 -1.3 -12.2 -13.6 a
C302 Carbon suboxide - 22,4 - 24.0 -1.6 -1.1 7.8 13.
CaH402 fl-Propiolactone - 67.6 - 70.6 -3.0 -3.3 -3.4 a
C3H602 Propionic acid - 108.4 - 106.4 2.0 2.1 -0.7 a
C3H602 Methyl acetate -97.9 -94.1 3.8 4.3 1.5 d
C3H802 Dimethoxymethane - 83.3 - 87.4 -4.1 -11.1 -20.0 a
C4H60a Diacetyl - 78.2 - 83.4 -5.2 -0.6 3.3 a
C4H1oO2 Diethyl peroxide -46.1 -40.0 6.1 7.0 7.7 a
C5H802 Acetylacetone -90.5 -91.6 -I.1 6.3 4.8 a
C6H402 p-Benzoquinone -29.3 -31.5 -2.3 -3.5 4.2 a
C7H602 Benzoic acid -70.1 -66.2 3.9 2.4 2.1 a
C7H602 o-Salicylic acid -70.1 -66.2 3.9 2.4 2.1 a
C~H203 Maleic anhydride -95.2 -90.l 5.1 6.7 18.8 a
C4H603 Acetic anhydride - 137.1 - 135.0 2.1 4.5 5.4 a
C2H204 Oxalic acid - I75.0 - I74.0 1.0 -0.1 2.6 a
CHN Hydrogen cyanide 32.3 33.0 0.7 3.0 -1.3 b
CHsN Methylamine - 5.5 - 5.2 0.3 -2.1 -1.9 a
C2H3N Acetonitrile 20.9 23.3 2.4 -1.7 -1.6 k
C2H3N Methyl isocyanide 35.6 54.7 19.1 24.8 14.8 k
CaHsN Ethyleneimine (Azirane) 30.2 31.6 1.4 -5.1 2.9 a
C2HTN Ethylamine - 11.4 -- 12.5 -1.1 -1.8 -3.8 a
CzHvN Dimethylamine -6.6 -7.9 -1.3 0.0 1.0 a
C3H3N Acrylonitrile 44.1 50.2 6.1 -0.3 0.9 a
C3HsN Ethyl cyanide 12.1 18.5 6.4 1.7 0.9 d
C3H9N Isopropylamine - 20.0 - 18.8 1.3 3.7 0.7 a
C3H9N Trimethylamine - 6.6 - 10.9 -4.3 3.8 4.9 a
C3HgN n-Propylamine - 16.8 - 17.9 -1.1 -1.4 -5.3 a
52

TABLE 1 (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

C4HsN Pyrrole 25.9 27.1 1.2 6.6 14.0 a

C4HgN Pyrrolidine -0,8 - 12.0 - 11,2 - 15.0 -9.6 a
C4HlIN ~-Butylamine - 28.9 - 25.2 3.7 13.4 7.7 a
CsHsN Pyridine 34.6 30.4 - 4.2 - 5.8 - 2.6 a
CTH5N Phenyl cyanide 51.5 58.5 7.0 0.5 1.9 a

CNO NCO 38.1 32.4 - 5.7 - 1.0 0.8 b

CHNO Hydrogen isocyanate -24.3 - 15.3 9.0 13.5 9.1 b
CH3NO Formamide -44.5 -41.8 2.7 4.3 -0.3 1

C3HTNO Dimethylformamide - 45.8 - 44.6 1.2 8.4 8.9 a

Nitromethane - 17.9 - 15.9 2.0 21.2 8.0 a
CH3NO2
CH3NOz Methyl nitrite -15,8 -9.1 6,7 -20.9 -16.0 a

CzHsNO/ Nitroethane - 23.5 - 20.9 2.6 20.2 6.6 a

Glycine -93.7 -96.0 -2.3 -2.0 -7.8 a
C2HsNO2
1-Nitropropane - 30.0 - 26.8 3.2 21.9 6.2 a
C3HTNOz
-33.2 -27.1 6.1 26.9 11.6 a
C3HTNO2 2-Nitropropane
-111.4 - 101.1 10.3 12.7 6.4 a
C3HvNO2 Alanine
-34.4 -32.1 2.3 21.6 4.0 a
C4HgNO2 1-Nitrobutane
2-Nitrobutane -39.1 -31.9 7.2 29.0 10.9 a
C4HgNO2
C6HsNO2 Nitrobenzene 15.4 14.5 -0.9 20.4 9.9 d
2-Nitrotoluene 9.3 4,7 - 4.6 20.6 7,9 a
C7HTNO2
C7HTNO2 3-Nitrotoluene 4.1 5.1 1.0 24.1 13.5 ,m
CvHvNO2 4-Nitrotoluene 7.4 4.7 - 2.7 22.4 9.9 m
Methyl nitrate -29.1 -32.4 -3.3 16.7 -2.2 a
CH3NO3
Ethyl nitrate -36.8 -36.4 0.5 18.9 -0.4 a
C2HsNO3
C2HsNO3 Nitroethanol -75.1 -61.4 13.7 29.0 10.5 n
Diazomethane 71.0 61.0 - 10.0 - 3.7 - 8.4 a
CH2N2
CH2N2 N = N-CH2- 79.0 91.7 12.7 -6.6 7.8 o
Methylhydrazine 22.6 17.9 -4.7 -8.2 -5.3 a
CH#N2
Cyanogen 73.8 77.5 3,7 - 7.2 - 5.9 a
C2N2
1,1-Dimethylhydrazine 20.0 15,1 -4.9 - 1.9 4,0 a
C2H8N2
1,2-Dimethylhydrazine 22.0 15.6 -6.4 -7.0 -0.5 a
C2HsN2
Dicyanoacetylene 126.5 128.1 1.6 - 15.1 - 6.7 a
C4N2
81.3 86.0 4.7 -6.6 -5.3 a
C4H2N2 Fumaronitrile
66.5 56.0 - 10.5 - 22.9 - 11.2 a
C4H4N2 Pyridazine
47.0 38.0 -9.0 - 12.0 -3.1 a
C4H4Nz Pyrimidine
Pyrazine 46.9 39.3 -7.6 -9.1 -2.7 a
C4H4N2
C6H14N2 azo-n-Propane 8.6 5.8 -2.8 --6.0 4.3 P
C2H6N202 N-Nitrodimethylamine - 3,2 1,3 4,5 25,5 24,9 d
16.2 10.7 -5.5 19.4 5.3 a
C6H6N202 p-Nitroaniline
CH2N204 Dinitromethane -13.3 --11.9 1.4 41.2 16.2 n
CzH4N204 1,l-Dinitroethane -24.1 -17.4 6.7 47.3 21.5 n
CzH4NzO4 1,2-Dinitroethane - 22.9 - 19.6 3.3 42.9 12.9 n

I,I-Dinitropropane - 25.9 - 22.0 3.9 45.3 I6.7 a

C3H6N204
C3H6N204 1,3-Dinitropropane -31.6 -26.6 5.0 44.5 12.8 n
CaH6N204 2,2-Dinitropropane - 27.0 - 23.0 4.0 50.2 21.5 n
C4 HsN2 04 1,1-Dinitrobutane -34.1 -27.2 6.9 48.9 18.1 n
 53

T A B L E l (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Catc. PM3 MNDO AMI

C4HsN~O4 1,4-Dinitrobutane - 38.9 - 32.7 6.2 45.8 11.2 n

C6H4N204 m-Dinitrobenzene 11.3 9.2 -2.1 43.9 21.8 a

C6H4N204 o-Dinitrobenzene 2O.2 9.2 - 11.0 36.4 12.9 m

C6H4N204 p-Dinitrobenzene 13.3 10.1 -3.2 41.5 20.0 m
CTH6N204 2,4-Nitrotoluene 4.7 2.0 - 2.7 44.2 21.5 m
CvH6N204 2,6-Nitrotoluene 9.6 6.4 - 3.2 42.3 20.5 m
CHN306 Trinitromethane -3.2 -4.7 - 1.5 61.9 28.2 q
C2H3N306 1,1,1-Trinitroethane - 12.4 - 10.0 2.4 68.8 33.5 n
C3HsN306 l,l,l-Trinitropropane -18.4 -8.1 10.3 76.0 36.9 n
C6H4N306 1,3,5-Trinitrobenzene 14.9 8.0 -6.9 37.3 3.5 m
CTHsN306 2,4,6-Trinitrotoluene 12.9 3.3 -9.6 61.8 28.4 a

C3HsN309 Glycerol trinitrate - 64.7 - 76.6 -- I 1.9 56.3 -6.5 a

CH2N4 1-H Tetrazole 79.9 86.3 6.3 -26.1 29.7 a

CN408 Tetranitromethane 18.5 6.4 -- 12.1 76.5 34.6 n

CsHsN40~2 Pentaerythritol tetranitrate -92.5 -98.2 - 5.7 I02.9 --2.8 a
CS Carbon sulfide 67.0 97.3 30.3 37,5 b
CH4S Thiomethanol - 5.4 - 5.5 -0.1 --1.9 a
C2H4S Thiirane 19.7 28.8 9.1 --0.9 a

C2H6S Thioethanol -11.0 -8.7 2.3 -2.4 a

C2H6S Dimethyl thioether -8,9 -11.0 -2.1 -8.1 a
CsH6S Thietane 14.6 7.5 -7.l -19.7 a

C3HsS Isopropanthiol -18.1 -13.3 4.8 1.9 b

C3H8S 1-Propanthiol - 16.2 - 14.1 2.1 - 1.9 a

C4H4S Thiophene 27.6 30.7 3.1 --1.1 a

C4H8S Tetrahydrothiophene -8.1 -10.4 --2.3 - 16.0 a

C4HloS Butanethiol -21.1 -19.5 1.6 - 1.8 b

C6H6S Thiophenol 26.9 27.7 0,8 -3.5 a

C6H128 Cyclohexanethiol - 23.0 - 20.6 2.4 --2.7 e

CSO Carbon oxysulfide -33.8 -23.8 10.1 I1.0 a

C2H480 Thiolacetic acid - 43.5 - 38.9 4.6 2.1 o

CHNS Hydrogen isothiocyanate 30.0 39.5 9.5 13.4 b
C2HaNS Methyl isothiocyanate 27.1 36.1 9.0 9.8 k
C2HaNS Methyl thiocyanate 38.3 28.3 10.0 - 15.2 k
CSz Carbon disulfide 28.0 36.9 8.9 8.9 b
C2H6S2 Ethanedithiol- 1 , 2 -2.2 1.2 3.4 -4.1 b
C2H682 2,3-Dithiabutane -5.6 -4.8 0.8 -9.2 a

C2N2S2 S2(CN)2 82.3 78.5 -3.8 11.7 a

C2H6S3 2,3,4-Trithiapentane 0.0 -6.9 -6.9 -13.2 r
C3H483 1,3-Dithiolan-2-thione 22.7 40.4 17.7 -11.3 b
CH3F Fluoromethane -56.8 -53.8 3.0 4.1 --4.2 s

C2HF Fluoroacetylene 30.0 18.1 11.9 - 14.3 - 14.8 b

CzH3F Fluoroethylene -32.5 -28.6 3.9 -2.0 -1.5 t

CzHsF Fluoroethane - 62.9 -- 60.2 2,7 -2.2 - 3.4 a

C3HvF 2-Fluoropropane -69.4 -66.8 2.6 2.8 -0.4 a

C6H~F Fluorobenzene -27.8 -20.2 7.5 2.5 4.4 a

C2H3OF Acetyl fluoride - 106.4 - 98.7 7.7 9.9 7.6 a
54

TABLE 1 (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

C7HsO2F p-Fluorobenzoic acid -- 118.4 - 108.9 9.5 5.3 6.4

CNF Cyanogen fluoride 8.6 6.5 -2.1 -10.9 --13.0
CH2F2 Difluoromethane - 108.1 - 103.8 4.4 -3.6 -8.0
CaF2 Difluoroacetylene 5.0 - 11.6 -16.6 -26.0 --~.6
CzHzF2 gem-Difluoroethylene - 80.5 - 73.0 7.5 -3.1 --2.2
C2H4F2 1,1-Difluoroethane -118.8 -111.9 6.9 5.4 0.2
C6H4F2 o-Difluorobenzene -70.3 -63.1 7.2 -0.3 3.9
C6H4F2 m-Difluorobenzene - 74.0 - 63.3 10.7 3.1 6.1
C6H4F2 p-Difluorobenzene -73.3 -63.3 10.0 2.3 5.3
COF2 Carbonyl fluoride - 152.7 - 141.6 11.1 14.1 6.4
CHF3 Trifluoromethane - 166,3 - 162.0 4.3 2.5 -6.2
C2HF3 Trifluoroethylene -117.3 -121.5 --4.2 -13.8 -13.3
C2H3F3 1,1,1-Trifluoroethane - 178.0 - 172.3 5.7 13.6 5.4
CvHsF3 Trifluoromet hylbenzene - 143.2 - 134.9 8.3 15.7 8.8
C2HO2F 3 Trifluoroacetic acid - 255,0 - 244.0 11.0 16.9 12.3
C2NF3 Trifluoroacetonitrile -118.4 -115.1 3.3 5.2 -1.1
CF4 Carbon tetrafluoride - 223.3 - 225.1 -1.8 9.1 -2.4
C2F4 Tetrafluoroethylene - 157,9 - 168.2 -10.3 -17,0 -16,9
COF 4 Trifluoromethyl hypofluorite - 182.8 - 187,3 -4.5 19.5 5.1
C6HF 5 Pentafluorobenzene - 192.5 - 188.6 3.9 -9.2 0.5
C2F6 Hexafluoroethane -321.2 -317.8 3.4 21.6 8.0
C6F6 Hexafluorobenzene - 228.5 - 229.3 -0.8 -14.9 -2.6
C3OF 6 Perfluoroacetone - 325.2 - 340.0 -14.8 3,3 -6.4
C4F8 Perfluorocyclobutane - 369.5 - 379.2 -9.7 5.8 2.3
CHsCI Methyl chloride - 20.0 - 14.7 5.3 -2,5 1.0
C2HC1 Chloroacetylene 51.1 46.6 -4.5 1.5 -3.3
C2H3C1 Chloroethylene 8.6 9.7 1.1 -3.7 -2.7
C2H5C1 Chloroethane -26.8 -22.1 4.7 -2.0 0.6
CTHsOCI Benzoyl chloride -26.1 -18.4 7.7 2,7 10.4
CNC1 Cyanogen chloride 31.6 31.6 0.0 1.7 -7.0
CH2FCI Fiuorochloromethane - 62.6 - 57.5 5.0 -5,4 -2.7
COFC1 Carbonyl fluoride chloride - 102.0 - 93.6 8.4 9.8 I0.0
CHF2C1 Difluorochloromethane - 115,6 - 109.7 5.9 1,2 1.4
CF3C1 Trifluorochloromethane - 169.2 - 169.3 -0.1 9.6 6.6
CHzClz Dichloromethane -23.0 -17.1 5.8 -5.1 -2.9
C2H2C12 gem-Dichloroethylene 0.6 4.0 3.4 -3.2 -3,9
C2H2C12 cis-Dichloroethylene 1.0 4.0 3.0 -3.6 -4.3
C2H2C12 trans-Dichloroethyle~e 1.2 3.6 2.4 -4.9 -4.6
C2H4C12 1,1-Dichloroethane - 30.9 - 26.5 4.4 -1.6 -0.2
C2H4C12 1,2-Dichloroethane - 31.0 - 24.7 6.3 -5.4 -2.8
C6H4C12 o-Dichlorobenzene 7.1 11,1 4.0 1.5 2.1
C6H4C12 m-Dichlorobenzene 6.1 10.2 4.1 0.5 2.1
C6H4C12 p-Dichlorobenzene 5.3 10.1 4,8 1.1 2.7
COC12 Carbonyl chloride -52.6 -49,1 3,5 0.0 5.3
CHFClz Fluorodichloromethane - 67.7 - 62.0 5.7 -1.5 2.5
CFzC12 Difluorodichloromethane - 117.5 - 116.1 1.4 7.3 10.5
 55

TABLE 1 (continued)

Empirical Chemical H e a t of f o r m a t i o n Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

CHC13 Chloroform - 24.7 - 20.9 3.8 -4.3 -4.3 1

CzHC13 Trichloroethylene - 2.0 - 2.3 -0.3 -4.4 -6.4 b
C2H3C13 1,1,1-Trichloroethane - 35.5 - 31.9 3.6 4.0 3.6 a
CFC13 Fluorotrichloromethane - 69.0 - 67.3 1.7 3.8 8.1 b
CC14 Carbon tetrachloride - 22.9 - 26.0 -3.1 --2.6 -5.2 b
C2C14 Tetrachloroethylene - 2.7 - 8.1 -5.4 --5.3 -9.7 a
C2C16 Hexachloroethane - 34.5 - 36.5 -2.0 7.1 -1.3 a
C6CI 6 Hexachlorobenzene - 8.6 - 9.1 -0.5 6.0 0.8 a
CH3Br Bromomethane - 9.1 - 2,0 7,1 --1.3 2.9 a
C2H3Br Bromoethylene 18.7 23.8 5.1 --2.9 -0.8 a
CzHsBr Bromoethane - 15.2 - 11.4 3.8 --1.8 2.1 a
C3HsBr 3-Bromopropene 10.9 15,3 4.4 --2.I 1.5 w
C3HTBr 1-Bromopropane - 20.5 - 16.7 3.8 -1.5 0.6 a
C3H7Br 2-Bromopropane - 23.5 - 20.9 2.6 2.8 5.6 x
C6H5Br Bromobenzene 25.2 31.0 5.8 -1.3 1.5 a
C2H3OBr Acetyl b r o m i d e - 45.6 - 43.5 2.1 2.4 11.3 a
CTHsOBr Benzoyl bromide - 11.6 - 7.7 4.0 0.7 12.9 a
CNBr Cyanogen bromide 43.3 53.7 10.4 --3.3 -10.8 a
CF3Br Bromotrifluoromethane - 155.1 - 157.9 -2.8 8.5 10.5 b
CC13Br Trichlorobromomethane - 9.3 - 14.1 -4.8 --5.8 -5.4 a
CH2Brz Dibromomethane - 3.5 7.9 11.4 --1.6 2.5 Y
COBr2 Carbonyl bromide - 20.1 - 25.3 -5.2 -11.5 2.3 a
C2F4Br2 1,2-Dibromotetrafluoroethane - 189.0 - 191.3 -2.3 18.6 27.8 x
CHBr 3 Bromoform 4.4 17.5 13.1 - 1.3 2.0 Y
CBr4 Carbon tetrabromide 35.1 32.9 -2.2 -21.3 -19.2 z
CH3I Methyl iodide 3.4 9.4 6.0 -- 1.5 2.3 a
C2HsI Iodoethane - 2.0 2.1 4.1 -2.5 0.9 a
C3HsI Allyl i o d i d e 22.8 30.2 7.4 -3.3 -0.4 a
C3HsI E- 1 - I o d o - 1 - p r o p e n e 22.3 25.4 3.2 -7.8 -2.6 c
CaHsI Z - 1- I o d o - 1- p r o p e n e 20.7 29.2 8.6 -5.4 -0.2 c
CaHvI l-Iodopropane - 7.1 - 3.0 4.2 --2.2 -0.7 a
CaHTI 2-Iodopropane - 9.8 - 5.3 4.5 2.3 4.1 a
C4H9I l-Butyl iodide - 12.0 - 8.3 3.7 -2.0 -2.7 a
C4H9I 2-Iodo-2-methylpropane - 17.2 - 12.5 4.7 10.7 8.9 c
C6H5I Iodobenzene 39.4 44.7 5.3 -6.9 -1.3 c
C6Hlli Iodocyclohexane - 11.9 - 11.7 0.2 -5.0 -8.1 c
C7H7I o-Iodotoluene 31.7 38.8 7.1 -4.7 0.1 c
C7H7I m-Iodotoluene 31.9 35.3 3.4 -7.1 -1.4 c
C7H7I p-Iodotoluene 29.1 35.3 6.2 -4.4 1.3 a
C7H7I Benzyl i o d i d e 25.1 37.6 12.5 3.6 5.8 a
C2H3OI Acetyl iodide - 30.2 - 29.9 0.3 3.3 9.5 c
C3HsOI 1-Iodo-2-propanone - 31.2 - 26.5 4.7 -1.6 0.2 c
CvHsOI Benzoyl iodide 2.5 8.0 5.5 2.6 11.6 c
CNI Cyanogen iodide 53.7 63.5 9.8 -14.1 -11.I c
CF3I Trifluoroiodomethane -- 140.5 -- 137.9 2.5 12.3 7.9 c
CH2I 2 Diiodomethane 27.0 33.5 6.5 --10.2 -5.5 aa
56

TABLE 1 (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

CzHzIz E- 1,2-Diiodoethene 49.6 55.0 5.4 -- 14.3 -- 5.5 C

C2H2Iz Z- 1,2-Diiodoethene 49.6 60.5 11.0 --14.5 --6.1 C
C2H412 1,2-Diiodoethane 16.0 23.3 7.4 --4.4 --0.2 C
C3H6I 2 1,2-Diiodopropane 8.5 20.8 12.3 - 1.2 1.2 a
C4H812 1,2-Diiodobutane 2.7 16.1 13.4 3.7 3.5 a

C6H4Iz o-Diiodobenzene 60.2 73.8 13.7 --15.4 --6.1 c

COI 2 Carbonyl iodide 9.6 - 1.5 --11.1 --15.1 --3.0 z
CHI3 Iodoform 50.4 60.6 10.2 - 18.4 -- 12.6 Y
CI4 Carbon tetraiodide 108.2 102.7 -- 5.5 - 61.3 -- 54.0 Z
CaH9AI Trimethylaluminum --20.9 -5.7 15.2 - 19.2 a

CHsSi Methylsilyl 30.5 23.3 -- 7.2 - 11.5 - 21.5 bb

CH6Si Methylsilane -7.8 -3.6 4.2 -- 5.8 -- 3.0 C
CzH~Si Vinylsilane - 1.9 19.7 21.6 8.2 13.3 V
C2HTSi Dimethylsilyl 14.3 3.9 -- 10.4 -- 13.6 --21.8 bb
C2HsSi Ethylsilane - 15.0 - 10.2 4.8 --6.7 --1.5 CC
C2HsSi Dimethylsilane -20.0 -20.8 -- 0.8 -- 9.2 - 5.6 C
C3H9Si Trimethylsilyl - 0.8 - 14.9 --14.1 --16.9 --22.1 bb
C3HloSi Trimethylsilane - 37.4 - 37.3 0.1 --6.9 -1.9 C

C4H12Si Diethylsilane - 43.6 - 32.6 11.0 --0.5 7.3 dd

C4H12Si Tetramethylsilane -55.7 -53.7 2.0 -- 3.4 3.6 ee
C5H12Si 1,1-Dimethylsilacyclobutane -33.7 -35.6 - - 1.9 14.7 -- 3.5 c

C6Ha6Si Triethylsilane - 39.5 - 55.4 -15,9 -24.5 --15.2 dd

CsH2oSi Tetraethylsilane - 64.4 - 77.9 - 13.5 -- 17.8 --7.4 dd
C3HloSiO Trimethylsilicon hydroxide - 119.4 - 115.4 4.0 --2.5 8.9 a

C2HvSiC1 Chlorodimethylsilane - 69.9 - 63.7 6.2 3.1 --0.1 dd

C3HgSiC1 Chlorotrimethylsilane -84.6 -81.6 3.0 --2.8 --0.2 a

CHaSiC12 Dichloromethylsilane -96.0 -88.9 7.1 --1.4 --2.5 a

C2H6SIC12 Dichlorodimethylsilane - 109.5 - 108.1 1.4 - 1.8 - - 5.8 c

CH3SiC13 Trichloromethylsilane - 131.2 - 132.9 -- 1.7 0.6 -- 12.4 c
C3H9SiBr Trimethylbromosilane - 70.0 - 68.5 1.5 7.7 3.6 a
C6Ha8Si2 Hexamethyldisilane -85.8 -83.2 2.7 12.1 19.7 C
C6HasSiaO Hexamethyldisiloxane - 185.8 - 182.8 3.0 --8.8 18.1 a
C6H19Si2N Hexamethyldisilazane - 113.9 - 120.8 --6.9 -9.7 --4.1 a

