Unesco - Eolss Sample Chapters: Schrödinger Equation and Quantum Chemistry
Unesco - Eolss Sample Chapters: Schrödinger Equation and Quantum Chemistry
Unesco - Eolss Sample Chapters: Schrödinger Equation and Quantum Chemistry
Contents
TE SS
1. Introduction
2. The Schrödinger equation
S
2.1 Foundation of wave mechanics
R
2.2 Properties of the Schrödinger equation
AP L
2.3 Generalization of the Schrödinger equation for many-body systems
H O
2.4 General remarks on the Schrödinger equation
3. Quantum Chemistry
C E
3.1 Hartree-Fock theory and molecular orbitals
3.2 Correlated wavefunctions
E O–
Bibliography
Biographical Sketch
M ES
Summary
SA N
Aims, topics and methods of quantum chemistry are discussed, together with the
relationship between quantum mechanics and quantum chemistry.
U
Quantum mechanical methods developed for studying static and dynamic properties of
molecules are described.
1. Introduction
Quantum chemistry is the science that studies molecules and processes involving
molecules using methods of quantum mechanics. If we consider that under the word
“molecules” one can include the largest part of chemical substances, like e.g. organic
and inorganic complexes, clusters and macromolecules, and if we consider that quantum
mechanics is a formal theory well-established in its general principles and almost
To summarize briefly the novelty of quantum mechanics and its deep revision of
concepts of Newtonian macroscopic science, we can follow U.Fano’s analysis “…
common to all modern science is a primary reliance on reproducible experimental
observations. Consider, however, that when Galileo rolled marbles down an incline,
measuring their speed and acceleration, his observation process clearly had negligible
influence on the results, thus keeping in effect the observer fully separate from the
objects under study. Atomistic experiments sample instead the behavior of matter or
radiation from equivalent sources in alternative ways, drawing conclusions from
TE SS
combined statistical results…. These procedures contrast with those employed to
determine the mechanical state of macroscopic objects, by measuring the position and
S
velocity of each individual object through actions viewed as fully compatible with one
R
another. Measurements performed sequentially, even though on equivalent objects, are
AP L
instead viewed as independent. This contrast of basic measurement procedures provides
H O
common to light polarization and atomic phenomena; it is the earmark of quantum
mechanics’s novelty… Atomic experiments have actually taught us that science should
C E
deal only with inferences from observations done by humans, caution being of the
essence when extrapolating such evidence on dubious grounds…This essay casts
E O–
is reproducible on a statistical basis, yet wholly random when observing single events,
viewing randomness as primitive, rather than as resulting from manifold unobserved
circumstances, introduces a novel concept to physics. C) Quantum mechanics describes
SA N
to correlate observable parameters, the wave functions are a particular – if major – class
of probability amplitudes…”. This last remark sends us back to the Schrödinger
equation and formal theory on which quantum chemistry is based. This is the subject of
selection 2. Here we discuss, first, the relationship between quantum mechanics and
quantum chemistry and, second, the main topics of quantum chemistry.
The first clear statement on the relationship between quantum mechanics and chemistry
is contained in the Dirac famous sentence: “…The underlying physical laws necessary
for the mathematical theory of a large part of physics and the whole of chemistry are
thus completely known, and the difficulty is only that the exact application of these laws
leads to equations much too complicated to be soluble. It therefore becomes desirable
that approximate practical methods of applying quantum mechanics should be
developed, which can lead to an explanation of the main features of complex atomic
systems without too much computation…”. This sentence contains the two main lines of
development of quantum chemistry: on the one hand, that based on the idea that a
TE SS
and the like...”
S
Quantum chemistry is now a mature science, whose evolution has followed the two
R
main lines sketched above: development of new quantum mechanical methods and
AP L
computational algorithm for calculating chemical properties, and definition of new
H O
concepts, the so-called “patterns of understanding”, for the interpretation of chemical
problems. Since what is seen as the beginning of quantum chemistry, the Heitler-
C E
London paper on the hydrogen molecule in 1927, for about forty years, quantum
chemistry has been essentially a “theory of valence”, i.e. a theory that was intended to
E O–
answer the following fundamental questions: Why do molecules form? Why do the
same atoms form different compounds in definite but different proportions? What is the
origin of the stereochemistry of molecules? During these years a solid theory developed,
endowed with predicting capabilities and able to answer in depth the above questions.
PL C
proposed for solving these problems and general concepts introduced for interpreting
obtained results are discussed in section 3.
U
At the end of the fifties electronic computers became available for chemical applications.
