Optics and Photonics J ournal, 2012, 2, 332-337

http://dx.doi.org/10.4236/opj.2012.24041 Published Online December 2012 (http://www.SciRP.org/journal/opj)

Simple Method of the Formation of the Hamiltonian
Matrix for Some Schrdinger Equations Describing the
Molecules with Large Amplitude Motions
George . Pitsevich, Alex E. Malevich
Belarusian State University, insk, Belarus
Email: pitsevich@bsu.by

Received September 8, 2012; revised October 7, 2012; accepted October 18, 2012
ABSTRACT
A simple approach to the formation of a Hamiltonian matrix for some Schrdinger equations describing the molecules
with large amplitude motions has been proposed. The algorithm involving one or several variables has been concretely
defined for the basis functions represented by Fourier series and orthogonal polynomials, taking Hermitian polynomials
as an example.

Keywords: Schrdinger Equation; Large Amplitude Motions; Hamiltonian Matrix
1. Introduction
Algebraic approaches to solving of Schrdinger equa-
tions have several advantages compared to other methods.
Provided the basis functions used for expansion of the
wave functions and the potential energy of a system are
adequately selected, the Schrdinger equation takes the
matrix form, in the end its solution being reduced to the
derivation of eigenfunctions and eigenvalues for the
Hamiltonian matrix. When studying the molecules whose
variables are changing with a large amplitude, the Ham-
iltonian matrix derivation is a nontrivial problem. This
work presents an algorithm to form the Hamiltonian ma-
trix for some Schrdinger equations describing mole-
cules and molecular systems with several variables of
this type. Such equations may be illustrated by the fol-
lowing:
( )
2
2
d
d
F U E

+
+ = + (1)
( ) ( )
( ) ( )
( )
2 2
2 2
, ,
, ,
,
s t s t
A B U s t s t
s t
E s t
c + c +
+ +
c c
= +
(2)
( ) ( ) ( )
( ) ( ) ( )
2 2 2
2 2 2
, , , , , ,
, , , , , ,
x y z x y z x y z
F
x y z
U x y z x y z E x y z
| | c + c + c +
+ +
|
|
c c c
\ .
+ + = +
(3)
In all cases it is assumed that with a change in the vari-
ables the kinematic parameters remain invariable. This
may be attained by an adequate selection of a coordinate
system strongly related to the molecule, both in the case
of symmetric [1-4] and low-symmetry [5-9] molecules.
Equation (1), in particular, describes internal vibrations
in a molecule of methanol, taking the effective vibra-
tional constant as F [10-12]. In some cases an invariable
character of the molecular kinematic parameters for large
amplitude motions may be considered as a physically
valid approximation. Specifically, Equation (3) may be
used for the description of motion of a hydrogen atom in
the process of hydrogen bonding if we neglect motion of
the oxygen atoms, the amplitude of which is in fact con-
siderably smaller than that of H motion as a mass ratio of
these atoms is 1:16. This paper presents quick ap-
proaches to construct the Hamiltonian matrix for some
basis functions.
2. Using of Fourier Series
Since the use of Fourier series for solving of equations of
the form given in (1) is frequently described in the lit-
erature and in some works the formation algorithm for
the Hamiltonian matrix is given in detail, we begin our
analysis from Equation (2). Let the potential energy be
given in the form:
( )
( )
,
i
, ,
, e ; ,
a b
ks lt
kl
k l a b
U s t u a b
+
=
= e

(4)
Then a wave function is derived as:
Copyright 2012 SciRes. OPJ
G. . PITSEVICH, A. E. MALEVICH 333
( )
( i
,
, e
ns mt
nm
n m
s t b

+
=
+ =

)
=
)
(5)
Substituting (4) and (5) into (2), we obtain:
( )
( )
( ) ( ) ( )
i 2 2
,
,
i
, , ,
e
e 0
ns mt
nm
n m
a b
n k s m l t
kl nm
n m k l a b
n A m B E b
u b

+
=

+ + +
= =
+
+

(6)
Next we define coefficients for the exponential
. In the second term the following condition
must be fulfilled:
( i
e
n s m t ' ' +
; n k n k n n
m l m l m m
' ' + = =
' ' + = =
(7)
Instead of (6), we have:
( )
( )
( )
i 2 2
,
i
,
, , ,
e
e 0
n s m t
n m
a b
n s m t
n n m m nm
n m n n m m a b
n A m B E b
u b
' ' +
' '