CHP Methinophosphine 35.8 46.5 10.7 6.4 11.6 b

CHsP Methylphosphine -7.0 -9.5 --2.5 7.7 5.8 ff
CzHTP Dimethylphosphine - 15.0 - 19.6 --4.6 -- 17.1 3.4 ff
C2HTP Ethylphosphine - 12.0 - 11.7 0.3 9.1 8.1 ff
C3H~P Trimethylphosphine 22.5 - 29.8 --7.3 --25.7 0.9 a

C,HllP Diethylphosphine -25.0 -23.3 1.7 -- 20.0 6.7 ff

C6H15P Triethylphosphine -11.8 -36.7 -- 24.9 -- 53.0 -- 18.2 a
 57

References to Table 1

a Cox, J.O. and Pilcher, G., Thermochemistry of Organic and Organometallic Compounds, Academic Press, New
York, NY, 1970.
b Chase, M.W., Davies, C.A., Downey, J.R., Frurip, D.R., McDonald, R.A. and Syverud, A.N., JANAF Thermo-
chemical Tables, 3rd ed., J. Phys. Chem. Ref. Data 14, Suppl. 1 (1985).
c Pedley, J.B. and Iseard, B.S., CATCH Tables of Silicon Compounds, University of Sussex, 1972.
d Stull, D.R., Westrum, Jr., E.F. and Sinke, G.C., The Chemical Thermodynamics of Organic Compounds, Wiley,
New York, NY, 1969.
e Turner, R.B., Goebel, P., Mallon, B.J., yon E. Doering, W., Coburn, J.F. and Pomerantz, M., J. Am. Chem. Soc.,
90 (1968) 4315.
f Good, W.D., Moore, R.T., Osborn, A.G. and Douslin, D.R., J. Chem. Thermodyn., 6 (1974) 303.
g Schleyer, P.v.R., Williams, J.E. and Blanchard, K.R., J. Am. Chem. Sot., 92 (1970) 3277.
h Average value from two determinations: Mansson, M., Rapport, N. and Westrum, E.F., J. Am. Chem. Soc., 92
(1970) 7296, Butler, R. S., Carson, A. S., Laye, P. G. and Steele, W. V., J. Chem. Thermodyn., 3 (1971) 277.
i Chao, J. and Zwolinski, B.J., J. Phys. Chem. Ref. Data, 5 (1976) 319.
j Levin, R.D. and Lias, S.G., Ionization Potentials and Appearance Potential Measurements, 1971-1981, Natl.
Stand. Ref. Data Ser., Natl. Bur. Stand. 71 (1982), Cat. No. CI3.48:71.
k Wagman, D.D., Evans, W.H., Parker, V.B., Hawlow, T., Bailey, S.M. and Schumm, R.H., Natl. Bur. Stand. (U.S.),
Tech. Note, No. 270-3 (1968) and Errata in NBS Technical Note 270-8 (1981).
1 Stull, D.R. and Prophet, H., Natl. Stand., Ref. Data Ser. (U.S., Natl. Bur. Stand.) NSRDS-NBS 37, 197l.
m Hwang, D., Tamura, M., Yoshida, T., Tanaka, N. and Hosoya, F., J. Energetic Matls, (1989) in press.
n Istomin, B.I. and Palm, V., ReaMs. Sposobnosi Org. Soedin., 10 (1973) 583.
o Benson, S.W., Cruickshank, F.R., Golden, D.M, Haugen, G.R., O'Neal, H.E., Rodgers, A.S., Shaw, R. and Walsh,
R., Chem. Rev., 69 (1969) 279.
p Engel, P.S., Wood, J.L., Sweet, J.A. and Margrave, M.L., J. Am. Chem. Soc. 96 (1974) 2381.
q Carpenter, G.A., Zimmer, M.F., Baroody, E.E. and Robb, R.A., J. Chem. Eng. Data, 15 (1970) 553.
r Hobrock, B.G. and Kiser, R.W., J. Phys. Chem, 67 (1963) 1283.
s Rodgers, A.S., Chao, J., Wilhoit, R.C. and Zwolinski, B.J., J. Phys. Chem. Ref. Data, 3 (1974) 117.
t Lake, R.F. and Thompson, H., Proc. R. Soc. London, Ser. A, 315 (1970) 323.
u Gordon, A.S., Int. L Chem. Kinet., 4 (1972) 541.
v Wagman, D.D., Evans, W.H., Parker, V.B., Schumm, R.H., Halow, I., Bailey, S.M., Churney, K.L. and Nuttall,
R.L., J. Phys. Chem. Ref. Data Suppl. 11 (1982) 2.
w Frost, D.C., McDowell, C.A. and Vroom, D.A., J. Chem. Phys., 46 (1967) 4255.
Pietro, W.J., Francl, M.M., Hehre, W.J., DeFrees, D.J., Pople, J.A. and Binkley, J.S., J. Am. Chem. Soe., 104 (1982)
5039.
x Natl. Bur. Stand. Selected Values of Chemical and Thermodynamic Properties, Rossini, 1st Feb. (1952).
y Kudehadker, S.A. and Kudehadker, A.P., J. Phys. Chem. Ref. Data, 4 (1975) 457.
z Ditter, G. and Niemann, U., Philips J. Res., 37 (1982) 1.
aa NBS Technical Note 270-3, Jan. (1968).
bb Walsh, R., Acc. Chem. Res., 14 (1981) 246.
cc Steele, W.C., Nichols, L.D. and Stone, F.G.A., J. Am. Chem. Soc., 84 (1962) 4441.
dd Pedley, J.B. and Rylance, G., Sussex-N.P.L. Computer Analysed Thermochemical Data: Organic and Organometall-
ic Compounds, Sussex University, 1977.
ee Steele, W.V., J. Chem. Thermodyn., 15 (1983) 595.
ff Wada, Y. and Kiser, R.W., J. Phys. Chem., 68 (1964) 2290.
58

TABLE 2
COMPARISON OF EXPERIMENTAL AND CALCULATED HEATS OF FORMATION FOR ORGANIC RADI-
CALS

Empirical Chemical H e a t of f o r m a t i o n Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

CH Methylidyne 142.4 146.8 4.4 1.2 2.6 a

CH 2 M e t h y l e n e , triplet 92.3 75.6 - 16.7 - 15.0 - 11.5 b
CH 3 M e t h y l radical 34.8 29.8 - 5.0 - 9.0 - 3.5 b
C2 C a r b o n , d i m e r , triplet 200.2 230.3 30.1 32.2 16.1 c
C2H 3 Vinyl 59.6 63.3 3.7 4.2 5.2 d
Call4 Methylmethylene 90.3 88.6 - 1.7 - 1.9 - 2.6 e
C2H5 Ethyl radical 25.0 17.3 - 7.7 - 12.2 - 6.9 f
C3 Carbon, trimer 196.0 206.6 10.6 24.3 16.4 b
C3H 5 Allyl 40.0 39.6 -0.4 -4.6 - 1.4 d
Call 7 i-Propyl radical 16.8 5.5 - 11.3 - 15.4 - 10.0 f
C4 C a r b o n , t e t r a m e r , triplet 232.0 265.9 33.9 44.0 47.9 b
C~.H 9 Isobutyl 4.5 - 5.9 - 10.4 - 11.7 - 7.4 f
CHO HCO 10.4 - 9.3 - 19.7 - 10.8 - 11.4 b
CH20 H y d r o x y m e t h y l e n e (trans) 27.2 16.4 - 10.8 - 13.7 - 10.5 e
CH30 Methoxy - 0.5 - 6.8 - 6.3 0.3 - 3.2 f
C2H20 HCCOH 36.0 23.2 - 12.8 - 16.7 - 11.5 e
CN Cyanide 104.0 128.0 24.0 25.3 10.4 b
CH4N CH3 - N H 37.0 27.3 - 9.7 - 4.3 - 2.9 f
CF Fluoromethylidyne 61.0 54.0 - 7.0 - 22.4 - 23.0 b
COF COF - 42.3 - 55.0 - 12.7 - 7.7 - 13.6 c
CHOF HCOF - 90.0 - 88.8 1.2 1.2 - 2.9 b
CF 2 Difluoromethylene - 45.0 - 49.1 - 4.1 - 20.2 - 23.0 g
CF3 Trifluoromethyl - 112.4 - 132.1 - 19.7 - 24.7 - 30.4 b
C2H2F3 CFaCH 2 - 123.6 - 131.2 - 7.6 - 6.0 - 7.8 h
CCI Chloromethylidyne 111.3 105.3 - 6.1 - 3.9 - 10.2 c
CHC1 Chloromethylene 80.0 83.2 3.2 0.9 - 2.3 b
COC1 COC1 - 15.0 - 16.1 - 1.1 -0.6 -0.4 b
CC12 Dichloromethylene 57.0 57.5 0.5 0.6 - 8.5 b
CC13 Trichloromethyl 21.0 1.6 - 19.4 - 20.5 - 25.7 c
CBr Bromomethylidyne 125.9 138.4 12.5 1.6 0.9 c
CBr2 Dibromomethylene 84.3 104.9 20.6 7.8 5.7 c
CBr3 Tribromomethyl 64.7 64.2 - 0.5 - 35.8 - 37.8 e
CI Iodomethylidyne 144.8 145.5 0.7 7.1 6.2 c
COI COI 63.5 - 2.9 - 66.4 - 62.5 - 58.8 c
CI2 Diiodomethylene 120.4 121.6 1.2 - 15.5 1.2 c
CI 3 Triiodomethyl 117.3 105.7 - 11.6 - 68.0 - 52.7 c
CP Carbon phosphide 107.5 119.6 12.1 22.1 23.6 b
 59

References to Table 2

a Levin, R.D. and Lias, S.G., Ionization Potentials and Appearance Potential Measurements, 1971 - 1981, Natl. Stand.
Ref. Data Set., Natl. Bur. Stand. 71 (1982), Cat. No. C13.48:71.
b Chase, M.W., Davies, C.A., Downey, J.R., Frurip, D.R., McDonald, R.A. and Syverud, A.N., JANAF Thermochem-
ical Tables, 3rd ed., J. Phys. Chem. Ref. Data 14, Suppl. 1 (1985).
c Ditter, G. and Niemann, U., Philips J. Res., 37 (1982) 1.
d Lossing, F.P., Can. J. Chem., 49 (1971) 357.
e Schroder, S. and Thiel, W., J. Am. Chem. Soc., 107 (1985) 4422. BEST ab-initio involving at least MP2 correction
to 6 - 31G* wavefunetions.
f Franklin, J.L., Dillard, J.G., Rosenstock, H.M., Herron, J.T., Draxl, K. and Field, F.H., Natl. Stand. Ref. Data Ser.,
Natl. Bur. Stand., No. 26 (1969).
g Vogt, J. and Beauchamp, J.L., J. Am. Chem. Soc., 97 (1975) 6682.
h Wu, E.C. and Rodgers, A.S., J. Phys. Chem., 78 (1974) 2315.
60
TABLE 3
C O M P A R I S O N OF E X P E R I M E N T A L A N D C A L C U L A T E D HEATS O F F O R M A T I O N F O R O R G A N I C IONS

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

CH3 Methyl ( + ) 261.0 256.5 -4.5 - 17.1 -8.6 a

C2H 3 Vinyl ( + ) 266.0 263.9 -2.1 -0.3 -4.5 b
C2H 4 Ethylene ( + ) 257.0 248.7 -8.3 -18.7 -13.1 c
Call5 Ethyl ( + ) (classical) 216.0 222.5 6.5 3.7 0.8 d
C2H5 Ethyl ( + ) (non-classical) 216.0 232.1 16.1 18.7 10.3 d
C3H 3 Propynyl ( + ) 281.0 275.3 - 5.7 - 15.6 - 7.3 c

C3H 3 Cyclopropenyl ( + ) 257.0 269.8 12.8 15.5 19.4 c

C3H 5 Allyl ( ÷ ) 226.0 232.7 6.7 - 4.6 0.2 c
C3H5 Propenyl ( + ) 237.0 238.2 1.2 3.1 -3.3 c
C3H5 Cyclopropyl ( + ) 235.0 261.8 26.8 23.2 25.6 c
C3H 7 Propyl ( + ) 208.0 214.4 6.4 4.4 - 0.2 c
C3H 7 2-Propyl ( + ) 192.0 197.3 5.3 8.7 -0.1 d
C4Hv 2-Butenyl ( + ) 200.0 212.6 12.6 7.0 6.3 c
C4H 7 Cyclobutyl ( + ) 213.0 225.6 12.6 8.4 13.2 c
C4H 9 n-Butyl ( + ) 201.0 208.0 7.0 5.8 - 1.1 c
C4H 9 1-Methyl propyl ( + ) 183.0 190.8 7.8 11.0 1.0 c
C,,H 9 Isobutyl ( + ) 176.0 178.7 2.7 12.0 - 1.2 c
CsH 5 Cyclopentadienyl ( - ) 21.3 15.9 - 5.4 - 2.4 3.9 e

CsH 9 Cyclopentyl ( + ) 188.0 193.5 5.5 6.2 -2.1 f

CsHI1 1-Pentyl ( + ) 194.0 202.4 8.4 7.9 - 1.2 c
CsHxl 2-Pentyl ( + ) 173.0 184.6 11.6 15.5 3.4 c
C5Hll 2-Ethylisopropyl ( + ) 156.0 171.9 15.9 25.6 10.5 c
CsH11 Neopentyl ( + ) 188.0 171.8 - 16.2 - 6.4 - 21.5 g

C6Hxl 1-Methylcyclopentyl ( + ) 165.0 174.5 9.5 13.6 2.4 c

C6HI1 Cyclohexyl ( + ) 177.0 186.1 9.1 9.9 -2.8 f
CTHv Benzyl ( + ) 216.0 227.4 11.4 2.0 6.1 h
CTH 7 Tropylium ( + ) 209.0 221.0 12.0 1.3 1.4 i
CvH~I 2-Norbornyl ( + ) 182.0 208.5 26.5 31.1 21.0 c
CHO HCO ( + ) 199.0 176.9 -22.1 -14.1 -11.5 j
CH30 CH2OH (+) 168.0 166.3 - 1.7 - 12.4 - 6.7 c

CH30 Methoxy ( - ) - 36.0 - 37.9 - 1.9 -3.7 -2.5 e

C2HsO Ethoxy ( - ) - 47.5 - 44.8 2.8 2.2 2.0 e

CaHsO Phenoxy ( - ) -40.5 -44.1 - 3.6 - 1.7 - 0.5 c

CHO 2 HCOO ( - ) - 106.6 - 110.9 -4.3 5.0 -2.8 e

C2H30 2 CH3COO (-) - 122.5 - 119.7 2.8 12.5 7.1 e
CH4N CH2-NH2 ( + ) 178.0 185.3 7.3 8.8 -1.7 k
CHiN CH3NH ( - ) 30.5 21.7 -8.8 -7.0 2.6 e
C2H6N MezN ( - ) 24.7 7.8 -16.9 -16.2 -2.3 e
CH2F Fluoromethyl ( + ) 200.3 200.3 0.0 - 17.5 - 19.9 1

CH3F Trifluoromethane ( + ) 233.3 228.2 -5.1 -10.1 -29.3 1

C2H4F CH3CHF ( + ) 166.0 172.9 6.9 - 1.3 - 8.9 m
CHFz Difluoromethyl ( + ) 142.4 145.5 3.1 -10.0 -20.5 1
CHzF 2 Difluoromethane ( + ) 185.2 180.4 -4.8 -6.8 -33.6 1
CaH3F 2 CH3CF 2 ( + ) 107.0 122.2 15.2 9.6 - 1.7 m
CF 3 Trifluoromethyl ( - ) - 163.4 - 178.8 - 15.4 - 15.4 - 15.4 n

CF 3 Trifluoromethyl ( + ) 99.3 99.6 0.3 1.6 - 17.2 1

CHF 3 Trifluoromethane ( + ) 151.9 149.4 - 2.5 6.8 - 30.7 1
CzH2F 3 CFaCH 2 ( + ) 114.0 I22.3 8.3 7.2 0.4 m
C2H2F 3 CH2F.CF 2 ( + ) 81.0 92.7 11.7 1.4 - 11.9 m
 61

References to Table 3

a Bowers, M.T., Gas Phase Ion Chemistry, Vol. 2, Academic Press, New York, NY, 1979.
b Lossing, F.P., Can. J. Chem., 49 (1971) 357.
c Halim, H., Heinrich, N., Koch, W., Schmidt, J. and Frenking, G., J. Comp. Chem., 7 (1986) 93 (and references the-
rein).
d Traeger, J.C. and McLoughlin, R.G., J. Am. Chem. Soc., 103 (1981) 3647.
e Bartmess, J.E. and McIver, R.T., Gas Phase Ion Chemistry, Vol. II, Academic Press, New York, NY, 1979.
f Harris, J.M., Shafer, S.G. and Worley, S.D., J. Comput. Chem., 3 (1982) 208.
g Lossing, F.P. and Maccoll, A., Can. J. Chem., 54 (1976) 990.
h Abboud, J.L., Hehre, W.J. and Taft, R.W., J. Am. Chem. Soc., 94 (1972) 6072.
i Vincow, G., Dauben, H.J, Hunter, F.R. and Volland, W.V., J. Am. Chem. Soc., 91 (1969) 2823.
j Chase, M.W., Curnutt, J.L., Hu, A.T., Prophet, H., Syverud, A.N. and Walker, L.C., J. Phys. Chem. Ref. Data, 3
(1974) 311.
k Franklin, J.L., Dillard, J.G., Rosenstock, H.M., Herron, J.Y., Draxl, K. and Field, F.H., Natl. Stand. Ref. Data Ser.,
Natl. Bur. Stand., No. 26 (1969).
I Blint, R.J., McMahon, T.B. and Beauchamp, J.L., J. Am. Chem. Soc., 96 (1974) 1269.
m Ridge, D.P., J. Am. Chem. Soc., 97 (1975) 5670.
n Harland, P.W. and Franklin, J.L., J. Chem. Phys., 61 (1974) 1621.
62

TABLE 4
COMPARISON OF E X P E R I M E N T A L A N D C A L C U L A T E D HEATS O F FORMATION FOR NORMAL
VALENT INORGANIC MOLECULES

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

H2 Hydrogen 0.0 - 13.4 - 13.4 0.7 - 5.2 a

HO Hydroxyl radical 9.5 3.0 - 6.5 - 9.0 - 8.6 b

H20 Water - 57.8 53.4 4.4 - 3.1 - 1.4 c

0 2 Oxygen (singlet) 22.0 18.4 -3.6 -9.9 -21.3 d
O2 Oxygen (triplet) 0.0 - 4.2 -4.2 -15.3 -27.0 a
H202 Hydrogen peroxide - 32.5 - 40.8 -8.3 -5.7 -2.8 c
0 3 Ozone 34.1 51.1 17.0 14.4 3.7 c
H2N Amidogen 45.5 35.4 -10.1 -8.5 -7.1 c
H3N Ammonia - 11.0 - 3.1 7.9 4.6 3.7 c
NO Nitrogen oxide 21.6 14.8 -6.8 -21.7 -20.4 c
NO2 Nitrogen dioxide 7.9 - 1.0 - 8.9 - 12.5 - 22.9 c
HNO 2 Nitrous acid, trans - 18.8 - 14.9 3.9 -21.9 -20.6 c
NO 3 Nitrate radical 17.0 22.9 5.9 27.9 16.2 c
HNO3 Nitric acid - 32.1 - 38.0 -5.9 14.6 -5.4 c
Nz Nitrogen 0.0 17.6 17.6 8.3 11.2 a
H2N 2 Diazene 36.0 37.8 1.8 -4.1 -4.5 e
H,,N z Hydrazine 22.8 20.6 -2.1 -8.6 -9.1 c
NzO Nitrous oxide 19.6 25.4 5.8 11.4 8.9 c
N20 3 Dinitrogen trioxide 19.8 23.7 3.9 -6.1 2.1 c
N20 4 Dinitrogen tetroxide 2.2 8.3 6.2 27.8 22.9 c
NzO s Dinitrogen pentoxide 2.7 - 19.0 -21.7 31.5 3.0 c
N3 Azide 99.0 106.0 7.0 3.4 8.4 b
HN a Hydrazoic acid 70.3 75.3 5.0 2.8 5.5 f
HS Hydrogen sulfide 33.3 38.2 4.9 4.0 c
H2S Hydrogen sulfide - 4.9 - 0.9 4.0 8.7 c
SO Sulfur monoxide (triplet) 1.2 - 13.6 - 14.8 3.0 c

$2 Sulfur dimer 30.8 28.7 -2.1 4.0 c

HzS 2 Hydrogen disulfide 3.7 8.6 4.9 2.8 f
S4 Sulfur tetramer 34.8 55.2 20.4 11.0 c
H2S3 Hydrogen trisulfide 7.3 26.4 19.1 1.1 f
H2S4 Hydrogen tetrasulfide 10.6 - 0.3 - 10.9 0.0 f

H2S 5 Hydrogen pentasulfide 13.8 2.0 - 11.9 66.2 f

Ss Ss 24.0 18.2 -5.8 -0.7 c
HF Hydrogen fluoride - 65.1 - 62.8 2.4 5.4 -9.1 c
OF FO 26,1 21,2 -4.9 -4.4 -3.5 g
HOF Hypofluorous acid - 23.5 - 29.2 -5.7 4.9 0.9 h
NOF Nitrosyl fluoride - 15.7 - 3.3 12.4 -9.1 -10.8 c
SF SF -4.1 -11.6 -7.5 4.8 c
F2 Fluorine 0.0 - 21.7 - 21.7 7.3 - 22.5 a
OF2 Difluorine oxide 5.9 - 4.8 - 10.7 12.3 4.6 c

NF2 NF2. 10.I 11.9 1.8 - 24.9 - 16.5 c

NzF z cis-Difluorodiazene 16.4 28.0 11.6 - 18.6 4.4 c
N2F a trans-Difluorodiazene 19.4 29.2 9.8 - 17.0 11.9 c
SF2 Sulfur difluoride - 70.9 - 91.9 -21.0 18.0 c
 63

T A B L E 4 (continued)

Empirical Chemical H e a t of f o r m a t i o n Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

$2F2 FSSF - 80.4 - 73.8 6.6 39.1 c

$2F2 SSF2 -95.9 -56.1 39.8 111.4 c
NF3 N i t r o g e n trifluoride -31.6 --24,4 7.2 -2.6 - 8.4 c
H4F4 H y d r o g e n fluoride t e t r a m e r -282.9 -280.2 2.7 38.6 - 30.2 c
N2F4 Tetrafluorohydrazine -2,0 --0.5 1.5 -- I7.7 7.8 c
HCI H y d r o g e n chloride -22.1 -20.5 1.6 6.8 -2.5 c
HOCI H y p o c h l o r o u s acid - 17,8 - 34.3 - 16.5 2.1 -4.0 c
NOC1 N i t r o s y l chloride 12.4 4.5 - 7.9 - 16.6 -7.7 c
SC1 SCI 41.8 28.6 -13.2 -25.3 i
FCI C h l o r i n e fluoride -12.1 -21.7 -9.6 20.3 1.6 c
F3C1 C h l o r i n e trifluoride --38.0 -22.1 15.9 116.7 58.2 c
FsC1 C h l o r i n e pentafluoride - 54.0 - 54.0 0.0 258.8 144.5 c
C12 Chlorine 0.0 - 11.6 - 11.6 - 10.7 - 14.2 a