Their rapid growth in speed and power has produced a real change in quantum
chemistry: on the one hand, the possibility of implementing quantum mechanical
methods already known, but too difficult or too sophisticated to be used without
computers on real problems; on the other hand, the possibility of applying these
methods to interpret, form first principles, the results of new spectroscopic techniques,
like those based on the use of laser sources, synchrotron radiation, strong magnetic
fields, collimated molecular beams, with the consequent development of new quantum
mechanical methods.
Quantum chemistry has also become an analytical tool suitable for applications in
different branches of chemistry and is used to obtain information bypassing direct
measurements: in physical chemistry, to predict thermodynamic properties of gases, to
calculate transition states and rate constants of chemical reactions, to obtain information
The quality of the results that can be obtained applying quantum chemistry to chemical
problems depends mainly on two elements: the modeling of the actual systems and the
methods used to study this model. As for the first point, one has to define the portion of
matter and the physical interactions that are important for the problem and construct a
model that takes into account the specific features of the quantities to be calculated and
the degree of accuracy achievable in the experiment that has to be interpreted. Thus, for
TE SS
example, an isolated molecule described as an ensemble of point charges of different
masses that interact via coulomb forces is a sufficiently accurate model for calculating
S
structural and spectroscopic properties of real molecules made up of atoms with nuclear
R
charges less then 20 and belonging to a low-density gas. For molecules with heavier
AP L
atoms, instead, accurate calculations of several properties require the inclusion of
H O
interaction due to the spin of the particles. Similarly, for molecules in condensed phases
or in solution, the environmental effects play an important role that has to be taken into
C E
account either via some mean-field approach or including explicitly in the calculation,
the molecular structure of a portion of the environment.
E O–
The second step is the choice of the quantum mechanical method appropriate for
studying the problem. This means, essentially, the choice of the approximations
necessary to approach the exact solutions of the Schrödinger equation and the capability
PL C
instead of measuring them directly, is relevant. First of all, in several cases the
interpretation of experimental data is dubious, since the experimental evidence can
rarely be directly referred to the property of interest. Quantum mechanical calculations,
instead, can be specifically organized in view of the properties of interest. Moreover,
there are performed on a model of the chemical system whose parameters can be
changed or adjusted in order to test their influence on the final results and the degree of
accuracy of the calculations can be increased to test the stability of the results.
Finally, one can observe that the continuous progress in experimental techniques has
made possible the direct measurement of quantities that previously were available only
form computation. Examples are electronic density that can be measured via diffraction
techniques, and time-evolution of simple chemical reactions, that can be followed using
time-resolved laser spectroscopy. The continuous feedback between theory and
experiment is a strong indication of the importance and vitality of quantum chemistry.
In this section we first summarize a way in which the Schrödinger equation can be
derived. This step represents the foundation of Wave Mechanics (Schrödinger, 1925-26)
as a specific version of quantum mechanics. The other equivalent formulation of
quantum mechanics, proposed and developed by Heisenberg, Born in Jordan (1925), is
known as Matrix Mechanics and is based on the association of a matrix with each
physical observable quantity. In section 3, we analyze properties of the Schrödinger
equation and the generalization of this equation for many-body systems.