' ' +
' '
' ' = =
' ' +
+

=
N
(8)
Then we construct the finite matrix with the dimen-
sions . This means that n and
m are varying within the limits from to c per unity.
From (8) we derive:
( ) ( )
2 2
2 1 2 1 ; c c c + + e
c
( )
( )
( )
i 2 2
,
i
,
, , ,
e
e 0
n s m t
n m
a b c
n s m t
n n m m nm
n m c n n m m a b
n A m B E b
u b
' ' +
' '
' ' +
' '
' ' = =
' ' +
+ =

(9)
Now we take (9) as a matrix equation of the form
ij j j j
H b E b = , where
j
b column vector that, ac-
cording to (5), gives the wave function corresponding to
the energy
j
E . It is clear that a pair of the indices
( ) , n m ' ' numbers rows of the Hamiltonian matrix and a
pair of the indices ( ) , n m
b
its columns. Next, to derive
the Hamiltonian matrix from (9), first we have to fix an
order of the coefficients
nm
in the column vector of the
wave function defined by Equation (5). For example, if
, the transposed column vector may be of the form: 1 c =
1, 1 1,0 1,1 0, 1 0,0 0,1 1, 1 1,0 1,1
; ; ; ; ; ; ; ; b b b b b b b b b b

=

(10)
Let us assume that in the same order from top to bot-
tom there is a change in the index pair ( ) , n m ' '
, a b
num-
bering rows of the Hamiltonian matrix. Then a matrix
element of H is numbered by two index pairs,
( ) ( ) , , , n m n m
. Considering that usually , for the
diagonal element
H
' '
c
( ) ; n n m m ' ' = = we can write:
( ) ( )
2 2
00 , , , n m n m
H n A m B u E
' ' ' '
' ' = + + (11)
and for nondiagional elements we can write:
( ) ( ) , , , ,
if and
n n m m n m n m
H u
n n a m m b
' ' ' '
=
' ' s s
(12)
( ) ( ) , , ,
0
if or
n m n m
H
n n a m m b
' '
=
' '
(13)
Numbering matrix elements of H by the ordinary
indices ( ) , i j
)
2
each of which is varying from 1 to
, we should establish for each of them a one-
to-one correspondence to a pair of numbers by the prin-
ciple:
(2 1 c +
( ) , i n
i i
m ' ' ;
( )
,
j j
j n m
3 i
. Specifically, in the
case given by (10) for = we have ;
3 3
1 n' = 1 m' = ;
and for 6 j = we have
6
;
6
. Now an algo-
rithm for the formation of the matrix
0 n = 1 m =
H takes the fol-
lowing form:
2 2
00 ii i i
H n A m B u E ' ' = + + (14)
,
if and
i j i j
ij n n m m
i j i j
H u
n n a m m b
' '
=
' ' s s
(15)
0 if or
ij i j i j
H n n a m m ' ' = b (16)
Let us write the Hamiltonian matrix in the explicit
form with the use of (14 - 16) for . Besides, we
assume that the index order is determined by the relation
of (10), and
1 c =
1 a b = = . Then we have:

0, 1 1,0 1, 1 00
0,1 0, 1 1,1 1,0 1, 1 00
0,1 1,1 1,0 00
1,0 1, 1 0, 1 1,0 1, 1 00
1,1 1,0 1, 1 0,1 0, 1 1,1 1,0 1, 1 00
1,1 1,0 0,1 1,1 1,0 00
1,0 1
0 0 0 0
0 0 0
0 0 0 0
0 0
0 0 0
0 0 0
u u u A B u
u u u u u A u
u u u A B u
u u u u u B u
u u u u u u u u u
u u u u u B u
u u






+ +
+
+ +
+
+
, 1 0, 1 00
1,1 1,0 1, 1 0,1 0, 1 00
1,1 1,0 0,1 00
0 0
0 0 0
0 0 0 0 0
u A B u
u u u u u A u
u u u
0
0
0