OC12 Chlorine monoxide 25.0 - 16.2 -41.2 6.3 -5.5 c

SC12 Sulfur dichloride -4.2 - 10.9 - 6.7 - 19.7 c
$2C12 CISSC1 -4.0 -7,7 - 3.7 - 16.8 c
HBr Hydrogen bromide -8.7 5.3 14.0 12.4 -1.8 c
HOBr H y p o b r o m o u s acid -- 20.0 -- 33,9 -4.7
- 13.9 - 2 . 7
J
NOBr Nitrosyl bromide 19.6 6.6 - 13.0 - 17.8 1.7 c
SBr SBr 56.1 48.3 -7.8 -26.7 i

FBr B r o m i n e fluoride -14.0 -21.3 -7.3 8.2 6.8 c

C1Br B r o m i n e chloride 3.5 -3.2 - 6.7 - 13.0 - 14.1 c
Br2 Bromine 7.4 4.9 -2.5 -9.1 - 12.7 c
SBrz Sulfur d i b r o m i d e 48.0 24.9 -23.1 -48.2 i

SzBr2 S2Br2 25.1 21.8 - 3.3 - 23.6 i

HI H y d r o g e n iodide 6.3 28.8 22.5 9.4 1.6 c

NOI N i t r o s y l iodide 26.8 18.2 -8.6 -5.9 5.5 c
SI SI 73.1 58.0 -15.1 -26.4 i

FI I o d i n e fluoride -- 22.6 - 8.0 14.6 13.4 13.5 c

CII I o d i n e chloride 4.6 10.8 6,2 - 11,4 -9.2 k
BrI Iodine b r o m i d e 9.8 15,6 5.9 -2.5 -3.8 c
I2 Iodine 14.9 20,7 5.8 6.3 4.9 c
SI2 Sulfur diiodide 81.9 51.6 - 30.3 - 52.4 i

$212 $212 59.0 46.3 -- 12.7 - 26.7 i

HA1 A1H 62.0 70.1 8.1 - 16.0 c

AlO A10 16.0 -7.8 - 23.9 - 17.8 c
HAlO A1 - O - H - 43.0 - 33.4 9.6 - 18.1 c
HAlO H-AI=O - 8.0 2.2 10.2 - 13.7 c
A102 A10 2 - 20.6 - 37.4 - 16.8 6.2 c
HAlO 2 A102H - 110.0 -105.1 4,9 16.2 c
AIN A l u m i n u m nitride 125.0 75.5 -49.5 15.9 c
AIF A l u m i n u m fluoride -63.5 -50.1 13.4 -20.1 c
AtOF AIFO -- 139.0 -- 124.7 I4.3 25.4 c
AIF 2 A1F 2 -- 166.0 - 162.7 3.3 --5.1 c
A1OF 2 A1F20 -- 265.0 -- 208.5 56.5 62.5 c
A1F3 A l u m i n u m trifluoride --289.0 --291.5 -2.5 -2.3 c
64

TABLE 4 (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

A1C1 Aluminum chloride - 12.3 - 5.5 6.8 - 15.5 C

A1OC1 A1C10 - - 83,2 - 72.4 10.8 14.5 C

A1FC1 A1C1F -117.0 -116.0 1.0 - 7.7 C

A1F2CI A1C1F 2 -238.8 -234.6 4.2 -2.8 C

A1C12 Aluminum dichloride - 67.0 - 68.2 - 1.2 - 7.6 C

AIFC12 A1C12F - 189.0 - 178.0 11.0 -2.4 C

AIC13 Aluminum trichloride - 139.7 - 122.1 17.6 -0.6 C

A1Br 3 Aluminum tribromide -98.1 -85.8 12.3 37.8 C

AII AII 16.2 49.3 33.I 15.0 C

AII 3 Aluminum triiodide - 46.2 - 39.9 6.3 57.8 C

Alz A12 116.4 79.6 - 36.8 14.9 C

A120 A120 -34.7 -28.6 6.1 -37.0 C

A1202 A1202 - 94.3 - 87.6 6.8 - 13,5 C

AlzF6 AIzF 6 -629.5 -631.4 - 1.9 - 2 . 1 C

AlzC16 AlzCI 6 -309.7 -311.2 -- 1.5 14.4 C

AI2Br6 A12Br6 - 224.0 - 224.9 -0.9 91.4 C

A1216 A12I 6 - 117.0 - 117.4 -0.4 124.4 C

HSi Sill 86.3 94.6 8.3 3.9 3.5 1

HeSi Silylene (singlet) 61.1 72.8 11.7 3.2 6.7 m

H2Si Silylene (triplet) 6.5 -2.9 -9.4 -2.7 - 30.7 n

H3Si Silyt 46A 42.9 --3.5 --9.5 -20.1 O

H4Si Silane 8.2 12.5 4.3 - 7.0 -4.1 C

SiO Silicon m o n o x i d e --23.9 --26.0 -2,1 1.3 21.8 i

SiO 2 Silicon d i o x i d e - 73.0 -- 88.9 -- 15.9 50,1 5.5 C

SiF Silicon fluoride 1.7 - 20.9 - 22.6 - 30.5 - 27.8 1

HaSiF Fluorosilane - 90.0 - 77.4 12.6 --6.4 4.2 P

SiOF SiOF - 136.1 -- 107.5 28.6 56.1 30.1 i
SiF 2 Silicon d i f t u o r i d e - 141.2 - 154.9 -13,7 --23.7 - 13.4 q
HzSiF2 Difluorosilane - 189.0 - 175.2 13.8 --3.5 7.5 q
SiOF/ SiOFz - 231.0 - 229.4 1.6 42.4 10.3 C

SiF3 Triftuorosilyl - 245.0 - 260.3 - 15.3 - 3.9 - 22.8 O

HSiF~ Trifluorosilane -287.0 -280.2 6.8 1.9 6.1 C

SiF 4 Silicon t e t r a f l u o r i d e - 386.0 -- 390.6 -4.6 15.6 4.0 q

SiCI Silicon c h l o r i d e 45.3 29.8 - 15.5 -- 15.6 - 18.0 1
H3SiCI Chlorosilane -32.4 -27.6 4.8 - 11.5 -5.3 q
SiOCI SiOC1 - 86.7 - 59.0 27.7 44.2 28.6 i
SIC12 Silicon d i c h l o r i d e - 40.6 - 49.4 -- 8.8 - 5.7 -6.1 q
H2SiC12 Dichlorosilane - 75.3 - 69.5 5.8 -8.2 -5.4 q
SiOC12 SiOClz - 167.7 - 121.2 46.5 73.6 49.4 i
SiCls Trichlorosilyl - 76.0 - 94.8 - 18.8 - 13.5 - 40.0 O

HSiCI 3 Trichlorosilane -119.3 -112.7 6.6 1.4 -5,8 q

SIC14 Silicon t e t r a c h l o r i d e - 158.4 - 156.4 2.0 10.8 - 12.5 C

SiBr Silicon b r o m i d e 50.0 41.0 -9.0 7.8 - 3.0 1

H3SiBr Bromosilane - 15.3 - 16.0 -0.7 -2.7 -5.8 r

SiOBr SiOBr -71.4 -47.1 24.3 42.8 31.1 i

SiBr 2 Silicon d i b r o m i d e -9.6 -27.4 - 17,8 20.8 3.6 q
 65

TABLE 4 (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

H2SiBr2 Dibromosilane -43.2 -47.2 -4.0 11.2 -2.6

SiOBr z SiOBr2 -137.4 -94.1 43.3 86.0 60.0
SiBr3 Silicon tribromide -56.1 -60.8 -4.7 41.1 -9.6
SiBr 4 Silicon tetrabromide - 99.3 -- 107.9 - 8.6 48.9 4.8
SiI Silicon iodide 76.4 71.0 -5.4 15.5 4.1
H3SiI Iodosilane -0.5 0.9 1.4 11.5 5.4
SiOI SiOI - 53.3 - 47.9 5.4 37.7 43.3
SiI/ Silicon diiodide 22.0 24.0 2.0 60.1 35.5
H2SiIz Diiodosilane -9.1 -12.4 -3.3 32.1 14.1
SiOIz SiOI/ - 99.4 - 50.2 49.2 89.7 80.7
SiI3 Silicon triiodide 0.5 4.7 4.2 44.9 11.9
HSiI3 Triiodosilane - 17.8 - 9.0 8.8 51.9 21.7
SiI,, Silicon tetraiodide - 26.4 - 14.2 12.2 68.8 28.0
Si/ Silicon dimer 140.9 135.7 - 5.2 74.4 - 1.4
H6Si2 Disilane 17.1 17.9 0.8 5.3 -0.9
Si2C16 Hexachlorodisilane - 243.5 - 229.7 13.8 30.8 1.9
Si2Br6 Hexabromodisilane - 182.8 - I64.7 18.1 116.9 38.9
Si3 Silicon trimer 152.2 152.8 0.6 15.0 32.7
HsSi3 Si3Ha 28.9 21.7 -7.2 2.9 -4.3
HP Phosphinidene 60.6 73.4 12.8 27.9 15.1
H3P Phosphine 1.3 0.2 - 1.1 2.6 8.9
H2P Phosphino 30.1 29.3 -0.8 1.3 4.1
PO Phosphorus oxide - 2.9 - 19.5 - 16.6 - 18.1 - 13.6
NP Phosphorus nitride 25.0 32.9 7.9 8.9 7.5
PF Phosphorus fluoride - 20.8 - 20,4 0.4 10.9 - 15.3
POF POF -111.8 -124.0 - 12.2 - 2.4 0.5
PF z Phosphorus difluoride - 119.0 - 144.4 -25.4 - 19.0 - 14.2
PF3 Phosphorus trifluoride -229.1 -252.2 -23.1 -0.2 0.2
PCI Phosphorus chloride 25.6 28.9 3.3 10.8 -4.3
POC1 POC1 - 64.7 - 76.4 - 11.7 - 10.1 - 0 . 4

PC1/ Phosphorus dichloride -21.3 -40.1 -18.8 -28.1 -26.3

PC13 Phosphorus trichloride - 69.0 - 88.5 - 19.5 - 27.4 - 20.0
PBr Phosphorus bromide 43.0 34.9 -8.1 -13.6 -15.4
POBr POBr - 50.2 - 59.9 -9.7 -3.7 10.6
PBr 2 Phosphorus dibromide 6.7 0.4 -6.3 --16.0 -11.1
PBr 3 Phosphorus tribromide - 34.9 - 28.2 6.7 - 3.2 11.6
PI PI (triplet) 54.6 51.0 -3.6 -3.1 -2.5
POI POI -33.4 -43.6 - 10.2 8.8 17.7
PI2 Phosphorus diiodide 41.3 36.6 -4.7 -4.8 -6.3
POI2 POI2 -40.1 -35.2 4.9 19.5 33.0
PI 3 Phosphorus triiodide 25.1 31.3 6.2 -0.3 -0.7
P2 Phosphorus dimer 42.8 32.0 - 10.8 - 1.7 - 18.2
H4P2 P2H4 5.0 - 3.7 -8.7 -7.9 1.6
P4 Phosphorus tetramer 31.1 42.9 11.8 5.1 19.3
P406 Phosphorus trioxide -529.2 -511.0 18.2 8.1
66

References to Table 4

a Element in its standard state.

b Chase, M,W., Curnutt, J.L., Hu, A.T., Prophet, H., Syverud, A.N. and Walker, L.C., J. Phys. Chem. Ref. Data, 3
(1974) 311.
c Chase, M.W., Davies, C.A., Downey, J.R., Frurip, D.R., McDonald, R.A. and Syverud, A.N., JANAF Thermochem-
ical Tables, 3rd ed., J. Phys. Chem. Ref. Data 14, Suppl. 1 (1985).
d Herzberg, G., Molecular Spectra and Molecular Structure, Vol. I. (Spectra of Diatomic Molecules), 2nd ed., Van Nos-
trand, New York, NY, 1950.
e Willis, C., Lossing, F.P. and Back, R.A., Can. J. Chem. 54 (1967) 1. Note:
Foner, S.N. and Hudson, R.L., J. Chem. Phys., 68 (1978) 3162 give the heat of formation of diazine as 50.7 +2 kcal/
mol.
f Wagman, D.D., Evans, W.H., Parker, V.B., Hawlow, T., Bailey, S.M. and Schumm, R.H., Natl. Bur. Stand. (U.S.),
Tech. Note, No 270-3 (1968) and Errata in NBS Technical Note 270-8 (1981).
g Berkowitz, J., Dehmer, P.M. and Chupka, W.A., J. Chem. Phys., 59 (1973) 925.
h Pople, J.A. and Curtiss, L.A., J. Chem. Phys. 90 (1989) 2833. Ref. gives - 19.9 kcal/mol for calculated heat of forma-
tion of HOF at OK.
i Ditter, G. and Niemann, U., Philips J. Res., 37 (1982) 1.
j De Kock, R.L., Higginson, B.R., Lloyd, D.R., Breeze, A., Cruickshank, D.WJ. and Armstrong, D.R., Mol. Phys.,
24 (1972) 1059.
k Stull, D.R. and Prophet, H., Natl. Stand. Ref. Data Ser. (U.S., Natl. Bur. Stand.) NSRDS-NBS 37, 1971.
1 Wagman, D.D., Evans, W.H., Parker, V.B., Schumm, R.H., Halow, I., Bailey, S.M., Churney, K.L. and Nuttall, R.L.,
J. Phys. Chem. Ref. Data Suppl. 11 (1982) 2.
m Vanderwielen, A.J., Ring, M.A. and O'Neal, H.E., J. Am. Chem. Soc., 97 (1975) 993.
n Heaton, M.M., J. Chem. Phys., 67 (1977) 5396.
o Walsh, R., Acc. Chem. Res., 14 (1981) 246.
p Schlegel, H.B., J. Phys. Chem., 88 (t984) 6254.
q Walsh, R., J. Chem. Soc. Faraday I, 79 (1983) 2233.
r Farber, M. and Srivastava, R.D., Chem. Phys. Lett., 60 (1979) 216.
s Potzinger, P., Ritter, A. and Krause, J., Z. Naturforsch., Teil A, 30 (1975) 347.
 67

TABLE 5
COMPARISON OF EXPERIMENTAL AND CALCULATED HEATS OF FORMATION FOR HYPERVALENT
MOLECULES

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

C2H6SO D i m e t h y l sulfoxide - 36.1 - 38.8 - 2.7 40.0 a

C4HloSO D i e t h y l sulfoxide - 49,1 - 46.6 2.5 41.5 a
SO 2 Sulfur d i o x i d e - 71.0 - 50.8 20.2 75,4 b
C2H6SO2 D i m e t h y l sulfone - 89.1 - 76.3 12.8 142.8 a
C4H~0SO 2 D i e t h y l sulfone - 102.5 - 80.8 21.7 143,1 a
SO3 Sulfur t r i o x i d e -94.6 - 104.8 - 10.2 153.1 b
C2H6SO 3 D i m e t h y l sulfite - 115.5 - 130.0 - 14.5 50.4 a
H/SO4 Sulfuric acid - 175,7 - 181.4 - 5.8 172.2 b
C2H6SO 4 D i m e t h y l sulfate - 164.1 - 172.1 - 8.0 158.6 a
O2F Fluorine dioxide 3.0 12.9 9.9 21.1 6.0 b
NOzF F l u o r i n e nitrite - 26.0 - 25.6 0.4 26.7 4.7 b
NO3F Fluorine nitrate 2.5 - 6.1 - 8.6 25.5 11.2 b
SOF SOF -63.3 -74.4 -11.1 35.0 c
SO2F SOzF - 113.2 - 102.5 10.7 127.2 c
SOF2 T h i o n y l fluoride - 130.0 - 138.2 - 8.2 84.3 b
SO2F 2 Sulfuryl fluoride - 181.3 - 184.3 - 3.0 203.3 b
NOF3 E3NO - 39.0 - 26.6 12.4 61.8 24.4 b
SF 3 Sulfur trifluoride - 130.0 - 134.3 -4.3 89.9 c
SOF 3 SOF 3 - 185.1 - 176.5 8,6 180.2 c
SF4 Sulfur t e t r a f l u o r i d e - 182.4 - 185.3 - 2.9 135.9 b
SOF4 SOF4 - 235,5 - 236.3 - 0.8 269.1 c
SF 5 Sulfur p e n t a f l u o r i d e - 217.1 - 232.5 - 15.4 208.2 b
SF 5 Sulfur p e n t a f l u o r i d e ( - ) - 291.0 - 303.0 - 12.0 159.4 d
SF6 Sulfur h e x a f l u o r i d e - 291.4 - 304.6 - 13.2 320.7 b
O2C1 Chlorine dioxide 25.0 1.5 - 23.5 111.0 80.9 b
NO2C1 Nitryl chloride 2.9 - 13.0 - 15.9 14.4 11.7 b
SOC1 SOC1 - 17.4 - 31.1 - 13.7 1.5 c
SO2C1 SO2C1 - 66.~ - 57.3 9.1 92.3 c
O3FC1 P e r c h l o r y l fluoride - 5.1 14.6 19.7 328.4 251.6 b
SOC12 Thionyl chloride - 50.8 - 47.6 3.2 28.6 e
SO2C12 Sulfuryl c h l o r i d e - 86.2 - 62.3 23.9 79.9 b
SC13 Sulfur t r i c h l o r i d e 8.8 - 19.l -27.9 -41.3 c
SOCI 3 SOC13 - 47.5 - 46.5 1.0 49.9 c
SCI 4 Sulfur t e t r a c h l o r i d e - 0.7 - 19.8 - 19.1 - 32.6 c
SOC14 SOC14 - 55.7 - 59.3 - 3.6 22.7 c
SC1 s Sulfur p e n t a c h l o r i d e - 8.6 9.2 17.8 4.3 c
SCI 6 Sulfur h e x a c h l o r i d e - 19.8 10.3 30.1 138.1 c
OBr BrO 30.1 20.8 - 9.3 5.3 5.6 f
SOBr SOBr -4.3 - 16.2 - 11.9 2.2 c
SO2Br SO2Br - 52.8 - 42.9 9.9 89.5 c
F3Br B r o m i n e trifluoride - 61.1 - 47.1 14.0 84.0 82.6 b
FsBr Bromine pentafluoride - 102.5 - 75.8 26.7 207.4 183.7 b
SOBr z Thionyl bromide - 11.5 - 18.6 - 7.1 16.1 c
SO2Br 2 Sulfuryl b r o m i d e - 59.5 - 46.2 13.3 127.9 c
68

T A B L E 5 (continued)

Empirical Chemical H e a t of f o r m a t i o n Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM 1

SBr3 Sulfur t r i b r o m i d e 50.2 16.4 -33.8 -51.3

SOBr3 SOBr 3 - 8.6 - 10.0 - 1.4 41.4
SBr4 Sulfur t e t r a b r o m i d e 53.0 18.8 - 34.2 - 42.9
SOBr~ SOBr 4 - 3.3 - 20.7 - 17.4 26.6
SBr5 Sulfur p e n t a b r o m i d e 55.9 44.0 - 11.9 23.5
SBr 6 Sulfur h e x a b r o m i d e 58.8 78.1 19.3 107.6
OI IO 41.8 31.0 - 10.9 4.8 - 4 . 9

SOI SOI 12.7 -0.4 - 13.1 7.0

SOzI SO2I - 34.9 - 32.0 2.9 64.5
FsI Iodine pentafluoride - 200.8 - 202.9 -2.1 298.8 267.9
FTI Iodine heptafluoride - 229.7 - 225.4 4.3 334.9 274.1
SOI2 T h i o n y l iodide 21.5 11.1 - 10.4 26.6
SO212 Sulfuryl iodide - 26.0 - 30.0 --4.0 51.7
SI3 Sulfur triiodide 100.3 54.7 - 45.6 - 60.0
SOI 3 SOI3 40.4 23.8 - 16.6 1.7
SI4 Sulfur t e t r a i o d i d e 120.2 83.8 -36.4 --38.0
SOI 4 SOI4. 60.0 27.6 -- 32.4 -- 17.0
SIs Sulfur p e n t a i o d i d e 130.9 129.6 - 1.3 - 1.0

SI6 Sulfur h e x a i o d i d e 158.9 167.0 8.1 48.7

C4H12SiF SiMe4F ( - ) C 3 v s y m m e t r y - 147.5 - 122.4 25.1 33.5 23.5
SiFs SiF 5 ( - ) - 507.1 - 504.4 2.7 17.7 3.6
SiF,,C1 SiF4CI ( - ) -- 465.3 - 465.2 0.1 17.5 8.0
SiC15 SiC15 ( - ) - 237.2 - 254.6 - 17.4 - 19.3 - 30.4
C3H9PO T r i m e t h y l p h o s p h i n e oxide - 102.2 - 82.7 19.5 59.3 0.7
PO2 P h o s p h o r u s dioxide - 71.0 - 76.7 - 5.7 24.0 13.2
CHsPO3 M e t h y l p h o s p h o n i c acid - 240.5 - 213.3 27.2 25.2 31.0
C2H-TPO 3 E t h y l p h o s p h o n i c acid - 239.4 - 218.5 20.9 19.9 26.0
C3H9PO 3 Trimethyl phosphite - 168.3 - 192.2 -23.9 -38.0 - 17.3
C6H15PO3 Triethyl p h o s p h i t e 195.9 - 208.5 --12.6 --27.5 -ll.2
C6H15PO4 Triethyl p h o s p h a t e - 284.5 - 252.9 31.6 75.0 8.3
C2H6PO2F Methyl methylphosphonofluoridate - 197.3 - 207.6 - 10.3 51.4 - 20.9
C3HsPO2F Ethyl m e t h y l p h o s p h o n o f t u o r i d a t e - 205.8 - 212.3 - 6.5 53.9 - 18.7
C4HloPO2F n-Propylmethylphosphonofluoridate -210.2 -217.4 -7.2 53.9 -21.0
C4HloPO2F i-Propylmethylphosphonofluoridate -214.6 -216.7 -2.1 60.8 -15.1
CsH12POzF s-Butyl m e t h y l p h o s p h o n o f l u o r i d a t e - 220.1 - 221.2 - 1.1 63.2 - 15.5
CsH12PO2F i-Propyl ethylphosphonoftuoridate - 219.8 - 217.5 2.2 59.4 - 12.1
CsH12PO2F n-Butyl m e t h y l p h o s p h o n o f l u o r i d a t e - 215.1 - 222.9 - 7.8 53.8 - 22.9
C6H14POzF N e o p e n t y l m e t h y l p h o s p h o n o f l u o r i d a t e - 224.2 - 226.6 -2.4 70.4 -- 13.7
POF 2 POF 2 - 213.6 - 188.9 24.7 61.7 39.4
CH3POF 2 Methylphosphonodifluoride - 233.2 - 225.3 7.9 86.1 5.6
POF 3 P h o s p h o r u s oxyfluoride - 289.5 - 297.7 - 8.2 90.0 - 3.3
PF4 P h o s p h o r u s tetrafluoride ( - ) - 325.0 - 332.1 -7.1 22.8 26.2
PF4 P h o s p h o r u s tetrafluoride - 287.9 - 303.9 -- 16.0 53.3 6.8
PF5 Phosphorus pentafluoride - 381.1 - 386.9 - 5.8 132.3 1.8
PF 6 P h o s p h o r u s hexafluoride ( - ) - 522.0 - 508.5 13.5 152.4 17.6
POC12 POC12 - 109.9 - 94.8 15.1 33.9 24.8
 69

TABLE 5 (continued)

Empirical Chemical Heat of formation Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM 1