TE SS
mathematical formulation of the analogy, suggested by de Broglie, between the
conditions for having stationary waves and the Bohr-Sommerfeld by de Broglie,
S
between the conditions for having stationary waves and the Bohr-Sommerfeld
R
conditions for the allowed motions in a system with f degrees of freedom. The latter can
AP L
be written as,
H O
∫ p dq = ni h (i = 1, 2,.... f ) (1)
C E
i i
non-negative integer number, h the Planck constant and the integral is taken over a
whole period of the coordinate. For a circular orbit of the electron in the hydrogen atom,
Eq. (1) gives the Bohr condition on the angular momentum L of the electron
PL C
∫ p ⋅ dq = ∫ pdq = 2π L = n h (2)
M ES
and, as a consequence, the terms of the Balmer formula for the discrete energy spectrum
of the hydrogen atom
SA N
2π 2 e4 m
U
En = − (n = 1, 2,...) (3)
h2n2
where –e and m are respectively the electron charge and mass. If one interprets the
quantity λ = h/p as the length of a wave associated with an electron of momentum
p (as suggested by de Broglie to interpret result of various experiments), the Bohr
condition (2) becomes
dq
∫ λ
=n (4)
that is exactly the condition for stationary waves in a vibrating string. This problem
requires the solution of the D’Alembert equation
∂2 1 ∂2
( − ) ψ ( x, t ) = 0 (5)
∂x 2 υ 2 ∂t 2
d2 ω 2π
u ( x ) + k 2u ( x ) = 0 ; k = = (7)
dx 2
υ λ
TE SS
and the boundary conditions for a string of length l, fixed at the ends, give the
S
eigenvalues of the problem
R
π 2π
AP L
k = kn = n = (8)
H O l λn
C E
Following the de Broglie hypothesis that the particles too possess the dual character of
light: as an electromagnetic wave is associated with each photon, so a material particle
E O–
E = ω; p = k (9)
M ES
ω E
k= ; p= (10)
c c
U
where c is the velocity of light, while for particles the relativistic invariance requires
mυ
p = ;
1 − υ2 c2
(11)
mc 2
E = = c p 2 + m2c 2
1− υ c2 2
ω m2c 2
= k2 + 2
(12)
c
TE SS
particle with a wave packet, superposition of plane waves
S
ψ (r , t ) = ∫ | f (k ) | ei[ k .r −ω ( k )t +α ( k )] d k
R
(13)
AP L
H O
localized in a volume ΔV around a centre r0 , that moves with a well-defined velocity υ g ,
C E
called group velocity. These quantities can be defined as follows
1
E O–
ΔV ∼ ; r0 =
ΔK (14)
t υ g − ∇ k α |k = k 0 ; υ g = ∇ k ω |k = k 0
PL C
Where ΔK is a volume inside which | f (k ) | is appreciably different from zero and takes
M ES
its maximum at k 0 . To compare the motion of the wave packet with that of a classical
particle, as suggested by the analogy between geometrical optics and classical
mechanics, it is necessary to establish the identity between rays of frequency ω and
SA N
variational principle, the exact correspondence between the two quantities requires that
between the two variational principles. In particular, trajectories of a classical particle of
momentum p can be obtained from the principle of least action
M2
δI = δ ∫
M1
p . dr = 0 (15)
The actual trajectory is that which makes stationary the integral I, calculated between
the two fixed points M 1 and M 2 . On the other hand, rays of geometrical optics can be
obtained from the Fermat principle that, for a given frequency ω and wave vector k
requires
M2
δJ = δ ∫ M1
k . dr = 0 (16)
The actual ray is the optical path that joins M 1 and M 2 and makes stationary the integral
J. Because of the formal analogy between the two variational principles, the exact
correspondence between rays and trajectories is obtained if p = β k . The constant of
proportionality β is determined by requiring the identity between the velocity or the
classical particle: υ = ∇ p E and group velocity υ g of the wave packet. The identity
requires β = and brings us back to Eq. (9) in a completely general way.
If we assume, now, that the wavefunction ψ (r , t ) completely defines the dynamical state
of the particle, in the sense that all its properties, at any given time t , can be deduced
from the knowledge ofψ (r , t ) , we are led to the central problem of the theory, that is the
TE SS
determination of the wavefunction at any instant t from the knowledge of ψ at a given
S
initial time t0 . The solution of this problem requires the knowledge of the equation of
propagation ofψ , i.e. of the wave equation.
R
AP L
H O
As one can see from Eq. (11, 12, 13) the wave packet fulfils the scalar, relativistic
equation
C E
⎛ 1 ∂2 ⎞ m2c 2
⎜ Δ − ⎟ ψ ( r , t ) = ψ (r , t ) ;
⎝ c 2 ∂t 2 ⎠ 2
E O–
(17)
∂2 ∂2 ∂2
Δ= 2 + 2 + 2
∂x ∂y ∂z
PL C
∂ ∂
i ↔ E; −i ↔ pα (18)
∂t ∂xα
SA N
U
Schrödinger initially proposed Eq. (17) as the appropriate wave equation. This equation,
however, which is known as the Klein-Gordon equation and plays an important role in
relativistic quantum mechanics, presents a few relevant difficulties in this context. It can
be observed, for example, that Eq.(17) is second order in time and, therefore, to define
the latter evolution of the wavefunction one has to specify not onlyψ (r , t0 ) ,but also its
time derivative at t0 . This fact contradicts the basic assumption that ψ (r , t ) can be
derived only from the knowledge ofψ (r , t0 ) . Because of these and other difficulties, Eq.
(17) cannot be adopted as the appropriate wave equation without reinterpreting the
physical meaning of ψ inside a more general theory.