CH3POC12 Methylphosphonodichloride - 124.1 - 128.9 -4.8 48.7 9.3 g

POCI 3 Phosphorus oxychloride - 132.8 - 140.2 -7.4 53.2 14.9 h
PSC13 Phosphorus thiochloride - 91.0 - 60.2 30.8 62.6 b
PCI~ Phosphorus tetrachloride - 80.5 - 102.8 - 22.3 - 22.7 - 29.5 c
PC15 Phosphorus pentachloride - 89.6 - 111.6 - 22.0 47.7 17.8 h
POBr z POBr 2 - 78.3 - 50.8 27.5 26.4 56.5 c
POBr3 Phosphorus oxybromide - 97.0 - 80.2 16.8 68.3 68.0 b
PSBr 3 Phosphorus thiobromide - 67.2 - 7.8 59.4 85.7 b
PBr4 Phosphorus tetrabromide - 17.4 - 17.4 0.0 - 11.9 - 8.5 c
PBr 5 Phosphorus pentabromide - 11.0 - 27.1 - 16.1 42.8 103.9 c
POI 3 Phosphorus oxyiodide - 39.7 - 8.3 31.4 71.3 75.9 c
PI 4 Phosphorus tetraiodide 60.2 44.7 - 15.5 - 36.1 - 33.7 c
PIs Phosphorus pentaiodide 97.7 88.5 - 9.2 - 16.0 - 42.1 c
P4Olo Phosphorus pentoxide - 694.1 - 712.6 - 18.5 262.4 - 6.8 b

References to Table 5

a Cox, J.O. and Pilcher, G., Thermochemistry of Organic and Organometallic Compounds, Academic Press, New York,
NY 1970.
b Chase, M.W., Davies, C.A., Downey, J.R., Frurip, D.R., McDonald, R.A. and Syverud, A.N., J A N A F Thermochemi-
cal Tables, 3rd ed., J. Phys. Chem. Ref. Data 14, Suppl. 1 (1985).
c Ditter, G. and Niemann, U., Philips J. Res., 37 (1982) 1.
d Larson, J.W. and McMahon, T.B., J. Am. Chem. Soc., 107 (1985) 766.
e Wagman, D.D., Evans, W.H., Parker, V.B., Hawlow, T., Bailey, S.M. and Schumm, R.H., Natl. Bur. Stand. (U.S.),
Tech. Note, No 2 7 0 - 3 (1968) and Errata in NBS Technical Note 2 7 0 - 8 (1981).
f NBS Technical Note 2 7 0 - 3 , Jan. (1968).
g Penski, E.C. and Domalski, E.S., C R D E C - T R - 8 7 0 6 3 (1987).
h Wagman, D.D., Evans, W.H., Parker, V.B., Schumm, R.H., Halow, I., Bailey, S.M., Churney, K.L. and Nuttall, R.
L., J. Phys. Chem. Ref. Data Suppl. I1 (1982) 2.
70

TABLE 6
COMPARISON OF EXPERIMENTAL AND CALCULATED HEATS OF FORMATION FOR INORGANIC
IONS

Empirical Chemical H e a t of f o r m a t i o n Difference Ref.

formula name
Exp. Calc. PM3 MNDO AM1

H Hydrogen ( + ) 365.7 353.6 - 12.1 - 39.0 - 50.8 a

HO Hydroxide(-) -33.2 -17.5 15.7 27.4 19.1 b
H30 Hydronium ( + ) 138.9 159.1 20.2 - 4.7 4.6 c
H4N Ammonium ( + ) 155.0 153.4 - 1.6 9.6 - 4.4 d
NO NO(+) 237.0 238.2 1.2 -6.4 --8.8 c
NO 2 Nitrogen dioxide ( + ) 233.0 208.4 - 24.6 7.6 - 11.9 e
NO 3 Nitrate anion - 74.7 -- 93.3 - 18.6 7.7 - 14.2 f
S S(- ) 16.8 20.7 3.9 29.8 c
HS HS(-) -17.1 -15.9 1.2 24.0 g
F Fluoride ( - ) - 61.0 - 31.2 29.8 43.9 64.4 c
NF 2 NF 2 (-) -29.5 - 31.0 - 1.5 - 14.1 4.3 h
CI Chloride ( - ) - 55.9 - 51.2 4.7 1.2 18.2 c
HFCI H y d r o g e n c h l o r i d e fluoride( - ) - 142.0 - 137.0 5.0 15,8 15.9 i
HCl 2 Hydrogen dichloride ( - ) - 142.0 - 1 l 1.0 31.0 47.2 47.6 i
Br Bromide ( - ) - 52.3 - 56.2 - 3.9 14.8 31.9 c
HBr HBr (+) 261.1 274.7 13.6 16.8 - 12.3 j
C1Br Bromine chloride ( + ) 261.0 247.6 - 13.4 4.8 - 14.6 f
Br2 Bromine ( + ) 253.5 263.0 9.5 12.5 - 7.7 j
I Iodide ( - ) - 46.5 - 64.6 - 18.1 40.1 44.3 c
A1 AI(+) 218.1 279.8 61.7 -24.2 c
AIO AIO ( + ) 237.3 202.4 - 34.9 - 14.2 c
AIO A10 ( - ) - 64.4 - 50.9 13.5 8.6 c
HAlO AIOH ( + ) 130.0 129.3 - 0.7 59.2 c
HAlO A1OH (-) -55.0 -105.7 -50.7 13.9 c
A10 2 A102 ( - ) -116.0 -117.1 -1.1 25.5 c
A1F A1F ( + ) 165.4 160.4 - 5.0 - 50.6 c
AIF 2 A1F 2 ( + ) 22.0 29.0 7.0 - 7.8 c
A1F 2 A1F 2 ( - ) -217.0 -229.6 - 12.6 7.4 c
AIOF 2 A1F/O ( - ) - 311.6 - 286.0 25.6 53.7 c
A1F4 A1F4( - ) - 476.0 - 469.2 6.8 35.4 c
AIC1 Aluminum chloride ( + ) 206.0 198.7 - 7.3 - 31.2 c
A1FC1 Aluminum chloride fluoride ( + ) 66.0 74.9 8.9 - 2.5 c
A1C12 A1C12 ( + ) 115.0 124.3 9.3 3.9 c
A1C12 A1C12 ( - ) - 1 l 5.0 - 145.0 - 30.0 - 15.6 c
A120 A120 ( + ) 155.9 168.4 12.6 -46,8 c
A120 z A1201 ( + ) 126.0 107.3 - 18.7 -47,1 c
H3Si Silyl ( - ) 14.0 - 2.8 - 16.8 32.5 - 15.8 k
H3Si Silyl ( + ) 234.1 223.3 - 10.8 - 43.3 - 11.2 k
 71

References to Table 6

a Weast, R.C. (Ed.), CRC Handbook of Chemistry and Physics, 60th ed., CRC Press, Boca Raton, FL, 1980.
b Bartmess, J. E. and McIver, R. T., Gas Phase Ion Chemistry, Vol. II, Academic Press, New York, NY 1979.
c Chase, M.W., Davies, C.A., Downey, J.R., Frurip, D.R., McDonald, R.A. and Syverud, A.N., JANAF Thermochemi-
ca1 Tables, 3rd ed., J. Phys. Chem. Ref. Data 14, Suppl. 1 (1985).
d Kebarle, P., Tamdagni, R., Kirooka, H. and McMahon, T.B., Int. J. Mass Spectrom. Ion Phys., 19 (1976) 71.
e Franklin, J.L., Dillard, J.G., Rosenstock, H.M., Herron, J.T., Draxl, K. and Field, F.H., Natl. Stand. Ref. Data Ser.,
Natl. Bur. Stand., No. 26 (1969).
f Wagman, D.D., Evans, W.H., Parker, V.B., Schumm, R.H., Halow, I., Bailey, S.M., Churney, K. L. and Nuttai1,
R i . , J. Phys. Chem. Ref. Data Suppl. 11 (1982) 2.
h Harland, P.W. and Franklin, J.L., J. Chem. Phys., 61 (1974) 1621.
i Larson, J.W. and McMahon, T.B., J. Am. Chem. Soc., 107 (1985) 766.
j NBS Technical Note 270-3, Jan. (1968).
k Nimlos, M.R. and Ellison, G.B., J. Am. Chem. Soc., 108 (1986) 6522-6529.
72

TABLE 7
STATISTICAL ANALYSIS OF DIFFERENCES BETWEEN EXPERIMENTAL AND CALCULATED HEATS OF
FORMATION

Type of Number of Average difference

compounds compounds
Signed Unsigned Root mean square

PM3 MNDO AM1 PM3 MNDO AM1 PM3 MNDO AM1

Hydrogen 456 6.2 12.8 7.3 1.5 14.9 --0.1 8.3 24.2 10.4
Carbon 459 6.2 12.8 7.9 1.2 3.5 -0.3 8.7 23.1 11.5
Nitrogen 117 6.5 18.8 9.6 0.6 10.0 13.3 8.3 27.0 12.5
Oxygen 234 9.0 32.4 14.4 -0.1 26.8 4.5 12.9 56.4 25.2
Fluorine 133 8.6 44.4 22.7 0.7 37.4 6.2 11.2 84.2 49,5
Silicon 78 I0.1 22.4 14.5 1.9 13.0 3.6 14.2 32.9 20.8
Phosphorus 70 12.4 37.7 19.5 -0.8 25.0 6.4 16.1 55.6 29.1
Sulfur 101 12.0 50.3 23.4 -1.4 36.9 - 10.4 16.2 79.8 34.0
Chlorine 93 9.6 24.6 18.8 --0.1 15.1 2.6 13.4 54.1 37.8
Bromine 67 11.4 27.8 23.4 1.5 16.4 6.2 15.5 46.5 39.9
Iodine 73 10.7 29.2 25.7 --0.3 10.9 4.2 15.8 60.7 53.5
Set of compounds 138 4.4 6.2 5.5 0.0 -1.4 0.7 6.3 9.1 7.3
used in Refs. 2
and 4
Compounds of 276 5.5 11.2 7.5 0.4 3.8 1.4 7.9 18.5 10.5
C, H, N, and O, only
Nitro compounds 29 5.2 39.6 15.7 2.5 38.1 14.5 6.2 44.1 18.5
Organophosphorus-V 15 10.9 53.9 15.5 3.6 50.2 -4.7 14.3 56.7 17.3
compounds
Normal valent 607 7.3 13.1 9.6 0.5 3.1 -0.2 11.2 24.3 14.8
compounds
Hypervalent 106 13.6 75.8 37.7 -0.8 67.2 8.9 17.3 104.5 62.3
All compounds 713 8.2 22.5 13.8 0.3 13.8 1.2 11.6 46.2 27.6
 73

TABLE 8
BARRIERS TO ROTATION AND INVERSION

Molecule Barrier

Exp. PM3 MNDO AM 1

Ethylene 65.0 35.90 38,83 45.01"

Ethylene 65.0 65.31 62.48 65.93 b
Ethane 2.9 1.43 1.01 1.25
Methylamine 2.0 1.10 1.07 1.27
Methanol 1.1 0.90 0.74 1.04
H O - O H (cis) 7.0 7.27 6.90 7.01
H O - O H (trans) 1.1 0,00 0.00 0.10

Formamide - 20 21.00 19.00 17.00 c

Formamide - 20 1.40 3.29 10.11 d

n-Butane (gauche) 0.8 0.53 0.58 0.71

n-Butane (eclipsed) 4- 6 4.00 3.23 3.28
Nitrobenzene 1,64 1.78 3.48 °
Ammonia (inversion) 6 9.98 11.58 4.24

" Barrier calculated with minimal configuration interaction. The transition state for the rotation of the C = C bond in-
volves two electronic states. When both states are used, the 'reaction' has a well-defined transition state.
b Barrier calculated without configuration interaction. Using only one configuration results in an intcrsystem crossing.
The D2d geometry does not correspond to a stationary point.
c Barrier calculated with molecular mechanics correction. The M M correction can be invoked by the keyword M M O K .
d Barrier calculated without molecular mechanics correction. The correction can be omitted by using the keyword
NOMM.
° For PM3 and AM1 the minimum energy configuration of nitrobenzene is with the NO2 coplanar with the aromatic
ring. For M N D O the NO2 is perpendicular to the aromatic ring. This is characteristic of most sp2-hybridized groups
attached to otherwise unsubstituted benzene rings.
74

TABLE 9
INTERMOLECULAR INTERACTION ENERGIES (kcal/mol)

Molecular pair Heat of association

PM3 MNDO AM1

H2 - N2 0.0 0.0 0.0

NH 3 - H2 0.0 0.0 0.0
CO2 - CO2 0.0 0.0 0.0
02 - 02 0.0 0.0 0.0
H2 - C O 2 0.0 0.0 0.0
NH3 - CO2 -0.5 - 0.4 - 1.8
C H 4 -- C H 4 - 0.9 0.0 - 0.4
H20- H2 -0.9 0.0 - 1.1
H20 - CO: - 0.9 - 0.8 - 2.7
NH3 - NH3 -0.9 -0.8 -2.4
CH20 - CH20 - 1.0 - 0.8 - 2.4
H20- NH 3 - 1.1 -0.5 -2.7
H20 - CH4 - 1.2 0.0 - 1.2
CH3OH - H20 - 1.5 - 0.7 - 2.7
CH20 - H20 - 1.6 - 0.7 - 3.7
H20- CH20 - 1.6 - 1.0 -4.0
H20 - CsHsN - 1,7 - 1.1 - 3,1
H20- O3 - 1.8 - 1.1 -6.8
H20- CH3OH - 1.9 -0.9 -4.8
NHzCOOH - NHzCOOH - 1.9 - 3.3 - 9.0
HCOOH - CH 4 - 2.3 0.1 - 0,8
C6H 6 - C6H 6 - 3.8 0.0 -- 0.4
NH2CHO - NH2CHO - 4.7 - 2.1 - 8,1
H=O - HCOOH - 5.3 - 1.0 - 7.4
HCOOH - NH 3 - 5.3 -0.9 -4.1
HCOOH - HCOOH - 8.6 - 1.7 - 6.4
 75

TABLE 10
C O M P A R I S O N OF E X P E R I M E N T A L A N D C A L C U L A T E D M O L E C U L A R G E O M E T R I E S F O R O R G A N I C
MOLECULES

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

CH 2 Methylene, singlet CH 1.110 1,092 - 0.018 -- 0,019 - 0.007 a

HCH 102.4 103.7 1.3 8.7 8,1
CH 2 Methylene, triplet CH 1.029 1.064 0.035 0.024 0.034 a
HCH 144.7 144.7 0.0 4.9 3.6
CH4 Methane CH 1.094 1,087 - 0.007 0,010 0.018 b
C2 Carbon, dimer CC 1.242 1.189 -0.053 -0.073 -0.078 a
C2H2 Acetylene CC 1.203 1.190 -0.013 -0.008 -0.008 b
CH 1.060 1.064 0.004 - 0.009 0.001
CzH 4 Ethylene CC 1.339 1.322 -0.017 -0,004 -0.013 b
CH 1.086 1.086 0.000 0.003 0.012
HCC 121.2 123.1 1.9 2.0 1.5
C2H 6 Ethane CC 1,536 1.504 - 0.032 - 0.0!5 - 0.036 b
CH 1.091 1.098 0.007 0,018 0.026
HCC 110.9 l 11.6 0.7 0.3 - 0.2

C3H 4 Allene CC 1.308 1.297 - 0.011 - 0,002 - 0,010 c

CH 1.087 1.086 - 0.001 0.003 0,013
HCC 120.9 122.3 1.4 2.0 1.4
C3H 4 Cyclopropene C2C3 1.509 1.484 - 0.025 0.003 - 0.020 d
C1C2 1.296 1,314 0.018 0.032 0.022
C1H 1.072 1.073 0.001 - 0.010 - 0.003
HC1C2 149.9 151.5 1.6 1,7 2.0
C3H 4 Propyne C2C1 1.206 1.191 -0.015 -0,009 -0,009 e
CIH 1.056 1,064 0.008 - 0,005 0,004
C3C3 1.459 1.433 - 0.026 - 0,014 - 0.032
C3H 1.105 1.098 - 0.007 0,006 0.016
HCC 111.0 110.7 -0.3 0.0 -0.5
C3H 6 Cyclopropane CC 1.510 1,499 -0.011 0.016 -0.009 f
CH 1.089 1.095 0.006 0.007 0.015
C3H 6 Propene C= C 1.336 1.328 - 0.008 0.004 - 0.005 g
C-C 1.501 1.480 - 0,021 - 0,005 - 0.025
CCC 124.3 123.4 -0.9 2.6 0.0
C3H 1.085 1.098 0.013 0.024 0.033
HC3C2 111.2 112.9 1.7 1.8 0.7
C2H 1.090 1.097 0.007 0.006 0.013
HC2C1 119.0 120.8 1.8 0.3 1.9
HC1 1.091 1.087 - 0.004 - 0.002 0.007
HC1C2 121.5 122.7 1.2 0.8 0.8
C3H8 Propane CC 1.526 1.512 -0.014 0.004 -0.019 g
CCC 112.4 111.7 -0.7 3.0 -0.6
C2H 1.115 1.108 - 0.007 0.000 0.007
HC2C1 109.5 109.9 0.4 - 0.7 0.0
C1H 1.096 1.097 0.001 0.014 0.021
HC 1C2 111.8 I 11.4 -- 0.4 - 1.5 -- 1.4
76

T A B L E 10 ( c o n t i n u e d )

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

C4H2 Diacetylene C1C2 1.205 1.193 -0,012 -0.006 -0.006 h

C2C3 1.376 1.371 - 0,005 - 0.008 - 0,020
CH 1,046 1.065 0.019 0.004 0.014
C4H4 CH2 = C = C = CH2 CH 1.083 1.087 0.004 0.007 0,017 i
C1C2 1.318 1.301 -0.017 -0.007 -0.016
C2C3 1.283 1.267 - 0.016 - 0.013 - 0.017
C4H4 Vinylacetylene C3C4 1,341 1.332 - 0.009 0.004 - 0.005 J
C2C3 1.431 1.414 -0.017 -0.014 -0.026
C2C3C4 123.1 122.5 -0.6 2.3 1.0
C1C2 1.208 1,193 -0.015 -0.010 -0.010
C4H 6 Bicyclobutane C1C2 1.498 1.507 0,009 0,029 0.012 k
C~1C3 1.497 1,481 - 0.016 0.039 - 0.002
C2C3C1C4 121.7 120.0 - 1.7 0.9 0.3
C1H 1.071 1.083 0.012 0.003 0.008
C2H 1.093 1.095 0.002 0.005 0.012
C4H 6 2-Butyne C2C3 1.213 1.193 -0.020 -0.013 -0.015 J
C1C2 1.467 1.432 - 0.035 - 0.023 - 0.042
CH 1.115 1.098 -0.017 -0.004 0.006
HCC 110.7 110.7 0.0 0.3 -0.1
C4H 6 1,3-Butadiene C1C2 1.344 1.331 -0.013 0.000 -0.009 1
C2C3 1.467 1.456 - 0.011 - 0.002 - 0.016
CCC 122.9 122.3 - 0.6 2.8 0.5
C4H8 1-Butene C2C3 1.347 1.328 - 0.019 - 0.006 - 0.016 m
C1C2 1.508 1.489 - 0.019 - 0.003 - 0.024
CCC 123.8 122.6 - 1.2 1.6 - 0.4
C4H8 Cyclobutane CC 1.548 1.542 - 0.006 0.001 - 0.005 n
C1C2C4C3 153.0 180.0 27.0 27.0 27.0
CH 1.105 1.100 - 0.005 0.000 0.005
C4H8 Isobutene C1C2 1.330 1.333 0.003 0.018 0.006 o
C2C3 1.507 1.487 - 0.020 0.002 - 0.023
C1C2C3 122.4 122.1 - 0.3 - 0.5 0.0
C4Hao n-Butane C1C2 1.533 1.512 -0.021 -0.002 -0.026 e
C2C3 1.533 1.521 - 0.012 0.007 - 0.019
CCC 112.8 111.6 - 1.2 2.0 - 1.2
C,Hlo Isobutane CC 1.525 1.520 - 0.005 0.016 - 0.011 p
CsH8 1,4-Pentadiene C1 C = C 1.339 1.328 - 0.011 0.001 - 0.008 q
C - C 1.511 1.489 - 0.022 - 0.005 - 0.027
C-C =C 115.5 123.1 7.6 11.I 8.4
C-C-C 113.1 114.4 1.3 -0.5 1.2
CI-C2-C3-C4 - 116.9 - 127.5 - 10.6 9.6 - 13.8
C2-C3-C4-C5 -4.3 14.2 18.5 110.0 16.2
CsH 8 1,4-Pentadiene C2 C = C 1.339 1.328 - 0.011 0.001 - 0.008 q
C-C 1.511 1.490 -0.021 -0.005 -0.025
C - C = C 115.5 123.1 7.6 11.2 8.3
C - C - C 108.9 110.8 1.9 3.8 2.9
C1-C2-C3-C4 - 122.2 - 129.3 - 7.1 15.1 - 11.9
 77

TABLE 10 (continued)

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

CsH 8 1,4-Pentadiene Cs C= C 1.339 1,328 - 0.011 0.001 - 0,008 q

C- C 1.511 1.490 - 0,021 - 0,005 - 0.025
C-C=C 115.5 123.0 7.5 l l .1 8.3
C-C-C 108.9 111.1 2.2 3.7 3.0
C1-C2-C3-C4 - 128.6 - 132.2 - 3.6 22,1 - 5.3
C5H12 Neopentane CC 1.539 1,527 -0,012 0.015 -0,018 f
CH 1.120 1,098 -0.022 -0.01l -0.004
HCC 110.0 111.3 1.3 1,7 0.3
C6H 6 Benzene CC 1.399 1.391 - 0.008 0.008 - 0,004 r
CH 1.084 1,095 0.011 0.006 0.016
C6H 6 Fulvene C3C4 1.476 1.471 - 0,005 0,000 0.000 s
C2C3 1,355 1.355 0.000 0.011 0.008
C1C2 1.470 1.478 0.008 0.021 0,013
C1C6 1.349 1.331 -0.018 -0.004 -0,017
C6Hlo Cyclohexene C1C2 1.335 1.334 - 0,001 0.011 0.002 t
C2C3 1.504 1.487 - 0,017 0.000 - 0.021
C3C4 1,515 1.52l 0.006 0,026 0.002
C4C5 1.550 1,519 -0,031 -0,011 -0.036
C5C4C2C 1 21.8 27.8 6.0 - 0.8 5.4
C6H12 Cyclohexane CC 1.536 1.521 -0.015 0,002 -0.021 u
CCC 111.4 111.0 -0.4 2.7 -0.1
CCCC 54.9 56.0 9.7 0.0 8.9
CH 1.104 1.107 -0.014 -0.007 0.000
CH' 1.121 1.108 -0.013 -0.007 0.001
CO Carbon monoxide CO 1,128 1,135 0.007 0.035 0.043 v
CH20 Formaldehyde CO 1,208 1.202 - 0,006 0.008 0.019 w
CH 1.116 1.091 -0.025 -0,010 - 0,006
HCO 121.8 12I .8 0.0 1.7 0.4
CH40 Methanol CO 1.425 1,395 -0,030 -0.034 -0.015 x
CH 1,094 1.097 0.003 0,025 0.025
HCO 108.5 112.2 3.7 3.8 2.4
OH 0.945 0.949 0.004 0,002 0,019
COH 107.0 107.5 0.5 4.6 0.2
C2H20 Ketene CO 1.161 1.175 0.014 0,023 0.032 y
CC 1.314 !.308 0,006 0,005 - 0.007
CH 1.083 1.084 0.001 0.002 0.012
HCC 118.7 122.0 3.3 3,0 2.7
C2H60 Dimethyl ether CC 1.410 1.406 - 0.004 - 0,014 0.007 z
COC 111.3 114.1 2.8 8.7 1.6
C3H40 Acrolein C3C2 1.335 1.330 - 0.005 0,008 - 0,001 aa
C2C1 1.478 1.479 0.001 0,007 -0,010
CCC 121.0 123.6 2.6 6.3 2.2
CO 1.208 1.211 0.003 0.016 0.026
OCC 124.0 124.0 0.0 1.5 0.0
78

TABLE 10 (continued)

Empirical Chemical Geometric Exp. Calc, Errors Ref.

formula name variable
PM3 MNDO AM1

C4H40 Furan CO 1.362 1.378 0.016 0.005 0.033 bb

CCO 106.6 106.9 0.3 1.0 0.0
C3C2 1.361 1.373 0,012 0,029 0.019
CCC 110.7 110.2 -0.5 -0.4 -0.6
CO2 Carbon dioxide CO 1.162 1.181 0.019 0.024 0.027 b
CH202 Formic acid C= O 1.202 1.211 0.009 0.025 0.028 cc
C-O 1.343 1.344 0.001 0.011 0.014
OCO 124.9 117.1 -7.8 -4.3 -7.3
OH 0.972 0.953 - 0.019 - 0.023 - 0.001
HOC 106,3 111.6 5,3 9.9 4.3
CH 1.097 1.095 - 0.002 0.008 0.006
HC - O 124,1 130.4 6,3 2.7 6,0
C2H202 trans-Glyoxal CO 1.207 1.207 0,000 0.013 0.022 aa
CC 1.525 1.526 0.001 0.004 - 0.017
CCO 121,2 120.5 -0,7 0,8 -0,2
C6H402 p-Benzoquinone C1C2 1.477 1.487 0.010 0.024 0.002 dd
C2C3 1.322 1.335 0.013 0.027 0.016
CCC 121.1 121.6 0.5 1.0 0.8
CO 1.222 1.217 - 0.005 0.004 0.014
CN Cyanide ( + ) CN 1.290 1.355 0.065 -0.149 -0.164 a
CN Cyanide CN 1.175 1.157 -0.018 -0.022 -0.027 a
CHN Hydrogen cyanide CN 1.154 1.156 0.002 0,006 0.006 ee
CH 1.063 1.070 0.007 - 0.008 0.006
CHsN Methylamine CN 1.474 1.469 - 0.005 - 0.014 - 0.042 ff
NH 1.011 0.999 - 0.012 - 0.003 - 0.011
HNC 112.0 109.8 - 2.2 - 2.2 - 0.7
HNH 105.9 108,7 2.8 - 0.4 3.1
C2H3N Acetonitrile CC 1.458 1.440 - 0.018 - 0.006 - 0.019 gg
CH 1.104 1.098 - 0.006 0.006 0.016
HCC 109.5 110.4 0.9 1.1 0.6
CN 1.157 1.159 0.002 0.005 0.006
C2H3N Methyl isocyanide CN - 1.424 1.433 0.009 0.000 - 0.029 gg
CH 1.101 1.097 - 0.004 0.014 0.024
HCN 109.1 109.7 0.6 1.1 1.0
-CN 1.166 1.181 0.015 0.025 0.015
C3HgN Trimethylamine CN 1.451 1.480 0.029 0.013 - 0.006 hh
CNC 110,9 112.3 1.4 5.1 2.1
C4HsN Pyrrole CN 1.370 1.397 0.027 0.028 0.022 1
CNC 107.7 109.7 2.0 2.0 1.1
C3C2 1.382 1,390 0,008 0.013 0.020
CCC 109.8 107.0 - 2.8 - 2.4 - 1.4
C4C3 1.417 1,390 -0.027 -0,022 -0.015
CHNO Hydrogen isocyanate NH 0.987 0.985 - 0,002 0.011 - 0.002 a
CN 1.207 1.251 0,044 0.042 0.025
CNH 128.1 123.7 -4.4 -7.9 -0.9
CO 1.171 1,181 0,010 0,014 0,031
OCN 180,0 168.7 -11.3 -12.5 -13,3
 79

TABLE 10 (continued)

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

CH3NO Formamide CN 1,376 1.413 0.037 0,033 - 0.009 ii

NH 1.002 0.994 -0.008 -0.002 -0,016
CH 1.102 1.102 0,000 0.006 0,012
CO 1.193 1.217 0,024 0.032 0,050
OCN 123.8 118.5 -5.3 -2.7 -1.9
C2N2 Cyanogen CN 1,154 1,159 0,005 0,008 0,008 b
CC 1.389 1.382 - 0.007 - 0.011 - 0.005
C2H6N 2 Dimethyldiazene NN 1.254 1.228 - 0,026 - 0.032 - 0.030 b
CN 1.474 1,467 - 0.007 0.000 - 0.022
CNN 111.9 119.3 7.4 5.0 7.9
C4H4N 2 Pyrimidine C4C5 1.393 1,396 0.003 0.016 0,015 kk
N3C4 1.350 1,354 0.004 0.003 0.000
N3C4C5 121.2 120.7 -0.5 0.1 1.1
C2N3 1.328 1.357 0.029 0.029 0.033
C2H 1.082 1,098 0,016 0.016 0.028
C5H 1.087 1.094 0.007 0.000 0.008
C6H 1.079 1.096 0.017 0.016 0.026
CaH3N 3 s-Triazine CN 1.338 1.358 0.020 0.019 0.026 11
NCN 126.8 121.6 -5.2 -3.4 - 1.1
CS Carbon sulfide CS 1.534 1.447 - 0.087 - 0.050 mm
CHzS Thioformaldehyde CS 1.611 1.539 - 0.072 - 0.074 ii
CH 1.093 1.095 0.002 - 0.001 0.006
HCS 121.6 126,1 4.5 2.4 1.6
CH4S Thiomethanol CS 1.818 1.801 -0.017 -0.101 nn
SH 1.329 1.306 - 0.023 - 0.027 - 0,009
HSC 100.3 100.0 -0.3 2.1 1.0
HCSH 180.0 179.9 -0.1 -0.2 -0.1
C2H6S Dimethyl thioether CS 1,802 1.802 0.000 - 0.079 oo
CSC 98.9 102.1 3.2 9.4 3.8
HCS 106.6 110.4 3.8 3.1 2.2
H'CS 106.6 106.0 - 0.6 - 0.6 - 0.6
C4H4S Thiophene CS 1.714 1.725 0.011 - 0.035 pp
CCS 92.2 91.4 -0.8 1.4 2.0
C3C2 1.370 1.366 - 0.004 0.004 - 0.003
CCC 111.5 112.1 0.6 0.4 -0.4
CSO Carbon oxysulfide CO 1.159 1.176 0.017 0.022 qq
CS 1.559 1.504 - 0.055 - 0.049 - 0.065
C2H3NS Methylisothiocyanate CS 1.597 1.498 - 0.099 - 0.088 rr
C=N 1.192 1.231 0.039 0.027 0.025
C-N 1,479 1.440 - 0.039 - 0.042 - 0.072
C-N=C 141.6 139.4 -2.2 -0.3 - 1.0
C2N2S Sulfur dicyanide CN 1.157 1.164 0.007 0.007 ss
CS 1.701 1.664 - 0.037 - 0.071 - 0.067
NCS 170.0 176.5 6.5 7.0 7.4
CSC 98.4 101.5 3.1 5.1 3.4
CS2 Carbon disulfide CS 1.553 1.481 -0.072 -0.061 oo
80

TABLE 10 (continued)

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AMI

C2H6S2 2,3-Dithiabutane CS t.810 1.804 - 0.006 - 0.085 tt

SS 2.038 2.021 -0.017 -0.103 -0.108
CSS 102.8 109.2 6.4 5.2 3.4
CSSC 84.7 88.0 3.3 19.8 10.0
SSC 102.8 109.2 6.4 5.1 3.4
CF Fluoromethylidyne CF 1.266 1.259 - 0.007 - 0.003 - 0.007 a
CHF Fluoromethylene CH 1.121 1.100 -0,021 - 0.001 0.006 a
CF 1.314 1.284 - 0.030 - 0.029 - 0.023
FCH I01.6 105.6 4.0 9.5 9.0
CH3F Fluoromethane CH 1.098 1.092 - 0.006 0.020 0.023 uu
CF 1.382 1.351 -0.031 -0.035 -0.007
FCH 108.5 108.6 0.1 2.1 1.0
C2H3F Fluoroethylene CC 1.333 1.333 0.000 0.018 0.007 vv
CH(g) 1.076 1.093 0.017 0.023 0.028
CEll(g) 127.7 126,2 -- 1.5 -4.7 --3.8
CH(t) 1.085 1.085 0.000 0.002 0.011
CCH(t) 123.9 121.2 -2.7 -3.0 -3.1
CH(c) 1.090 1.086 - 0.004 - 0.003 0.006
CCH(c) 121.4 123.8 2.4 3.1 1.6
CF 1.348 1.338 - 0.010 - 0.024 0.003
FCC 121.0 122.0 1.0 2.3 2.2
C3H3F Fluoroallene C1C2 1.301 1.310 0.009 0.019 0.012 ww
C1H 1.083 1.094 0.0I 1 0.016 0.022
HC1C2 124.3 124.8 0.5 - 1.4 - 1.4
C1F 1.360 1.340 -- 0.020 - 0.035 -- 0.008
FC1C2 121.9 122.2 0.3 1.0 1.8
C2C3 1.309 1.294 --0.015 -0.006 --0.014
C3H 1.086 1.087 0.001 0.005 0.015
HC3C2 120.8 122.3 1.5 2.1 1.5
CNF Cyanogen fluoride CN 1.159 I. 159 0,000 0.001 0.006 xx
CF 1.262 1.297 0.035 0.011 0.045
CF 2 Difiuoromethylene CF 1.300 1.298 - 0.002 0.004 0.012 a
FCF 104.9 106.3 1,4 3.4 1.1
COF z Carbonyl difluoride CO 1.174 1.199 0,025 0.045 0.046 a
CF 1.312 1.322 0,010 0.004 0.016
FCO 126.0 124.6 -- 1.4 - 1.9 -1.5
CSF 2 Thiocarbonyl difluoride CS 1.589 1.600 0,011 -0.014 oo
CF 1.315 1.338 0,023 0,006 0.027
FCS 126.5 128.9 2,4 - 0.5 1.I
CHF~ Trifluoromethane CH 1.098 1.110 0,012 0.038 0.032 a
CF 1.333 1.346 0,013 0.020 0.035
FCH 110.3 113.1 2.8 1.2 2.7
CzNF 3 Trifluoroacetonitrile CC 1.461 1.487 0,026 0.037 0.025 a
CF 1.335 1.350 0,015 0.020 0.036
CCF 111.4 113.4 2,0 0.5 2.1
CN 1.153 1.155 0,002 0.006 0.006
 81

TABLE 10 (continued)

Empirical Chemical Geometric Exp. Calc, Errors Ref.

formula name variable
PM3 MNDO AM1

CF~ Carbon tetrafluoride CF 1.321 1.337 0.016 0.026 0.037 yy

C2F 4 Tetrafluoroethylene CC 1.311 1.355 0.044 0.070 0.057 vv
CF 1.319 1,326 0.007 -0.001 0.021
FCC 123,8 125.l 1.3 0.4 1.6
C~F6 Hexalluoroethane CC 1,560 1.608 0.048 0.114 0.061 a
CF 1.320 1,341 0.021 0.026 0.043
FCC 109.5 111.4 1.9 1.2 2.6
CHCI Chloromethylene CH 1.120 1.100 - 0.020 - 0.020 - 0.010 a
CC1 1.689 1,554 -0.135 0.050 -0.042
C1CH 103.4 115.5 12.1 6.0 7.7
CH3C1 Chtoromethane CCI 1,781 1.764 - 0.017 0.014 - 0.040 a
CH 1.096 1.094 - 0.002 0.006 0.016
HCC1 110.9 109.9 - 1.0 -- 2.8 - 2.6
CHF2C1 Chlorodifluoromethane CH 1.090 1.108 0,018 0.036 0.037 zz
CCI 1.740 1.822 0.082 0.099 0.069
C1CH 107.0 109,9 2.9 - 2.2 - 1.8
CF 1.350 1.346 - 0.004 - 0.008 0.019
FCCI 110.5 110,1 --0.4 -0.4 1.9
FCCIH 120.0 122.6 2.6 1.0 2.0
CH2C12 Dichloromethane CCI 1.772 1.758 --0.014 0.014 --0.031 yy
CICC1 111.8 107.9 - 3.9 - 0.6 1.2
CH 1.103 1.102 -0.001 0.000 0.010
COCI/ Carbonyl chloride CO 1,166 1.198 0.032 0.034 0.056 a
CC1 1.746 1.737 - 0.009 0.014 - 0.027
C1CO 124.4 124.2 -0.2 -0.5 - 1.I
CFzC12 Dichlorodifluoro-
methane CC1 1.770 1.808 0.038 0.055 0.037 a
CICC1 108.5 106.6 - 1.9 - 1.7 - 1.8
CF 1.330 1.345 0,015 0.007 0,040
FCC1 109.8 l 11.4 1.6 0.8 2.l
CHCI3 Chloroform CC1 1.782 1.753 -- 0,029 0.000 -- 0.034 yy
CICH 107.5 110.4 2,9 1.1 0.2
CFCI 3 Trichlorofluoromethane CF 1.330 1,349 0.019 - 0.003 0.046 a
CC1 1.760 1,779 0.019 0.046 0.026
CCI 4 Carbon tetrachloride CCt 1.760 1,747 -0,013 0,022 0,000 yy
C2C16 Hexachloroethane CC 1.550 1.512 -0.038 0.016 0.007 a
CC1 1.740 1.754 0.014 0.050 0.020
CICC 109.0 110.2 1.2 2.4 0.7
CH3Br Bromomethane CBr 1.933 1.951 0.018 - 0.055 - 0.028 oo
CH t .086 1,090 0.004 0.016 0.024
HCBr 107,7 108.4 0.7 0.8 1.1
C2H3OBr Acetyl bromide CC 1.516 1,477 - 0.039 - 0.001 - 0.026 oo
CBr 1.973 t,966 -- 0,007 - 0.086 -- 0.026
BrCC 111,0 106,3 --4.7 2.0 2.3
CO 1.183 1.185 0.002 0.026 0,042
CCO 127.1 134.3 7,2 0.2 - 3.0
82

TABLE 10 (continued)

Empirical Chemical Geometric Exp, Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

CNBr Cyanogen bromide BrC 1,789 1.796 0.007 - 0,046 - 0.029 a

CN 1.158 1.155 - 0.003 0.003 0.006
CF3Br Trifluorobromomethane CBr 1.909 1.960 0.05l 0.029 0.134 a
CF 1.328 1,335 0.007 0.019 0.039
FCBr 110.3 1 I0.8 0.5 0.8 3.4
CH2Br 2 Dibromomethane CH 1.079 1.095 0.016 0.023 0.031 oo
HCH 113.6 111.4 -2.2 -2.7 -3.3
CBr 1.927 1.912 - 0.015 - 0.059 - 0.025
BrCH 106.5 112.5 6.0 2.0 1.7
C4OzBr 2 1,2-Dibromocyclobu-
tene-3,4-dione C = C2 1.356 1.343 - 0.013 0.008 0.015 aaa
C2Br4 Tetrabromoethylene CC 1.362 1.450 0.088 - 0.020 - 0.018 oo
CBr 1.881 1,865 - 0.016 - 0.060 - 0.024
BrCC 122.4 111.9 - 10.5 1.3 0.0
CH~I Iodomethane CH 1.084 1.093 0.009 0.020 0.025 oo
CI 2.132 2.028 -0.104 -0.117 -0.082
HCH 111.2 109,9 - 1.3 -2,8 - 1.4
CNI Cyanogen iodide CN 1.159 1.155 - 0.004 0.005 0.003 a
CI 1.994 1.908 - 0.086 - 0.103 - 0.067
CF3I Trifluoroiodomethane CI 2.130 2.052 - 0.078 -0.005 0.045 a
CF 1.332 1.340 0.008 0.022 0.037
FCI 110.6 112.1 1.5 1.9 3.5
CP Carbon phosphide CP 1.562 1.389 - 0.173 - 0.145 - 0.029 a

References to Table 10

a Chase, M.W., Davies, C.A., Downey, J.R., Frurip, D.R., McDonald, R.A., Syverud, A.N., J A N A F Thermochemi-
cal Tables, 3rd ed., J. Phys. Chem. Ref. Data 14, Suppl. 1 (1985).
b Herzberg, G., Molecular Spectra and Molecular Structure, III, (Electronic Spectra and Electronic Structure of Po-
lyatomic Molecules), Van Nostrand, New York, NY 1966.
c Maki, A.G. and Toth, R.A., J. Mol. Spectr., 17 (1965) 136.
d Stigliani, W.M., Laurie, V.W. and Li, J.C., J. Chem. Phys., 62 (1975) 1890.
e Bonham, R.A. and Bartell, L.S., J. Am. Chem. Soc., 81 (1959) 3491.
f Bastiansen, O., Fritsch, F.N. and Hedberg, K., Acta Crystallogr., 17 (1964) 538.
g Lide, D.R. and Christensen, D., J. Chem. Phys., 35 (1961) 1374.
h Callomon, J.H. and Stoicheff, B.P., Can. J. Phys., 35 (1957) 373.
i Almenningen, A., Bastiansen, O. and Traetteberg, M., Acta Chem. Scand., 15 (1961) 1557.
j Fukuyama, T., Kuchitsu, K. and Morino, Y., Bull. Chem. Soc. Jap., 42 (1969) 379.
k Cox, K.W., Harmony, M.D., Nelson, G. and Wiberg, K.B., J. Chem. Phys., 50 (1969) 1976.
1 Haugen, W. and Traetteberg, M., Acta Chem. Scand., 20 (1966) 1226.
m Almenningen, A., Anfinsen, I.M. and Haaland, A., Acta Chem. Scand., 24 (1970) 43.
n Meiboom, S. and Snyder, L.C., J. Chem. Phys., 52 (1970) 3857.
o Scharpen, L.R.H. and Laurie, V.W., J. Chem. Phys., 39 (1963) 1972.
p Lide, D.R., J. Chem. Phys., 33 (1960) 1519.
 83

q McCelland, B.W. and Hedberg, K., J. Am. Chem. Soc., 109 (1987) 7304.
r Tamagawa, K., Iijima, T. and Kimura, M., J. Mol. Struct., 30 (1976) 243.
s Baron, P.A., Brown, R.D., Burden, F.R., Domaitle, P.J. and Kent, J.E., J. Mol. Spectrosc., 43 (1970) 401.
t Chiang, J.F. and Bauer, S.H., J. Am. Chem. Soc., 91 (1969) 1898.
u Bastiansen, O., Fernholt, L., Seip, H.M., Kambara, H. and Kuehitsu, K., J. Mol. Struct., 18 (1973) 163.
v Herzberg, G., Molecular Spectra and Molecular Structure, I, (Spectra of Diatomic Molecules), 2nd ed., Van Nos-
trand, New York, NY 1950.
w Takagi, K. and Oka, T., J. Phys. Soc. Jap., 18 (1963) 1174.
x Lees, R.M. and Baker, J.G., J. Chem. Phys., 48 (1968) 5299.
y Cox, A.P., Thomas, L.F. and Sheridan, J., Spectrochim. Acta, 15 (1959) 542.
z Blukis, U., Kasai, P.H. and Myers, R.J., J. Chem. Phys., 38 (1963) 2753.
aa Kuchitsu, K., Fukuyama, T. and Marino, Y., J. Mol. Struct., 1 (1968) 463.
bb Bak, B., Christensen, D., Dixon, W.B., Hansen-Nygarrd, L., Rastrup-Anderson, J. and Shottlander, M., J. Mol.
Spectrosc., 9 (1962) 124.
cc Kwei, G.H. and Curl, R.F., J. Chem. Phys., 32 (1960) 1592.
dd Trotter, J., Acta Crystallogr., 13 (1960) 86.
ee Lister, D.G., Tyler, J.K., Hog, J.H. and Larson, N.W., J. Mol. Struct., 23 (1974) 253.
ff Lide, D.R., 5. Chem. Phys., 27 (1957) 343.
gg Costain, C.C., J. Chem. Phys., 29 (1958) 864.
hh Wollrab, J.E. and Laurie, V.W., J. Chem. Phys., 51 (1969) 1580.
ii Johnson, D.R., Lovas, F.J. and Kirchhoff, W.H., J. Phys. Chem. Ref. Data, 1 (1972) I011.
jj Morley, J.O., J. Chem. Soc. Perkin Trans. II, (1987) in press.
kk Cradock, S., Liescheski, P.B., Rankin, D.W.H. and Robertson, H.E., J. Am. Chem. Soc., 1I0 (1988) 2758.
11 Lancaster, J.E. and Stoicheff, B.J., Can. J. Phys., 34 (1965) 1016.
mm Sutton, L.E., Tables of Interatomic Distances and Configurations in Molecules and Ions, (Special Publication No.
11 + 18), Chem. Soc., London (1958), (1965).
nn Kilb, R.W., J. Chem. Phys., 23 (1955) 1736.
oo Callomon, J.H., Hirota, E., Kuchitsu, K., Lafferty, W.J., Maki, A.G. and Pore, C,S., Structure Data on Free Polya-
tomic Molecules, (Landolt-Bornstein, New Series, Group II, Vol. 7), Springer-Verlag, Berlin, 1976.
pp Bak, B., Christensen, D., Hansen-Nygarrd, L. and Rastrup-Anderson, J., J. Mol. Spectrosc., 7 (1961) 58.
qq Watson, J.K.G., J. Mol. Spectrosc., 48 (1973) 479.
rr Anderson, D.W.W., Rankin, D.W.H. and Robertson, A., J. Mol. Struct., 14 (1972) 385).
ss Rosmus, P., Stafast, H. and Bock, H., Chem. Phys. Lett., 34 (1975) 275.
tt Sutter, R., Driezler, H. and Rudolph, F.Z., Naturforschung A, 20 (1965) 1676.
uu Duncan, J.L., J. Mol. Struct., 6 (1970) 447.
vv Carlos, J.L., Karl, R.R. and Bauer, S.H., J. Chem. Soc. Faraday Trans. 2 (1974) 177.
ww Ogata, T., Fujii, K., Yoshikawa, M. and Hirota, F., J. Am. Chem. Soc., 109 (1987) 7639.
xx Tyler, J.K. and Sheridan, J., Trans. Faraday Soc., 59 (1963) 2661.
yy Rodgers, A.S., Chao, J., Wilhoit, R.C. and Zwolinski, B.J., J. Phys. Chem. Ref. Data, 3 (1974) 117.
zz Chen, S.S., Wilhoit, A.C. and Zwolinski, B.J., J. Phys. Chem. Ref. Data 5 (1975) 571.
aaa Hagen, K. and Lunelli, B., J. Phys. Chem., 93 (1989) 1326.
84

TABLE l I
COMPARISON OF EXPERIMENTAL AND CALCULATED MOLECULAR GEOMETRIES FOR INORGANIC
MOLECULES

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

H2 Hydrogen HH 0.741 0.699 - 0.042 - 0.078 - 0.064 a

H20 Water OH 0.957 0.951 - 0.006 - 0.014 0.004 b
HOH 104.5 107.7 3.2 2.3 - 1.0
0 2 O x y g e n , triplet state OO 1.216 1.169 - 0.047 - 0.082 - 0.130 c
H20 2 Hydrogen peroxide OO 1.475 1.482 0.007 - 0.179 - 0.175 d
OH 0.950 0.945 - 0.005 0.011 0.033
HOO 94.8 96.5 1.7 12.2 11.2
HOOH 119.8 180.0 60.2 60.5 8.1
H40 2 Water dimer OO 3.000 2.769 - 0.231 0.905 - 0.383
H3N Ammonia NH 1.012 0.999 -0.013 -0.005 -0.014 a
HNH 106.7 108.1 1.4 - 1.4 2,4
NO N i t r o g e n oxide NO I. 151 1.127 - 0.024 - 0.028 - 0.036 a
NO 2 N i t r o g e n dioxide NO 1,197 1.181 -0.016 --0.023 -0.038 a
ONO 136.0 137.8 1.8 - 2.8 0.4
HNO2 Nitrous acid (cis) N-O 1.460 1.339 -0.121 -0.163 -0.169 a
N=O 1.200 1.175 -0.025 -0.031 -0.038
ONO 114.0 113.3 -0.7 3.1 2.6
OH 0.980 0.960 - 0.020 - 0.017 0.003
HON 103.0 109.9 6.9 16.7 12,5
HNO 2 N i t r o u s acid (trans) N - O 1.460 1.383 - 0.077 - 0.148 -0.141 a
N = O 1.200 1.167 - 0.033 - 0.034 - 0.042
ONO 118.0 109.1 - 8.9 - 4.7 - 5.2
OH 0.980 0.950 - 0.030 - 0.022 - 0.005
HON 105.0 104.6 --0.4 5.0 2.0
HNO 3 Nitric acid N = O 1.206 1.203 - 0.003 0.005 - 0.011 a
O=N=O t30.0 132,7 2.7 -3.5 --I.0
N - O 1.405 1.410 0.005 - 0.065 - 0.072
OH 0.960 0.953 - 0.007 0.002 0.022
NOH 102,0 109,0 7,0 12.0 7.7
N2 Nitrogen NN 1.094 1.098 0.004 0.010 0.012 c
H4N 2 Hydrazine NN 1.449 1.440 - 0.009 - 0.052 - 0.071 a
NH 1,022 1.001 - 0.021 - 0.001 - 0.008
HNN 112.0 106.5 - 5.5 - 4.8 - 4.6
HNNH 90.0 180.3 90.3 90.2 90.2
N20 N i t r o u s oxide NN 1.128 1.124 - 0.004 0.000 0.000 a
NO 1.184 1.197 0.013 -0.003 -0.009
H2N20 2 NH2-NO 2 NN 1.427 1.431 0.004 - 0.020 - 0.060 e
NO 1.206 1.2l l 0.005 0.003 - 0.002
NH 1.005 1.000 - 0.005 0.013 - 0.003
ONO 130.1 127.2 2.9 - 6.0 - 6.1
N20 3 D i n i t r o g e n trioxide NN 2.080 1.409 - 0.671 - 0.706 - 0.728 a
NO 1.100 1.172 0.072 0.063 0.061
NNO 110.0 118.2 8.2 11.0 12.8
NO' 1.180 1.268 0.088 0.079 0.101
O'NO' 134.0 150.4 16.4 14.8 13.5
 85