As a matter of fact, indeed, the assumption that the dynamical state of a single particle is
represented by a single wavefunction is fully justified only in the non-relativistic limit,
where the law of conservation of the number of particles is satisfied. Assuming the
p2
E = mc 2 + + = mc 2 + Ek +
2m
E mc 2 k2 mc 2
ω = = + + = + ω′ + (19)
c 2m
TE SS
( )
mc 2 t
ψ r , t = e−i ψ ′ (r , t ) =
S
3
(20)
∫ | f (k ) |
− i mc t i [ k .r −ω ′ ( k ) t +α ( k )]
e e dk
R
AP L
H O
And using this expression in Eq.(17), one finds that ψ ′(r , t ) fulfils the following
C E
equation
2m ∂
) ψ ′ (r , t ) = 0
E O–
(Δ + i (21)
∂t
We see that, in the non-relativistic limit, the energy is shifted by a constant and the
PL C
time and does not present the drawbacks of the Klein-Gordon equation. Therefore, it can
be chosen as the appropriate wave equation. This equation, indeed, is the time-
dependent Schrödinger equation for a free particle of mass m.
SA N
The Generalization of Eq. (21) for a particle moving with non-relativistic velocities in a
U
p2
E = + V ( r ) = H ( p, r ) (22)
2m
∂
i ψ (r , t ) E ψ (r , t ) ;
∂t (23)
− 2 Δ ψ (r , t ) p 2 ψ (r , t )
⎛ ∂ 2
⎞
⎜i + Δ − V ⎟ ψ (r , t ) 0 (24)
⎝ ∂t 2m ⎠
This equation is approximately satisfied by a wave packet of the type (13) under the
conditions of validity of geometrical optics, but we are led, in a very natural way, to
postulate that the wavefunction ψ (r , t ) fulfils in general the equation
TE SS
S
∂
i ψ (r , t ) =
R
∂t
AP L
(25)
⎡ 2
⎤
−
H O
⎢ 2m Δ − V ( r ) ⎥ ψ ( r , t ) = H ψ ( r , t )
ˆ
⎣ ⎦
C E
That is the time-dependent Schrödinger equation for a particle of mass m moving in the
E O–
scalar potential V (r ) . We can also postulate that this equation applies generally to all
motions, independently of the energy of the system. The main feature of this equation is
the presence of an operator H , that has the same formal structure as the classical
PL C
∂
pα → − i (26)
∂xα
SA N
as in Eq.(18). One immediately sees that, if the Hamiltonian does not depend explicitly
U
on time, i.e. the system is conservative Eq. (25) admits solutions of the type
ψ (r , t ) = u (r ) e −iEt / (27)
that represents periodic time-dependent states of definite E (stationary states), with the
spatial motion described by a wavefunction solution of the equation
2
H u (r ) = E u (r ) ; Hˆ = − Δ + V (r ) (28)
2m
-
-
Bibliography
TE SS
Dirac P.A.M. (1958) The principles of Quantum Mechanics, Oxford University press. [This is a classic
book by a father of quantum mechanics that presents the basic principles and the mathematical structure
S
of the quantum theory]
R
Fano U. (1995) Quantum Mechanics: Then and Since, La Rivista del Nuovo Cimento, vol.11, Editrice
AP L
Compositori, Bologna. [A critical analysis of basic concepts and definitions in quantum mechanics]
H O
Levine I.N. (1991) Quantum Chemistry, prentice-Hall International, New Jersey. [A modern textbook that
treats in a clear and complete way the main topics of quantum chemistry]
C E
Messiah A. (1970) Quantum Mechanics, North-Holland Publishing Company, Amsterdam. [A classic
textbook on quantum mechanics that illustrates basic principles, explains mathematical structure and
E O–
provides a rigorous formulation of several quantum mechanical methods for studying molecules].
M ES
Biographical Sketch
Renato Colle was born in Belluno, Italy, on 9 October 1949, studied at Scuola Normale Superiore (SNS)
of Pisa during 1973-75 for Ph.D., and worked there as Assistant Professor of Chemistry during 1975-81.
SA N
In 1978 and 1980 respectively he visited Technisch-Chemisches Labor der ETH of Zurich and the
Department of Physics, University of Chicago, as a Research Associate. During 1981-90 he was
U
Associate Professor of Chemical Physics at the SNS. In 1985 he was a Visiting Professor at the
Department of Physics, University of Chicago. Since 1991 he is a Full Professor of Chimica Fisica dei
Materiali Solidi" at the University of Bologna and was Scientific Director of a Joint Research Project of
the European Community during1994-98.