TABLE I 1 (continued)

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

NaO4 Dinitrogen tetroxide NN 1.750 1.759 0.009 - 0.135 0.068 a

NO 1,180 1,195 0.015 0,009 -0.008
ONN 113.2 1 I4.6 1.4 2.1 1.0
N3 Azide NN 1.181 1.174 -0.007 -0.007 -0.004 a
H2S Hydrogen sulfide HS 1.328 1.290 - 0.038 - 0.028 a
HSH 92.2 93.5 1.3 5.9 6.6
SO2 Sulfur dioxide SO 1.432 1.442 0.010 0.044 a
OSO 119.5 106.1 - 13.4 - 12.7 - 16.6
SO3 Sulfur trioxide SO 1.430 1.384 - 0,046 0.061 a
H2SO~ Sulfuric acid S- O 1.550 1,668 0.118 0.078 a
OH 0.970 0.947 - 0.023 - 0.023 - 0.006
SOH 105.0 117.7 I2,7 12.2 1.5
S=O 1.420 1.668 0.248 0.209 0.287
NS Sulfur nitride SN 1.495 1.452 - 0.043 - 0.055 a
$2 Sulfur dimer SS 1.889 1.857 -0.032 -0.114 a
H2S2 H2S 2 SS 2.055 2.034 -0.021 -0.131 f
SH 1.327 1.311 -0.016 -0.023 -0.005
HSS 91.3 103.2 11.9 11.2 9.9
HSSH 90.5 93.4 2.9 9.1 21.7
$6 $6 SS 2.057 2.048 - 0.009 - 0.110 g
SSS 102.2 107.5 5.3 2.9 2.9
SSSS 74.5 64.6 - 9.9 - 5.1 - 5.0
$8 Ss SS 2.048 1.973 -0.075 -0.113 h
SSS 107.9 i16.0 8.1 0.4 0.4
SSSS 98.6 87.6 - 11.0 - 0.2 - 0.2
HF Hydrogen fluoride HF 0.917 0.938 0,021 0,039 -0.091 i
NOF Nitrosyl fluoride NF 1.520 1.367 -0.153 -0.215 -0.153 a
NO 1.130 1.162 0.032 0.031 0.018
FNO 110.2 111.6 1.4 3.7 2.1
Fz Fluorine FF 1.412 1.350 -0.062 -0.146 0.015 a
HzF2 Hydrogen fluoride dimer HF 0.920 0.939 0.019 0.036 - 0.093 a
H'F 1.870 1.743 - 0.127 1.093 0.417
H'FH 108.0 147.0 39.0 71.5 - 18.1
OFa FzO OF 1.412 1.378 -0.034 -0.131 -0.058 a
FOF 103.2 100.9 - 2.3 5.9 - 0.7
SF2 Sulfur difluoride SF 1.592 1.560 - 0.032 - 0.020 a
FSF 98.2 96.5 - 1.7 1.3 - 1.3
SOF 2 Thionyl fluoride SO 1.412 1.467 0.055 0.068 a
SF 1.585 1. 574 - 0,011 0.023 0.058
FSO 106.8 101.0 - 5.8 - 3.9 - 7.0
FSF 92.8 93.9 1.1 4.2 2.0
S02F2 Sulfuryl fluoride SF 1.530 1.547 0.017 0,080 a
FSF 96.1 95.3 -0.8 1.7 1.1
SO 1.405 1.400 -0.005 0.103 0.171
OSO 124.0 126.4 2.4 0.8 3.5
86

TABLE 11 (continued)

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

S2Fz FSSF SF 1.635 1.584 - 0.051 - 0.063 a

SS 1.888 2.008 0.120 0.078 0.060
FSS 108.3 112.0 3.7 - 1.8 - 2.1
FSSF 87.9 87.4 - 0.5 - 1.3 - 0.8
$2F2 SSF2 SS 1.860 1.919 0.059 0.035 a
SF 1.598 1.592 - 0.006 0.003 0.054
FSS 107.5 114.5 7.0 1.4 1.3
FSF 92.5 90.8 - 1.7 3.9 0.2
NF3 Nitrogen trifluoride NF 1.371 1.354 - 0.017 - 0.056 - 0,011 a
FNF 102.2 105.0 2.8 4.0 0.4
SF4 Sulfur tetrafluoride SF 1.545 1.633 0.088 0.061 a
FSF 101.6 81.0 -20.6 - 13.0 - 18.6
SF' 1.646 1.591 - 0.055 - 0.005 0.022
FSF' 87.8 81.0 - 6.8 0.9 -- 4.8
SF6 Sulfur hexafluoride SF 1.564 1.560 - 0.004 0.091 a
HC1 Hydrogen chloride HCI 1.275 1.268 - 0.007 0.073 0.009 a
OCI Chlorine monoxide CIO 1.546 1.548 0.002 0.073 0.090 a
NOC1 NOC1 CIN 1.950 1.764 -0.186 -0.167 -0.219 a
NO 1.170 1.156 -0.014 -0.033 -0.033
CNC1 114.0 118.9 4.9 3.5 5.4
NO2CI NO2C1 C1N 1.830 1.818 -0.012 -0.020 -0.059 a
NOC1 1.210 1.197 -0.013 -0.019 -0.024
FC1 Chlorine fluoride C1F 1.628 1.582 - 0.046 0.022 0.019 a
O3FC1 C103F C1F 1.630 1.690 0.060 0.105 0.051 a
C10 1.460 1.453 - 0.007 0.270 0.328
OC1F 95.2 102.8 7.6 11.0 3.0
FaC1 Chlorine trifluoride C2v C1F 1.598 1.671 0.073 0.101 0.085 a
C1F' 1.698 1.671 - 0.027 0.001 - 0.015
FC1F' 87.5 120.0 32.5 32.5 32.5
C12 Chlorine C1C1 1.986 2.035 0.049 0.010 - 0.068 a
0C12 C120 C10 1.701 1.700 0.001 -0.018 0.032 a
C1OC1 110.8 109.2 - 1.6 2.1 0.3
SC12 Sulfur dichloride SC1 2.015 2.031 0.016 - 0.043 a
C1SCI 102.7 101.6 - 1.1 3,6 3.1
SOC12 Thionyl chloride SO 1.443 1.479 0.036 0.026 j
SC1 2.076 2.080 0.004 - 0.038 - 0.102
C1SO 106.3 104,7 - 1.6 0.3 - 0.2
82C12 C1SSC1 SC1 2.057 2.044 -0.013 -0.081 a
SS 1.931 1.965 0.034 - 0.011 - 0.004
C1SSC1 108.2 113.1 4.9 0.0 - 0.3
HBr Hydrogen bromide HBr 1.415 1.471 0.056 0.025 0.006 k
NOBr BrNO BrN 2.140 1.888 - 0.252 - 0.271 - 0.218 1
NO 1.146 1.147 0.001 -0.007 -0.011
BrNO 114.5 120.8 6.3 4.4 8.0
FBr BrF BrF 1.755 1.774 0.019 - 0.029 0.022 a
 87

T A B L E 11 ( c o n t i n u e d )

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

F3Br Bromine trifluoride BrF 1.721 1.786 0.065 0.036 0.087 a

BrF' 1.806 1.786 - 0.020 - 0.049 0.010
FBrF' 86.2 120.0 33.8 33.8 - 4.8
FsBr Bromine pentafluoride BrF(ax) 1.680 1.755 0.075 0.086 0.134 a
BrF(eq) 1.790 1.774 - 0.016 - 0.019 0.014
C1Br Bromine chloride BrCI 2.136 2.176 0.040 - 0.056 - 0.072 a
Brz Bromine BrBr 2.283 2.443 0.160 - 0. I 15 - 0.099 a
HI Hydrogen iodide HI 1.609 1.677 0.068 - 0.042 - 0.022 k
FI Iodine fluoride IF 1.906 1.889 - 0.017 - 0.004 - 0.025 a
Fsl Iodine pentafluoride IF(ax) 1.844 1.867 0.023 0.139 0.087 m
IF(eq) 1.869 1.882 0.013 0.088 0.029
F(ax)IF(eq) 81.9 102.6 20.7 - 3.8 - 4.6
FTI Iodine heptafluoride IF(ax) 1.760 2.698 0.938 0.731 0.866 n
IF(eq) 1.860 1.913 0.053 0.218 0.114
CII Iodine chloride IC1 2.327 2.192 - 0.135 - 0.065 - 0.109 a
BrI Iodine bromide IBr 2.485 2.561 0.076 -0.135 -0.131 a
I2 Iodine II 2.666 2.668 0.002 - 0.151 - 0.128 a
HA1 A1H A1H 1.648 1.663 0.015 - 0.222 a
A10 A10 A10 1.618 1.533 -0.085 -0.143 a
A1F A l u m i n u m fluoride A1F 1.654 1.652 - 0.002 - 0.094 a
A1F 3 Aluminum trifluoride AIF 1.630 1.644 0.014 - 0.038 a
AIF 4 A1F4(-) A1F 1.690 1.688 - 0.002 - 0.041 a
AICI Aluminum chloride A1C1 2.130 1.947 - 0.183 - 0.055 a
A1C13 Aluminum trichloride A1C1 2.060 1.966 - 0.094 0.005 a
A1Br Aluminum bromide A1Br 2.295 2.292 - 0.003 - 0.093 a
A1Br 3 Aluminum tribromide AIBr 2.270 1.875 - 0.395 - 0.095 a
AII 3 Aluminum triiodide AII 2.499 2.487 - 0.012 - 0.174 a
AI 2 A1z A1A1 2.467 2.554 0.087 -0.175 a
A120 AI/O A10 1.730 1.677 - 0.053 - 0.124 a
H2Si Silylene (singlet) Sill 1.519 1.513 - 0.006 - 0.139 - 0.062 o
HSiH 92.1 94.9 2.8 5.2 8.9
H4Si Silane Sill 1.481 1.488 0.007 - 0.105 - 0.020 a
SiN Silicon nitride SiN 1.572 1.464 - 0.108 - 0.021 - 0.087 a
SiF2 Difluorosilylene SiF 1.591 1.575 -0.016 -0.013 0.021 a
FSiF 101.0 95.3 - 5.7 -4.1 - 3.9
HSiF3 Trifluorosilane Sill 1.447 1.507 0.060 - 0.072 - 0.007 a
SiF 1.562 1.590 0.028 0.031 0.047
FSiH 110.6 112.6 2.0 3.0 1.2
SiF 4 Tetrafluorosilane SiF 1.552 1.580 0.028 0.032 0.052 a
SiC1 Chlorosilylidyne SiC1 2.063 1.946 - 0.117 0.009 - 0.077 a
SiC12 Dichlorosilylene C1SiC1 109.7 101.9 - 7.8 - 4.2 - 5.3 p
SIC14 Silicon t e t r a c h l o r i d e SiC1 2.017 2.041 0.024 0.063 0.022 a
H3SiBr Bromosilane SiBr 2.210 1.901 - 0.309 0.018 0.030 1
Sill 1.481 1.491 0.010 -0.113 -0.016
HSiBr 107.9 108.3 0.4 - 1.0 2.4
SiBr4 Silicon t e t r a b r o m i d e SiBr 2.150 1.796 - 0.354 0.040 0.093 a
88

TABLE 11 (continued)

Empirical Chemical Geometric Exp. Calc. Errors Ref.

formula name variable
PM3 MNDO AM1

H3SiI Iodosilane SiI 2.437 2.012 - 0.425 - 0.051 - 0.003 a

Sill 1.486 1.492 0.006 - 0.116 - 0.020
HSiI 108.5 108.0 - 0.5 - 0.1 1.3
SiI4 Silicon tetraiodide SiI 2.430 2.467 0.037 - 0.097 - 0.005 a
Si2 Silicon dimer SiSi 2.246 2.297 0.051 - 0.259 - 0.019 a
H6Si 2 Disilane SiSi 2.331 2.396 0.065 -0.158 0.086 q
Sill 1.492 1.487 - 0.005 - 0.113 - 0.026
HSiSi 110.3 109.7 - 0.6 0.9 - 0.7
H3P Phosphine PH 1.420 1.324 - 0.096 - 0.080 - 0.057 a
HPH 93.8 97.1 3.3 2.3 2.7
PO Phosphorus oxide PO 1.476 1.459 - 0.017 - 0.053 - 0.039 a
PF 3 Phosphorus trifluoride PF 1.570 1.558 - 0.012 - 0.014 - 0.027 a
FPF 97.8 95.8 -2.0 1.1 0.2
PF5 Phosphorus pentafluo-
ride PF(ax) 1.577 1.553 - 0.024 0.025 - 0.028 a
PF(eq) 1.534 1.528 - 0.006 0.039 0.001
PC13 Phosphorus trichloride PC1 2.039 2.064 0.025 - 0.050 - 0.120 a
C1PC1 100.3 99.7 - 0.6 4.9 5.1
PC15 Phosphorus pentachlo-
ride PCl(ax) 2.190 2.093 - 0.097 - 0.078 - 0. l 15 a
PCl(eq) 2.040 2.052 0.012 - 0.007 - 0.069
PBr 3 Phosphorus tribromide PBR 2.220 2.150 -0.070 -0.131 -0.119 1
BrPBr 101.0 101.3 0.3 4.7 5.8
Pz Phosphorus dimer PP 1.894 1.715 - 0.179 - 0.200 - 0.271 a
P4 Phosphorus tetramer PP 2.210 2.197 -0.013 -0.158 -0.170 a
P406 Phosphorus trioxide PO 1.650 1.708 0.058 - 0.046 a
OPO 99.0 96.5 - 2.5 - 3.0

References to Table 11

a Chase, M. W., Davies, C. A., Downey, J. R., Frurip, D. R., McDonald, R. A. and Syverud, A. N., J A N A F Thermo-
chemical Tables, 3rd ed., J. Phys. Chem. Ref. Data 14, Suppl. 1 (1985).
b Herzberg, G., Molecular Spectra and Molecular Structure, III (Electronic Spectra and Electronic Structure of Polya-
tomic Molecules), Van Nostrand, New York, NY 1966.
c Herzberg, G., Molecular Spectra and Molecular Structure, I (Spectra of Diatomic Molecules), 2nd ed., Van Nostrand,
New York, NY 1950.
d Oelfke, W. C. and Gordy, W., J. Chem. Phys., 51 (1969) 5336.
e Tyler, J. K., J. Mol. Spectrosc., 11 (1963) 39.
f Winnewisser, G., Winnewisser, M. and Gordy, W., J. Chem. Phys., 49 (1968) 3465.
g Donohue, J., Caron, A. and Goldish, E., J. Am. Chem. Soc., 83 (1961) 3748.
h Caron, A. and Donohue, J., Acta Crystallogr., 18 (1965) 562.
i Kuipers, G. A., Smith, D, F, and Nielsen, A. N., J. Chem. Phys., 25 (1956) 275.
j Hargittai, I., Acta Chem. (Budapest), 49 (1969) 351.
k Huber, K. P. and Herzberg, G., Molecular Spectra and Molecular Structure, IV (Constants for Diatomic Molecules),
Van Nostrand, Reinhold, New York, NY 1979.
 89

1 Callomon, J. H., Hirota, E., Kuchitsu, K., Lafferty, W. J., Maki, A. G. and Pote, C. S., Structure Data on Free
Polyatomic Molecules (Landolt-Bornstein, New Series, Group II, Vol. 7), Springer-Verlag, Berlin, 1976.
11 Greenwood, N. N. and Earnshaw, A., Chemistry of the Elements, Pergamon Press, Oxford, 1984, p. 976.
n Stull, D.R. and Prophet, H., Natl. Stand. Ref. Data Ser. (U.S., Natl. Bur. Stand.) NSRDS-NBS 37, 1971.
o Meadows, J. H. and Schaefer III, H. F., J. Am. Chem. Soc., 98 (1976) 4383.
p Davis, R. W. and Gerry, M. C. L., J. Mol. Spectrosc., 60 (1976) 117.
q Shotton, K. C., Lee, A. G. and Jones, W. J., J. Raman Spectrosc., 1 (1973) 243.
90
TABLE 12 UNSIGNED AVERAGE ERRORS IN BOND LENGTHS (/~)

H C N O F A1 Si P S C1 Br I

H 1
PM3 0.042
MNDO 0.078
AM1 0.064

50 72
PM3 0.009 0.017
MNDO 0.010 0.014
AM1 0.015 0.017

N 7 21 8
PM3 0.011 0 . 0 1 8 0.092
MNDO 0,139 0 . 0 2 2 0.120
AMI 0.051 0 . 0 2 4 0.122

O 8 18 18 3
PM3 0.014 0 . 0 1 2 0 . 0 3 4 0.095
MNDO 0.014 0.021 0 . 0 8 5 0.389
AM1 0,012 0 , 0 3 0 0 . 0 6 2 0.229

3 19 2 1 1
PM3 0.056 0 . 0 1 5 0 . 0 8 5 0 , 0 3 4 0.062
MNDO 0.389 0 . 0 1 6 0 . 1 3 5 0.131 0.146
AM1 0.200 0,026 0.082 0 , 0 5 8 0,015

A1 1 2 3 1
PM3 0.015 0.069 0 . 0 0 6 0,087
MNDO 0.222 0.133 0 . 0 5 8 0.175

Si 6 1 1 3 2
PM3 0.016 0.011 0.108 0.024 0.058
MNDO O.110 0 . 0 6 4 0.021 0.025 0.209
AMI 0.025 0 , 0 5 0 0,087 0,040 0.052

P 1 1 2 3 2
PM3 0.096 0.173 0.038 0.014 0.096
MNDO 0.080 0.145 0.050 0,026 0.179
AMI 0.057 0,029 (see 0.019 0.221
text)
S 3 I0 1 7 8 8
PM3 0.026 0 . 0 4 7 0 . 0 4 3 0.074 0.033 0.046
MNDO 0.026 0.063 0.055 0.084 0.043 0.087

C1 1 10 2 3 4 2 2 3 3 1
PM3 0.007 0 . 0 3 7 0 . 0 9 9 0 . 0 0 3 0 . 0 5 2 0 , 1 3 8 0 . 0 7 0 0 , 0 4 5 0.011 0,049
MNDO 0.073 0 . 0 3 6 0 . 0 9 3 0 . 1 2 0 0 . 0 5 7 0,030 0 . 0 3 6 0 , 0 4 5 0 . 0 5 4 0.010
AM1 0.009 0 . 0 3 3 0 , 1 3 9 0 . 1 5 0 0.042 0.049 0.101 0.068

Br 1 6 1 5 2 2 1 1 1
PM3 0.056 0 . 0 1 9 0.252 0.039 0 . 1 9 9 0 . 3 3 2 0.070 0.040 0.160
MNDO 0.025 0 , 0 5 6 0.27l 0.044 0 . 0 9 4 0 . 0 2 9 0.131 0.056 0.115
AMI 0,006 0.044 0.218 0.053 0.061 0.072 0.099

1 3 5 1 2 1 1 1
PM3 0.068 0.089 0.209 0 . 0 1 2 0.231 0,135 0.076 0.002
MNDO 0.042 0,075 0.236 0.174 0.074 0.065 0 . 1 3 5 0.151
AM1 0.022 0,065 0.224 0.004 0,109 0.131 0.128
 91
TABLE 13 S I G N E D AVERAGE ERRORS IN B O N D L E N G T H S (/k)

H C N O F A1 Si P S C1 Br I

H 1
PM3 - 0.04
MNDO - 0.08
AM1 - 0.06

50 72
PM3 0.00 - 0.01
MNDO 0.01 0.00
AM1 0.01 -0.01

N 7 21 8
PM3 -0.01 0.01 -0.09
MNDO 0.14 0.00 -0.12
AM1 0.04 --0.01 -0.10

O 8 18 18 3
PM3 -0.0t 0.01 0.00 -0.09
MNDO -0.01 0.02 0.02 0.21
AM1 0.01 0.03 -0.01 -0.23

3 19 2 1 1
PM3 -0.03 0.00 -0.08 -0.03 -0.06
MNDO 0.39 0.00 -0.14 -0.13 -0.15
AM1 0.08 0.02 -0.08 -0.06 0.01

A1 1 2 3 1
PM3 0.02 - 0.07 0.00 0.09
MNDO -0.22 -0.13 --0.06 --0.18

Si 6 I 1 3 2
PM3 0.01 0.01 -0.I1 0.01 0.06
MNDO --0.11 -0.06 -0.02 0.02 -0.21
AM1 -0.03 -0.05 -0.09 0.04 0.03

P I l 2 3 2
PM3 -0.t0 -0.17 0.02 --0.01 -0.10
MNDO -0.08 -0.14 -0.05 0.02 --0.18
AM1 - 0.06 - 0.03 (see -- 0.02 - 0.22
text)
S 3 10 1 7 8 8
PM3 -0.03 -0.04 -0.04 0.06 - 0.01 0.0l
MNDO -0.03 -0.06 -0.05 0.08 0.02 - 0.06

C1 I I0 2 3 4 2 2 3 3 1
PM3 -0.01 -0.01 -0.10 0.00 0.01 -0.14 -0.05 -0.02 0.00 0.05
MNDO 0.07 0.04 - 0.09 0.11 0.06 -0.02 0.04 -0.05 -0.05 0.01
AM1 0.01 0.00 --0.14 0.15 0.03 -0.03 -0.10 - 0.07

Br 1 6 1 5 2 2 1 1 I
PM3 0.06 0.01 -0.25 0.02 -0.20 -0.33 -0.07 0.04 0.16
MNDO 0.02 -0.05 -0.27 0.00 -0.09 0.03 -0.13 -0.06 -0.11
AM1 0.01 0.00 - 0.22 0.05 0.06 - 0.12 -0.07 -0.10

I 1 3 5 1 2 l 1 1
PM3 0.07 - 0.09 0.20 -0.01 -0.19 -0.14 0.08 0.00
MNDO - 0.04 - 0.08 0.23 -0.17 -0.07 -0.07 -0.i4 -0.15
AM1 -0.02 -0.03 0.21 0.00 -0.11 -0.13 -0.13
92

TABLE 14
A V E R A G E ERRORS IN C A L C U L A T E D G E O M E T R I E S

In bonds No. Average error (A) in bond lengths

involving
PM3 MNDO AMP

Hydrogen 83 0.005 0.015 0.006

Carbon 211 0.002 0.002 0.002
Nitrogen 61 0.012 0.015 0.014
Oxygen 62 0,006 0.017 0.011
Fluorine 57 0.011 0.023 0.017
Aluminum 12 0.044 0,074
Silicon 19 0,045 0.030 0,019
Phosphorus 13 0.041 0,050 0,044
Sulfur 40 0.007 0.010
Chlorine 33 0,021 0.020 0.028
Bromine 21 0.059 0,046 0.042
Iodine 15 0.055 0,063 0.049

Atoms No. Average error (degrees) in bond angles

PM3 MNDO AM1

Unsigned Signed Unsigned Signed Unsigned Signed

H C H 4 1.200 - 0.550 4.775 2,025 4,100 1.750

H C C 19 1.342 0.826 1.658 0.468 1,342 0.237
H C O 3 3.333 3,333 2.733 2.733 2.933 2.933
H C F 3 2.300 2.300 4.267 4.267 4.233 4.233

H C C1 4 4.725 4,225 3.025 0.525 3.075 0.875

H C Br 2 3,350 3.350 1.400 1.400 1.400 1.400
C C C 16 2.319 1,581 4,137 4.013 2.431 2,144
C C O 4 2.100 1.500 0.725 0,525 0.950 - 0.950
C C F 5 1.300 1.300 1.080 1,080 2.060 2.060

C C Br 2 7.600 - 7,600 1.650 1.650 1.150 1.150

N C O 2 8.300 - 8.300 7.600 - 7,600 7,600 - 7,600
F C C1 2 1.000 0,600 0.600 0.200 2,000 2.000
C1 C CI 2 2.900 2.900 1.150 - 1.150 1.500 - 0.300
H N H 2 2.100 2.100 0,900 - 0.900 2.750 2.750

H N C 2 3.300 - 3.300 5.050 - 5.050 0,800 - 0.800

C N C 3 1.867 0,400 2,467 2.267 1.400 0.733
N N O 3 8.667 8.667 9,300 9,300 9,100 9,100
O N O 5 3.400 - 1,600 4,020 -2,780 3.060 - 1.860
H O C 2 2.900 2,900 7.250 7.250 2.250 2.250

H O N 3 4.767 4.500 11.233 11.233 7.400 7.400

C O C 2 1,550 1.550 4.850 4,850 0.800 0.800
C S C 2 1.950 1.150 3.250 3.250
C S S 2 6.400 6.400 5.150 5.150
O S O 2 7.900 - 5,500 6.750 5.950

F S F 6 5.450 -5.083 4.167 -0,167

F S S 2 5.350 5.350 1.600 - 0,200
S S S 2 6.700 6.700 1.650 1.650

a AM 1 results are for systems which do not include aluminum or sulfur.

 93

TABLE 15
COMPARISON OF EXPERIMENTAL AND CALCULATED DIPOLE MOMENTS

Empirical Chemical Dipole m o m e n t Errors Ref.

formula name
Exp. Calc. PM3 MNDO AM1

C3H~ Cyclopropene 0.45 0.39 - 0.06 0.03 - 0.09

C3H 4 Propyne 0.78 0.36 - 0.42 - 0.66 - 0.38
C3H 6 Propene 0.37 0.23 -0.14 -0.33 -0.14
C3H 8 Propane 0.08 0.01 - 0.07 - 0.08 - 0.08
C4H 6 Bicyclobutane 0.68 0.43 0.25 - 0.27 - 0.26
C4.H6 Cyclobutene 0.13 0.15 0.02 - 0.05 0.04
CsH 6 Cyclopentadiene 0.42 0.53 0.11 -0.24 0.11
CsH s Cyclopentene 0.20 0.15 -0.05 -0.15 -0.03
C6H 6 Fulvene 0.42 0.66 0.24 0.27 0.27
C7H8 Toluene 0.36 0.26 -0.I0 -0.30 -0.10
H20 Water 1.85 1.74 -0.I1 -0.07 0.01
CO Carbon monoxide 0.11 0.18 0,07 0,09 - 0.05
CH/O Formaldehyde 2.33 2.16 -0.17 -0.17 -0.01
CH40 Methanol 1.70 1.49 -0.21 -0.22 -0.08
C2H20 Ketene 1.42 1.06 -0.36 -0.38 -0.07
C2H40 Acetaldehyde 2.69 2.54 -0.15 -0.31 0.00
C2H40 Ethylene oxide 1.89 1.77 -0.I2 0.03 0.02
C2H60 Ethanol 1.69 1.45 -0.24 -0.29 -0.14
C2H60 Dimethyl ether 1.30 1.25 -0.05 -0.03 0.13
C3H20 Propynal 2.47 2.62 0.[5 0.38 0.35
C3H40 Acrolein 3.12 2.36 - 0.76 - 0.66 - 0.59
C3H60 Acetone 2.88 2.78 -0.10 -0.37 0.04
C~H~O Furan 0.66 0.22 -0.44 -0,24 -0.17
C~H 1oO Diethyl ether 1.15 1.33 0.18 0,21 0.30
C6H60 Phenol 1.45 1.14 -0.31 -0.29 -0.22
CvHsO Anisole 1.38 1.08 -0,30 -0.31 -0.13
CH20 2 Formic acid 1.41 1.51 0.I0 0.08 0.07
C2H40 2 Acetic acid 1.74 1.83 0.09 -0.06 0.12
C2H402 Methyl formate 1.77 1.59 -0.18 -0.15 -0.26
C3H602 Propionic acid 1.75 1.81 0.06 - 0.04 0.08
C3H60 z Methyl acetate 1.72 1.82 0.10 0.03 0.02
O3 Ozone 0.53 1.7I 1.18 0.65 0.67
H3N Ammonia 1.47 1.55 0.08 0.28 0.38
CHN Hydrogen cyanide 2.98 2.70 -0.28 -0.48 -0.62
CHsN Methytamine 1.31 1.40 0,09 0,17 0.18
C2H3N Acetonitrile 3.92 3.21 -0.71 - 1.29 - 1.03
C2H3N Methyl isocyanide 3.85 3.69 -0.16 -1,68 -1.02
C2HsN Ethyteneimine (Azirane) 1.90 1.69 -0.21 -0.15 -0.15
C2HvN Ethylamine 1.22 1.43 0.21 0.30 0.33
C2HvN Dimethylamine 1.03 1.27 0,24 0.14 0.20
C3HN CH,C,CN 3.72 3.33 - 0,39 - 0.50 - 0.68
C3H3N Acrylonitrile 3.87 3.25 - 0.62 - 0.90 - 0.87
C3H9N Trimethylamine 0.61 1.15 0.54 0.14 0.41
C4HsN Pyrrole 1.74 2.18 0.44 0.07 0.21
CsHsN Pyridine 2,22 1.94 - 0.28 - 0.26 - 0.25
94

TABLE 15 (continued)

Empirical Chemical Dipole moment Errors Ref.

formula name
Exp. Calc. PM3 MNDO AM1

C6HTN Aniline 1.53 1.30 - 0.23 - 0.07 0.01 a

CH3NO Formamide 3.73 3.12 - 0.61 - 0.62 - 0.03 a
C3HTNO Dimethylformamide 3.82 3.06 - 0.76 - 0.65 - 0.27 a
HNO 2 Nitrous acid, trans 1.86 2.08 0.22 0.42 0.45 e
HNOz Nitrous acid, cis 1.42 1.27 --0.15 0.14 0.02 e
HNO 3 Nitric acid 2.17 2.32 0.15 0.61 0.40 a
CH2N2 Diazomethane 1.50 1.92 0.42 -0.25 -0.17 a
CH2N2 N=N-CHz- 1.59 1.87 0.28 - 0.04 0.04 a
CH6N2 Methylhydrazine 1,66 0.32 -- 1.34 - 1.42 - 0.89 f
NzO Nitrous oxide 0.17 0.26 0.09 0.59 0.47 a
CH3N3 Methyl azide 2.17 2.28 0.11 -0.54 -0.23 a
HzS Hydrogen sulfide 0.97 1.78 0.81 0.51 g
CH2S Thioformaldehyde 1.65 2.07 0.42 0.88 h
CH4S Thiomethanol 1.52 1.95 0.43 0.01 g
C2H6S Thioethanol 1.52 1.98 0.46 - 0.04 g
C2H65 Dimethyl thioether 1.50 1.96 0.46 0.07 h
C4H4S Thiophene 0.53 0.67 0.14 0.01 g
CSO Carbon oxysulfide 0.71 0.38 -0.33 -0.31 h
SO2 Sulfur dioxide 1.57 3.63 2.06 2.16 g
C2H692 2,3-Dithiabutane 1.98 2.57 0.59 - 0.09 g
HF Hydrogen fluoride 1.83 1.40 -0.43 0.16 -0.09 i
CH3F Fluoromethane 1.86 1.44 -0.42 --0.10 -0.24 J
C2HF Fluoroacetylene 0.70 1.11 0.41 0.88 0.36 k
C2H3F Fluoroethylene 1.43 1.37 - 0.06 0.27 - 0.05 h
C2HsF Fluoroethane 1.96 1.58 -0.38 -0.09 -0.27 h
C6HsF Fluorobenzene 1.66 1,60 - 0.06 0.30 - 0.08 h
HOF Hypoftuorous acid 2.23 1.68 - 0.55 - 0.42 - 0.63 1
CHOF HCOF 2.02 2.46 0.44 0.48 0.55 h
CNF Cyanogen fluoride 2.17 1.63 - 0.54 - 1.28 - 0.96 k
NOF Nitrosyl fluoride 1.81 0.26 - 1.55 - 1.30 - 1.43 h
NO2F Fluorine nitrite 0.47 0.89 0.42 0.19 0.35 g
CH2F2 Difluoromethane 1.96 1.81 -0.15 0.26 0.08 h
C2H4F2 1,1-Difluoroethane 2.30 2.12 -0.18 0,20 -0.03 h
C6H4F2 o-Difluorobenzene 2.59 2.74 0.15 0.77 0.09 m
OF2 Difluorine oxide 0.30 0.38 0.08 0.02 - 0.19 h
COF2 Carbonyl fluoride 0.95 1.08 0.13 -0.14 0.33 g
N2F2 cis-Difluorodiazene 0.16 0.63 0.47 -0.14 0.50 g
CHF 3 Trifluoromethane 1.65 1,88 0.23 0.58 0.43 J
C2HF3 Trifluoroethylene 1.30 1.49 0.19 0.52 0.13 n
C2H3F3 1,1,1-Trifluoroethane 2.32 2.40 0.08 0.55 0.27 h
C2HOzF3 Trifluoroacetic acid 2.28 1.96 -0.32 0.17 -0.42 h
NF3 Nitrogen trifluoride 0.24 0.26 0.02 - 0.04 - 0.20 h
C/NF3 Trifluoroacetonitrile 1.26 0.32 - 0.94 - 0.90 - 1.23 o
COF 4 Trifluoromethyl hypofluorite 0.33 0.28 - 0.05 - 0.24 0.02 p
HC1 Hydrogen chloride 1.12 1.38 0.26 0.36 0.26 q
C2HCI Chloroacetylene 0.44 0.14 -0.30 0.34 -0.17 r
 95

TABLE 15 (continued)

Empirical Chemical Dipole moment Errors Re f.

formula name
Exp. Calc. PM3 MNDO AM1

FC1 Chlorine fluoride 0.88 1.42 0.54 0.59 0.03 s

SC12 Sulfur dichloride 0.36 0.59 0.23 0.61 g
HBr Hydrogen bromide 0.83 1.27 0.44 0.24 0,55 a
CH3Br Bromomethane 1.82 1.55 - 0.27 - 0.26 -- 0.34 a
C2HBr Bromoacetylene 0.00 0.28 0.28 0.43 0.57 r
CzH3Br Bromoethylene 1.42 1.33 -0.09 -0.11 -0.12 s
C2HsBr Bromoethane 2.03 1.85 -0,18 --0.37 -0,37 a
C3HTBr 1-Bromopropane 2.18 1.81 -- 0,37 - 0.47 -- 0.48 a
C6HsBr Bromobenzene 1.70 1.18 - 0.52 - 0.28 - 0.25 a
OBr BrO 1.61 3.32 1.71 0.38 0.75 t
C2HaOBr Acetyl bromide 2.43 2.95 0.52 0.02 0.16 s
FBr Bromine fluoride 1.42 2.25 0.83 0.68 0.04 s
CF3Br Bromotrifluoromethane 0.65 0.90 0.25 0.53 0.37 u
CIBr Bromine chloride 0.52 0.06 --0.46 0.23 -0.07 v
CHzBr2 Dibromomethane 1.43 1.45 0.02 --0.06 --0,I1 s
CF2Br2 Difluorodibromomethane 0.66 0.54 -0.12 0.17 -0.06 s
CHBr3 Bromoform 0.99 0.96 - 0.03 - 0.08 - 0.08 s
HI Hydrogen iodide 0.44 0.97 0.53 0.57 0.83 a
CH3I Methyl iodide 1.65 1.44 -0.21 -0.28 -0.30 a
C2HsI Iodoethane 1.91 1.83 -0.08 -0.50 -0.41 a
CsHTI l-lodopropane 2.04 1.78 -0.26 -0.57 -0.50 a
C6HsI Iodobenzene 1.70 0.81 -0.89 -0.10 -0.27 a
CF3I Trifluoroiodomethane 1.00 1.55 0.55 1.13 0.67 w
BrI Iodine bromide 0.74 0.53 -0.21 -0.02 -0.11 v
CH2I 2 Diiodomethane 1.62 1.20 - 0.42 - 0.47 - 0.50 a
A1F Aluminum fluoride 1.53 3.29 1.76 - 1.22 x
CzH6Si Vinylsilane 0.66 0.24 -0.42 -0.12 -0.1I a
C2HsSi Ethylsilane 0.81 0.37 - 0.44 - 0.71 - 0.45 g
CzHsSi Dimethylsilane 0.75 0.46 - 0.29 - 0.57 - 0.30 a
C3HloSi Trimethylsilane 0.52 0.37 -0.15 -0.35 -0.13 a
H2SiF 2 Diftuorosilane 1.54 1.41 -0.13 0.69 0.03 a
HSiF 3 Trifluorosilane 1.27 1.51 0.24 1.53 0.28 a
H2SiCIz Dichlorosilane 1.18 2.68 1.50 2.29 0.58 a
HSiC13 Trichlorosilane 0.86 2.38 1.52 1.87 0.56 a
H2SiBr2 Dibromosilane 1.43 3.09 1.66 1.90 0.38 s
C6H19Si2N Hexamethyldisilazane 0.37 0.35 -0.02 -0.10 0.24 a
H3P Phosphine 0.58 1.18 0.60 0.77 1.71 a
CHsP Methylphosphine 1.10 1.15 0.05 0.51 0.92 h
C2HTP Dimethylphosphine 1.23 1.16 - 0.07 0.58 0.56 h
C3HgP Trimethylphosphine 1.19 1.08 --0.11 0.74 0.37 h
PE3 Phosphorus trifluoride 1.03 2.25 1.22 1.41 0.63 a
POF 3 Phosphorus oxyfluoride 1.76 2.04 0.28 -0.56 0.88 a
H4P2 P2H4 0.92 2.19 1.27 - 0.92 - 0.92 Y
96

References to Table 15

a Nelson, R.D., Lide, D.R. and Maryott, A.A., Nat. Stand. Ref. Data Set., Natl. Bur. Stand. No. 10 (1967).
b Baron, P.A., Brown, R.D., Burden, F.R., Domaille, P.J. and Kent, J.E., J. Mol. Spectrosc., 43 (1970) 401.
c Nygaard, L., Nielsen, J.T., Kirchheimer, J., Maltesen, G., Rastrup-Andersen, J. and Sorensen, G.O., J. Mol. Struct.,
3 (1969) 491.
d Sorensen, G.O., Mahler, L. and Rastrup-Andersen, J., J. Mol. Struct., 20 (1974) 119.
e Cox, A.P., Brittain, A.H. and Finnigan, D.J., Trans. Faraday Soc., 67 (1971) 2179.
f Lattimer, R.P. and Harmony, M.D., J. Phys. Chem., 53 (1970) 4575.
g McClellan, A.L., Tables of Experimental Dipole Moments, Vol. 2, Rahara Enterprises, E1Cerrito, 1974.
h McClellan, A.L., Tables of Experimental Dipole Moments, Freeman, San Francisco, CA, 1963.
i Muenter, J.S. and Klemperer, W., J. Chem. Phys., 52 (1970) 6033.
j Wofsy, C.S., Muenter, J.S. and Klemperer, W., J. Chem. Phys., 55 (1971) 2014.
k Tyler, J.K. and Sheridan, J., Trans. Faraday Soc., 59 (1963) 2661.
1 Rock, S i . , Pearson, E.F., Appleman, E.H., Norris, C.L. and Flygare, W.H., J. Chem. Phys., 59 (1973) 3940.
m Nygaard, L., Hansen, E.R., Hansen, R., Rastrup-Anderson, J. and Sorensen, G.O., Spectroehim. Acta, Part A, 23
(1967) 2813.
n Bhaumik, A., Brooks, W.V.F. and Dass, S.C., J. Mol. Struct., 16 (1973) 29.
o Foreman, P.B., Chien, K.R., Williams, J.R. and Kukolich, S.G., J. Mol. Spectrose., 52 (1974) 251.
p Buckley, P. and Weber, J.P., Can. J. Chem., 52 (t974) 942.
q Hojer, G. and Meza, S., Acta Chem. Scand., 26 (1972) 3723.
r Scharffenberg, P., Theor. Chim. Acta, 49 (1978) 115.
s Weast, R.C. (Ed.), CRC Handbook of Chemistry and Physics, 60th ed., CRC Press, Boca Raton, FL, 1980.
t Byfleet, C.R., Carrington, A. and Russell, D.K., Mol. Phys., 10 (1971) 271.
u Cox, A.P., Duxbury, G., Hardy, J.A. and Kawashima Y., J. Chem. Soc. Faraday Trans. 2.76 (1980) 339.
v Nair, K.P.R., Kemiai Kozl 52 (1979) 431.
w Hellwege, K.H. and Hellwege, A.M. (Eds.) Molecular Constants from Microwave Spectroscopy (Landolt-Bornstein,
New Series II, Vol. 4), Springer-Verlag, Berlin 1967.
x Lovas, F.J. and Tiemann, E., J. Phys. Chem. Ref. Data, 3 (1974) 609.
y Durig, J.R., Carreira, L.A. and Odom, J.M., J. Am. Chem. Soc., 96 (1974) 2688.
 97

TABLE 16
COMPARISON OF EXPERIMENTAL AND CALCULATED IONIZATION POTENTIALS

Empirical Chemical Ionization potential Errors Ref.

formula name
Exp. Calc. PM3 MNDO AM1

H2 Hydrogen 15.40 16. l 1 0.71 0.35 - 0.48

CH 4 Methane 13.60 13.64 0.04 0.27 - 0.29
C2H2 Acetylene 11.40 11.61 0.21 -0.39 0.10
C2H4 Ethylene 10.51 10.64 0.13 -0.34 0.04
C2H 6 Ethane 12.00 11.98 - 0.02 0.70 - 0.23
C3 Carbon, trimer 11.10 11.75 0.65 -0.06 0.23
C3H 4 Allene 10.07 10.18 0.11 -0.05 0.07
C3H 4 Cyclopropene 9.86 9,88 0.02 0,02 - 0.04
Call 4 Propyne 10.37 10.89 0.52 0,35 0.37
C3H 6 Cyclopropane 11.00 11.78 0.78 0.43 0.48
C3H 6 Propene 9.88 i0.10 0.22 0.08 0.11
C3H8 Propane 1t.50 11.51 0.01 0.84 -0.18
C4H2 Diacetylene 10.17 10.47 0.30 -0.18 0,20
C4H4 CH z = C = C = CH z 9.10 9.34 0.24 -- 0.09 0.24
C4H 6 1,2-Butadiene 9.15 9.72 0.57 0.69 0.52
C4H 6 1-Butyne 10.20 10.77 0.57 0.48 0.42
C4H 6 2-Butyne 9.60 10.34 0.74 0.87 0.57
C4H 6 Cyclobutene 9.43 9.82 0.39 0.34 0.29
C4H6 1,3-Butadiene 9.08 9.47 0.39 0.06 0.25
C4H8 1-Butene 9.70 10.03 0,33 0,24 0.23
C4H 8 Cyctobutane 10.70 11.02 0.32 1.11 0.29
C4Hlo n-Butane 11.20 11.35 0.15 1.01 -0.03
C4Hlo Isobutane 11.40 11.59 0.19 0.72 -0.11
CsH 6 Cyclopentadiene 8,57 9.23 0.66 0,47 0.51
C5H8 Cyclopentene 9.18 9.52 0.34 0.54 0.26
CsH~o 1-Pentene 7.90 10.15 2.25 2.07 2.09
CsH~o 2-Methyl-l-butene 7,40 9,85 2,45 2.41 2.30
C5H10 2-Methyl-2-butene 8.70 9.39 0.69 0.94 0.55
C5Hlo 3-Methyl- 1-butene 9.60 10.26 0.66 0,38 0.49
C5Hlo Cyclopentane 10.50 11.07 0.57 1.56 0.47
CsHx2 2-Methylbutane 10.30 11.44 1.14 1.70 0.90
C5H12 Neopentane 11.30 12.06 0.76 0.82 0.23
C5H~2 n-Pentane 10.30 11.30 1.00 1.86 0.81
C6H 6 Benzene 9.25 9.75 0.50 0.14 0.40
C6Hlo Cyclohexene I0.30 9.59 - 0.71 - 0.55 - 0.81
C6Hlz Cyclohexane 10.30 11.29 0.99 1.44 0.64
CTH 8 Cycloheptatriene 8.50 8.95 0.45 0.22 0.25
CTH 8 Toluene 8.82 9.44 0.62 0.46 0.51
CsHto Ethylbenzene 8.80 9.52 0.72 0.48 0.58
C8H14 Bicyclo(2.2.2)-octane 9.45 10.94 1.49 1.95 1.11
CaoH8 Naphthalene 8.15 8.84 0.69 0.42 0.56
CloH16 Adamantane 9.60 10.85 1.25 1.67 0.83
C~4Hlo Anthracene 8.16 8.25 0.09 - 0.11 - 0.04
H20 Water 12.62 12.32 -0.30 -0.43 --0.16
H20 Water, second I.P. 14.70 14.52 - 0.18 - 0.23 0.25
98

TABLE 16 (continued)

Empirical Chemical Ionization potential Errors Ref.

formula name
Exp. Calc. PM3 MNDO AM1

H20 Water, third I,P. 18.50 17.58 -0.92 0.61 -0.30 f

H20 Water, fourth I.P. 32.20 36.83 4.63 7.83 4.22 f
14.01 13.03 -0.98 -0.58 -0.70 a
CO Carbon monoxide
10.I0 10.63 0.53 0.94 0.68 a
CH20 Formaldehyde
10.96 I1.14 0.18 0.46 0.17 a
CH40 Methanol
9.64 9.46 -0.18 -0.35 -0.04 a
C2HzO Ketene
10.21 10.71 0.50 0.67 0.51 a
C2H40 Acetaldehyde
C2H40 Ethylene oxide 10.57 11.34 0.77 0.92 0.76 b
10.60 10.90 0.30 0.70 0.28 a
C2H60 Ethanol
10.04 10.69 0.65 1.00 0.57 a
C2H60 Dimethyl ether
10.11 10.68 0.57 0.53 0.58 a
C3H40 Acrolein
9.72 10.77 1.05 1.04 0.95 a
C3H60 Acetone
10.00 10.62 0.62 0.81 0.59 a
C3H60 Propanal
8.88 9.38 0.50 0.26 0.44 a
C4H40 Furan
9.90 10.50 0.60 0.49 0.57 a
C~H~O 2-Butenal
9.83 10.63 0.80 0.97 0.76 a
C4H80 Butanal
9.60 10.79 1.19 1.18 0.96 a
C4HloO Diethyl ether
10.16 10.92 0.76 0.93 0.65 a
C5H1~O 3-Pentanol
9.70 10,05 0.35 0,04 0.30 a
C7H60 Benzaldehyde
8.40 9. l l 0.71 0.44 0.60 a
C~HsO Anisole
CO2 Carbon dioxide 13.78 12.73 - 1.05 - 0.99 - 0.57 b
11.51 11.56 0.05 0.23 0.31 a
CHzO2 Formic acid
10.59 10.57 -0.02 0.16 0.07 a
C2H202 trans-Glyoxal
10.80 11.44 0.64 0.77 0.82 a
C2H402 Acetic acid
C2H,O2 Methyl formate 11.02 11.35 0.33 0.59 0.55 g

10.60 10.77 0.17 0.09 0.29 a

C2H602 Dimethyl peroxide
10.60 10.22 -0.38 -0.53 -0.10 a
C302 Carbon suboxide
10.60 9.96 - 0.64 - 0.73 - 0.54 a
C3H402 /~-Propiolactone
10.50 11.34 0.84 1.00 0.99 a
C3H602 Propionic acid
10.60 11.27 0.67 0.86 0.80 a
C3H602 Methyl acetate
8.38 10.89 2.51 2.40 2.36 a
CsHsOz Acetylacetone
9.80 10.13 0.33 -0.03 0.28 a
C7H602 Benzoic acid
03 Ozone 12.75 12.69 -0.06 -0.05 0.35 h
10.84 11.71 0.87 0,86 1.18 a
C4HzO~ Maleic anhydride
11.20 11.67 0.47 0.59 0.76 a
CzHzO~ Oxalic acid
H3N Ammonia 10.85 9.70 - 1.15 0.34 -0.43 b
CHN Hydrogen cyanide 13.60 12.60 - 1.00 -0.19 0.08 b
9.60 9A0 - 0.20 0,96 0.15 a
CH5 N Methylamine
12.21 12.33 0.12 0.58 0.25 a
C2H3N Acetonitrile
11.32 11.70 0.38 0.92 0.67 a
CzH3N Methyl isocyanidc
9.90 9.92 0.02 0.78 0.41 a
CzHsN Ethyleneimine (Azirane)
9.50 9.50 0.00 1.01 0.39 a
C2HTN Ethylamine
C2HTN Dimethylamine 8.93 9.22 0.29 I. 11 0.46 i
11.75 ll.58 -0.17 -0.40 -0.10 a
C3HN CH.C.CN
10.91 10.89 - 0.02 - 0.30 - 0.05 a
C3H3N Acrylonitrile
11.90 12.0l 0.ll 0.69 0.09 a
C3HsN Ethyl cyanide
 99

TABLE 16 (continued)

Empirical Chemical Ionization potential Errors Ref.

formula name
Exp. Calc. PM3 MNDO AM1

C3H9N Trimethylamine 8.54 9.07 0.53 1.05 0.58 i

C4HsN Pyrrole 8.21 8.93 0.72 0.35 0.45 a

CsHsN Pyridine 9.67 10.10 0.43 0.02 0.26 a
C6H7N Aniline 7.70 8.61 0.91 1.05 0.82 a
CvHsN Phenyl cyanide 9.70 10.10 0.40 0.I1 0.32 a
CHNO Hydrogen isocyanate 11.60 10.59 -- 1.01 --0.50 -0.36 a
CHaNO 2 Nitromethane 11.30 12.17 0.87 0.24 0.68 a
CH3NO 2 Methyl nitrite 11.00 10.64 - 0.36 0.42 0.19 a
C3HTNO2 Alanine 8.I0 9.88 1.78 2.7I 2.27 a
C6HsNOz Nitrobenzene 9.90 10.60 0.70 0.41 0.66 a
N2 Nitrogen 15.60 13.80 - 1.80 -0.73 - 1.28 a
CH2N2 Diazomethane 9.00 9.22 0.22 - 0.33 -0.18 a
CH6N2 Methylhydrazine 9.30 8.92 --0.38 0.36 - 0.34 a
C2N2 Cyanogen 13.36 12.87 -0.49 -0.16 -0.05 a
C4H4N2 Pyridazine 9.30 9.94 0.64 1.19 1.37 a
C4H4N: Pyrimidine 9.73 10.29 0.56 0.65 0.85 a
C4H~N2 Pyrazine 9.90 10.16 0.26 0.12 0.35 a
N204 Dinitrogen tetroxide II.40 11.6I 0.2I 0.65 -0.I9 a
N205 Dinitrogen pentoxide 12.30 12.38 0.08 0.88 1.32 a
CH3N3 Methyl azide 9.81 9.96 0.15 0.10 - 0.22 a
HS Hydrogen sulfide 10.40 9.79 -0.61 -0.55 J
H2S Hydrogen sulfide 10.43 9.63 - 0.80 - 0.44 k
CS Carbon sulfide 11.30 10.60 -0.70 -0.21 a
CH2S Thioformaldehyde 9.34 9.06 - 0.28 0.64 1

CH4S Thiomethanol 9.44 9.21 - 0.23 0.33 k

C2H2S Thioketene, H2CCS 8.89 8.81 - 0.08 0.34 m
C2HaS Thiirane 8.87 9.23 0.36 0.79 k
C2H68 Thioethanol 9.21 9.19 -0.02 0.52 n
C2H6S Dimethyl thioether 8.65 8,88 0.23 0.93 k
C3H6S Thietane 8.65 8.95 0.30 0.88 o
C3HsS Isopropanthiol 9.14 9.22 0.08 0.57 a
C3H8S 1-Propanthiol 9.19 9.19 0.00 0.54 a
CaH~S Thiophene 8.95 9.54 0.59 0.55 k
C4H8S Tetrahydrothiophene 8.60 8.79 0.19 0.85 a
C4H10S Butanethiol 9.15 9.19 0.04 0.58 a
C6H6S Thiophenol 8.47 8.78 0.3I 0.36 o
CSO Carbon oxysulfide 11.20 10.71 - 0.49 - 0,69 a
C2H~.SO Thiolacetic acid 9.70 10.05 O.35 0.83 a
CaH6SO Dimethyl sulfoxide 9.01 9.35 0.34 0.79 a
SO2 Sulfur dioxide 12.30 10.55 - 1.75 - 0.44 k
SO3 Sulfur trioxide 11.00 12.91 1.91 2.16 k
CHNS Hydrogen isothiocyanate 9.94 9.38 -0.56 -0.23 a
C2H3NS Methyl isothiocyanate 9.37 9.17 - 0.20 0.09 a
C2HaNS Methyl thiocyanate 9.96 9.68 -0.28 0.33 a
H2S2 Hydrogen disulfide 10.01 9.86 -0.15 0.21 p
CS2 Carbon disulfide I0.08 9.83 - 0.25 - 0.42 k
100

TABLE 16 (continued)

Empirical Chemical Ionization potential Errors Ref.

formula name
Exp. Calc. PM3 MNDO AM1

C2H6S2 Ethanedithiol- 1 , 2 9.30 9.32 0.02 0.54

CzH6S2 2,3-Dithiabutane 8.71 9.40 0.69 1.07
C2N2S2 Sz(CN)2 11.05 10,56 - 0.49 - 0.07
C3H4S3 1,3-Dithiolan-2-thione 8.40 9.25 O.85 1.05
Ss $8 9.04 9.O9 0.05 1.51
HF Hydrogen fluoride 16.06 16.14 0.08 - 1.24 - •.97
CHaF Fluoromethane 13.31 12.92 -0.39 -0.26 - 1.21
C2HF Fluoroacetylene 11.30 11.56 0.26 -0.24 -0.15
CEH3F Fluoroethylene 10.58 10.60 0.02 - 0.41 - 0.34
CEHsF Fluoroethane 12.43 12.07 -0.36 0.18 -0.85
C3H7F 2-Fluoropropane 11.08 12.10 1.02 1.25 0,44
C6HsF Fluorobenzene 9.19 9.81 0.62 0.28 0.35
CvHsO2F p-Fluorobenzoic acid 9.90 10.16 0.26 - 0.07 0.05
NOF Nitrosyl fluoride 12.94 11.54 - 1.40 -0.01 -0,42
NOzF Fluorine nitrite 13.51 13.37 -0.14 -0.52 -0.12
SF SF 10.00 9.75 -0.25 0.13
CH2F2 Difluoromethane 13.17 12.86 -0.31 -0.08 -1.15
C2Fz Difluoroacetylene 11.20 11.54 0.34 - 0.03 - 0.27
CzH2F2 gem-Difluoroethylene 10.72 10.54 -0.18 -0.54 -0,73
C2H4Fz 1,l-Difluoroethane 12.80 12.82 0.02 - 0.07 - 0.87
C6H4F2 o-Difluorobenzene 9.68 9.98 0.30 0.01 - 0.05
C6H~.F2 m-Difluorobenzene 9.68 10.02 0.34 0.05 0,04
C6H4Fz p-Difluorobenzene 9.30 9.87 0.57 0.26 0.19
OF2 Difluorine oxide 13.26 13.47 0.21 0.26 -0.36
NzFe trans-Difluorodiazene 13.40 11.91 -1.49 -0.40 -1,15
SF2 Sulfur difluoride 10.20 9.81 - 0.39 0.54
CSF2 Thiocarbonyl difluoride 10.46 10.37 - 0.09 0.68
SOF2 Thionyl fluoride 12.58 11.03 - 1.55 -0.13
SO2F 2 Sulfuryl fluoride 13.04 13.10 0.06 0.71
$2F2 FSSF 10.68 10.47 -0.21 0.47
CHF3 Trifluoromethane 14.80 14.36 - 0.44 - 0.23 - 1,49
C2HF3 Trifluoroethylene 10.54 10.68 0.14 -0.08 -0,45
CzH3F3 1,1,1-Trifluoroethane 13.80 14.38 0.58 0.21 -0,68
CTHsF3 Trifluoromethylbenzene 9.68 10.34 0.66 0.39 0.57
C2HOzF3 Trifluoroacetic acid 12.00 12.47 0.47 0.73 0,49
NF 3 Nitrogen triftuoride 13.73 12.24 - 1.49 0.20 - 0,55
CF4 Carbon tetraftuoride 16.23 16.79 0.56 0.58 -0,91
C2F4 Tetrafluoroethylene 10.50 10.84 0.34 0.42 - 0,22
COF,, Trifluoromethyl hypofluorite 13.60 14.18 0.58 0.63 0,07
N2F4 Tetrafluorohydrazine 12.00 12.46 0.46 1.04 0.47
SF4 Sulfur tetra fluoride 12.05 10.41 - 1.64 0.66
C6HF5 Pentafluorobenzene 9.75 10.55 0.80 0.65 0,32
C2F6 Hexafluoroethane 14.60 14.48 -0.12 -0.10 - 1,37
C6F6 Hexafluorobenzene 10.90 10.85 -0.05 -0.13 -0.53
C3OF6 Perfluoroacetone 12.10 12,71 0,61 0.90 0,24
SF~ Sulfur hexafluoride 15.70 16.39 0.69 0.68
 101

TABLE 16 (continued)

Empirical Chemical Ionization potential Errors Ref.

formula name
Exp. Calc. PM3 MNDO AM1

HC1 Hydrogen chloride 12.75 l 1.06 - 1.69 0.25 - 0.42 q

CH3C1 Methyl chloride 11.30 10.48 - 0.82 0.94 0.04 a
C6HsC1 Chlorobenzene 9.08 9.39 0.31 0.54 0.48 u
C7H5OC1 Benzoyl chloride 9.90 10.29 0.39 0.12 0.34 a
NOC1 Nitrosyl chloride 10.90 10.59 -0.31 1.07 0.66 a

NO2C1 Nitryl chloride 11.40 12.42 1.02 1.61 154 a

FC1 Chlorine fluoride 12.02 11.13 - 0.89 1.35 0.37 v
CH2FC1 Fluorochloromethane 11.74 10.85 -0.89 0.80 -0.16 a

SO2FC1 SO2C1F 12.61 12.43 -0.18 0.76 a

CHF2CI Difluorochloromethane 12.60 11.36 - 1.24 0.63 -0.32 a
CF3C1 Trifluorochloromethane 10.30 11.88 1.58 3.83 2.94 a
CH2C12 Dichloromethane 11.30 10.58 -0.72 1.19 0.09 a

COC12 Carbonyl chloride 11.84 11,23 - 0.61 0.86 0,53 a

SC12 Sulfur dichloride 9.70 9.60 - 0.10 1.00 a
SOC12 Thionyl chloride 11.10 10.65 -0.45 1.24 a
SO2C12 Sulfuryl chloride 12.40 10.55 - 1.85 -0.55 a
S2Cl 2 CtSSC1 9.40 9.96 0.56 1.67 a
CHFC12 Fluo rodichloromethane 12.00 10.99 - 1,01 0.96 0.03 a
CF2C12 Difluorodichloromethane 12.30 11.33 --0.97 1.06 0.21 a
CHC13 Chloroform 11.48 10.84 - 0.64 1.43 0.29 a
CFC13 Fluorotrichloromethane 11.90 11.16 -0.74 1.39 0,55 a

C2Cl 6 Hexachloroethane 11.20 10.84 - 0.36 1.88 0.98 a

HBr Hydrogen bromide 11.71 12.13 0.42 0,39 -0.25 q
CH3Br Bromomethane 10.53 11.01 0.48 1.03 0.27 a
C2H3Br Bromoethylene 9.90 10.44 0.54 0.35 0.25 a
C2HsBr Bromoethane 10.28 10.91 0.63 1,20 0.41 a
C3H7Br l-Bromopropane 10.18 10.93 0.75 1.29 0.51 a
C6HsBr Bromobenzene 9.25 9.81 0.56 0.30 0.35 a
C2H3OBr Acetyl bromide 10.55 11.20 0.65 0.88 0,63 w
CF3Br Bromotrifluoromethane 12.10 12.23 0.13 1.15 0.23 a
Br 2 Bromine 10,70 l 1,24 0.54 0.96 0.24 a
CH2Br2 Dibromomethane 10.50 10.59 0.09 1.20 0.46 a
CF2Br 2 Difluorodibromomethane 11.18 11.17 -0.01 1.10 0.13 a
C2F4Br2 1,2-Dibromotetraftuoroethane 14.44 12.00 - 2.44 - 1.65 -- 2.53 a
CHBr 3 Bromoform 10.50 10.84 0.34 1.37 0.57 a
CBr 4 Carbon tetrabromide 10.30 11.22 0.92 1.73 0.92 a
HI Hydrogen iodide 10.39 9.97 -0.42 0.82 0.52 x
CH3I Methyl iodide 9.50 9.47 - 0.03 1.35 1.01 a
C2H5I Iodoethane 9.34 9.44 0.10 1.48 1.09 a
C3H7I 1-Iodopropane 9.27 9.45 0.18 1.53 I. 16 a
C3HvI 2-Iodopropane 9.40 9.43 0.03 1.39 1.00 a
C,H9I 1-Butyl iodide 9.20 9.45 0.25 1.60 1.23 a
C6HsI Iodobenzene 8.70 9.05 0.35 0.85 0.95 a
C6Hlli Iodocyclohexane 8.91 9.42 0.51 1.86 1.47 a
CvHvI o-Iodotoluene 8.53 9.02 0.49 0.98 0.99 a
CvH7I m-Iodotoluene 8.55 9.01 0.46 0.96 0.97 a
102

TABLE 16 (continued)

Empirical Chemical Ionization potential Errors Ref.

formula name
Exp. Calc. PM3 MNDO AM 1

C7H7I p-Iodotoluene 8.38 8.94 0.56 1.07 1.03 a

CTHvI Benzyl iodide 8.91 9.35 0.44 0.55 0.60 a

CF3I Trifluoroiodomethane 10.45 10.28 -0.17 2.03 1.52 a

CII Iodine chloride 10.10 9,74 - 0.36 1.48 1.02 Y
BrI Iodine bromide 9.85 9.84 - 0.01 1.43 0.99 Y
I2 Iodine 9.34 9.53 0.19 1.53 1.33 z
I2 Iodine, second I.P. 11.03 10.27 - 0.76 1.66 1.45 z
I2 Iodine, third I.P. 12.95 12.53 - 0.42 0.40 0.25 z
CH2I 2 Diiodomethane 9.46 8.98 - 0.48 1,41 1,I0 a

C212 Diiodoacetylene 9.03 8.64 - 0.39 1.55 1.47 aa

C2H412 1,2-Diiodoethane 9.50 9.66 0,16 1.52 1.21 a

C3H9A1 Trimethylaluminum 9.76 10.12 0.36 0.92 bb

C2H6Si Vinylsilane 10.40 10.01 -0.39 --0.17 -0.01 a

C2HsSi Ethylsilane 10.95 10.64 -0.31 0.19 0.12 k

C2HsSi Dimethylsilane 11.20 10.65 -0.55 0.27 -0.03 a

C3HloSi Trimethylsilane 10.80 10.51 -0.29 0.60 0.15 a

C4H12Si Tetramethylsilane 10.40 10.42 0.02 0.94 0.52 a

CaH9SiF Fluorotrimethylsilane 11.00 10.47 - 0.53 0.49 0.06 a

H2SiF2 Difluorosilane 12.85 10.82 - 2.03 - 0.63 -- 1.24 cc

SiF4 Silicon tetrafluoride 15.81 15.19 - 0.62 0.02 -- 1.09 dd
H2SiCIz Dichlorosilane 11.70 10.27 - 1.43 0.82 --0.14 cc
C2H6SIC12 Diehlorodimethylsilane 10.79 10.27 - 0.52 1.43 0.70 a

SIC14 Silicon tetrachloride 11.79 11.23 -0.56 2.02 1.19 w

C6H19Si2N Hexamethyldisitazane 8.66 8.38 -0.28 1.03 0.83 ee

HaP Phosphine 9.98 8.67 - 1.31 1.36 0.44 x
CP Carbon phosphide I0.50 10.90 0.40 1.06 1.08 a

CHP Methinophosphine 10.79 10.74 -- 0.05 0.44 0.43 a

CH5P Methylphosphine 9.72 8.66 - 1.06 1.07 0.10 a

C2HvP Dimethylphosphine 9.10 8.68 -0.42 1.14 0.21 a
C3HgP Trimethylphosphine 8.60 8,81 0.21 1.25 0.30 k
CaHgPO3 Trimethyl phosphite 9.22 9.69 0.47 1.69 0.79 a
PF3 Phosphorus trifluoride 9.71 10.70 0.99 3.42 0.72 x
POF3 Phosphorus oxyfluoride 12.77 12.05 - 0.72 0.40 - 0.04 a

PC/3 Phosphorus triehloride 10.50 10.42 --0.08 1.67 0.13 a

POC13 Phosphorus oxychloride 11.85 12,29 0.44 0.93 1.27 a

10.80 11.86 a
PCI5 Phosphorus pentachloride 1.06 1.49 1.90

PBr 3 Phosphorus tribromide i0.00 10.84 0,84 1.44 0.60 a

PI3 Phosphorus triiodide 9.15 9.97 0.82 1.29 1.27 a
Phosphorus dimer 10.62 8.91 -- 1.71 0.82 0.84 a
P2
P4 Phosphorus tetramer 9.54 10.17 0.63 1.96 2.14 a
 103

References to Table 16

a Levin, R.D. and Lias, S.G,, Ionization Potentials and Appearance Potential Measurements, 1971-1981, Natl.
Stand. Ref. Data Ser., Natl. Bur. Stand. 7l, (1982), Cat. No. C13.48:71.
b Turner, D.W., Baker, C., Baker, A.D. and Brundle, C.R., Molecular Photoelectron Spectroscopy, Wiley-Inter-
science, London, 1970.
c Bischof, P. and Heilbronner, E., Helv. Chim. Acta, 53 (1970) 1677.
d Katrib, A. and Rabalais, J.W., J. Phys. Chem., 77 (1973) 2358.
e Murrell, LN. and Schmidt, W., J. Chem_ Soc., Faraday Trans. 2, 68 (1972) 1709.
f Potts, A.W. and Price, W.C., Proc. R. Soc. London, Ser. A, 326 (1972) 165.
g Sweigart, D.A. and Turner, D.W., J. Am. Chem. Soc., 94 (1972) 5592.
h Brundle, C.R., Chem. Phys. Lett., 26 (1974) 25.
i Maier, J.P. and Turner, D.W., J. Chem. Soc., Faraday Trans. 2, 69 (1973) 521.
j Chase, M.W., Davies, C.A., Downey, J.R., Frurip, D.R., McDonald, R.A. and Syverud, A.N., JANAF Thermo-
chemical Tables, 3rd ed., J. Phys. Chem. Ref. Data 14, Suppl. l (1985).
k Rosenstock, H.M., Draxel, K., Steiner, B.W. and Herron, J.T., Energetics of Gaseous Ions, J. Phys. Ref. Data, 6,
Suppl. 1 (1977).
1 Solouki, B., Rosmus, P. and Bock, H., Chem. Phys. Lett., 98 (1976) 6054.
m Bock, H., Solouki, B., Bert, G. and Rosmus, P., J. Am. Chem. Soc., 99 (9177) 1663.
n Omura, I., Higasi, K. and Baba, H., Bull. Chem. Soc. Japan, 29 (1956) 504.
o Gleiter, G. and Spanget-Larsen, J., Topics in Current Chemistry, 86 (1979) 139.
p Solouki, B. and Bock, H.~ Inorg. Chem., 16 (1977) 665.
q Frost, D.C., McDowell, C.A. and Vroom, D.A., J. Chem. Phys., 46 (1967) 4255.
Pietro, W.J., Francl, M.M., Hehre, W.J., DeFrees, D.J., Pople, J.A. and Binkley, J.S., J. Am. Chem. Soc., i04 (1982)
5039.
r Banna, M.S., Mills, B.E., Davis, D.W. and Shirley, D.A., J. Chem. Phys., 61 (1974) 4780.
s Frost, D.C., Lee, S.T, McDowell, C.A. and Westwood, N.P.C., J. Electron Spectrosc. Relat. Phenom,, 7 (1975)
331.
t Brundte, C.F., Robin, M.B., Kuebler, N.A. and Basch, H , J. Am. Chem. Soc., 94 (1972) I451.
u Fujisawa, S., Ohno, K., Masuda, S. and Harada, Y., J. Am. Chem. Soc., 108 (1986) 6505-6511.
v Anderson, C.P., Mamantov, G., Bull, W.E., Grimm, F.A., Carver, J.C and Carlson, T.A., Chem. Phys. Lett., 12
(1971) 137.
w Green, J.C., Green, M.L.H., Joachim, P.J., Orchard, A.F. and Turner, D.W., Philos. Trans. R. Soc. London, Ser.
A, 111 (1970) A268.
x Weast, R. C. (Ed.), CRC Handbook of Chemistry and Physics, 60th ed., CRC Press, Boca Raton, FL, 1980.
y Hellwege, K.H. and Hellwege, A.M. (Eds.), Molecular Constants from Microwave Spectroscopy (Landolt-Bornstein,
New Series II, Vol. 4), Springer-Verlag, Berlin, 1967.
z Cornforth, A.B., Frost, D.C., McDowell, C.A., Ragle, J.L. and Stonehouse, I.A., J. Chem. Phys., 54 (1971) 2651.
aa Heilbronner, E., Hornung, V. and Kloster-Jensen, E., Helv. Claim. Acta, 53 (1970) 331.
bb Baker, G.K., Lappert, M.F., Pedley, J.B., Sharp, G.J. and Westwood, N.P.C., J. Chem. Soc. Dalton Trans., 1765
(1975).
cc Frost, D.C., Herring, F.G., Katrib, A., McLean, R.A.N., Drake, J.E. and Westwood, N.P.C., Can. J. Chem., 49
(1971) 4033.
dd Bassett, P.J. and Lloyd, D.R., J. Chem. Soc. A, 641 (1971).
ee Pedley, J.B. and Rylance, G., Sussex-N.P.L. Computer Analysed Thermochemical Data: Organic and OrganometalI-
ic Compounds, Sussex University, 1977.