A B u


+ +
+
+ +


Next we consider the case of three variables. Let the Schrdinger equation be of the form:
Copyright 2012 SciRes. OPJ
G. . PITSEVICH, A. E. MALEVICH 334
( ) ( ) ( )
( ) ( ) ( )
2 2 2
2 2
, , , , , ,
, , , , , ,
2
s t r s t r s t r
A B C
s t
U s t r s t r E s t r
c + c + c +

c c c
+ + = +
r
)
(17)
Then we define an algorithm to form the Hamiltonian
matrix when using three-dimensional Fourier series. Let
the potential energy be given as:
( )
( )
, ,
i
, , , ,
, , e ;
, ,
a b c
hs kt lr
hkl
h k l a b c
U s t r u
a b c N
+ +
=
=
e

(18)
A wave function takes the form:
( )
( i
, ,
, , e
ns mt qr
nmq
n m q
s t r b

+ +
=
+ =

(19)
Substituting (18) and (19) into (17), we obtain:
( )
( )
( ) ( ) ( ) ( )
i 2 2 2
, ,
, ,
i
, , , , , ,
e
e 0
ns mt qr
nmq
n m q
a b c
n h s m k t l q r
hkl nmq
n m q h k l a b c
n A m B q C E b
u b

+ +
=

+ + + + +
= =
+ +
+ =

(20)
Let us find coefficients for the exponential
) ( i
e
n s m t q r ' ' ' + +
.
The following condition must be fulfilled:
;
;
;
n h n h n n
m k m k m m
l q l q l l
' ' + = =
' ' + = =
' ' + = =
(21)
Instead of (20), we have:
( )
( )
(
i 2 2 2
, ,
i
, ,
, , , , , ,
e
e
0
n s m t q r
n m q
a b c
n s m t q r
n n m m l l nmq
n m q n n m m l l a b c
n A m B q C E b
u b
' ' ' + +
' ' '

' ' ' + +

' ' '
' ' ' = =
' ' ' + +
+
=

)
N
)
(22)
We construct the finite matrix with the dimensions
, i.e. n, m, and q are varying
within the limits from to d per unity. From Equa-
tion (22) we get:
( ) ( )
3 3
2 1 2 1 ; d d d + + e
d
( )
( )
(
i 2 2 2
, ,
i
, ,
, , , , , ,
e
e
0
n s m t q r
n m q
a b c d
n s m t l r
n n m m l l nmq
n m q d n n m m l l a b c
n A m B q C E b
u b
' ' ' + +
' ' '
' ' ' + +
' ' '
' ' ' = =
' ' ' + +
+
=

(23)
Now three indices ( ) , , n m l ' ' ' number rows and three
indices ( ) , , n m l
1 c =
number columns of the Hamiltonian
matrix. To derive a Hamiltonian matrix from (23), we
again fix an order of the coefficients
nmq
in the column
vector for the wave function defined by (19). For exam-
ple, if , the transposed column vector may be of the
form:
b
1, 1, 1 1, 1,0 1, 1,1 1,0, 1 1,0,0
1,0,1 1,1, 1 1,1,0 1,1,1 0, 1, 1 0, 1,0
0, 1,1 0,0, 1 0,0,0 0,0,1 0,1, 1 0,1,0 0,1,1
1, 1, 1 1, 1,0 1, 1,1 1,0, 1 1,0,0 1,0,1
1,
; ; ; ; ;
; ; ; ; ; ;
; ; ; ; ; ;
; ; ; ; ; ;
b b b b b b
b b b b b b
b b b b b b b
b b b b b b
b




=

;
1, 1 1,1,0 1,1,1
; ; b b

(24)
We assume that a change of three indices , , n m q ' ' ' ,
numbering rows for the Hamiltonian matrix is in the
same order from top to bottom. Then a matrix ele-
ment of H is numbered by two pairs of three indices,
( ) , ,
. Considering that, as previously, we
have , then for the diagonal elements
( , , , n m q n m q
H
' ' '
d
)
, , a b c
( ) q ' ; ; m q n n m ' ' = = = we can write:
( ) ( )
2 2 2
000 , , , , , n m q n m q
H n A m B q C u
' ' ' ' ' '
' ' ' E = + + + (25)
And for nondiagonal elements we can write:
( ) ( ) , , , , , , ,
if ; and
n n m m q q n m q n m q
H u
n n a m m b q q c
' ' ' ' ' '
=
' ' ' s s s
(26)
( ) ( ) , , , , ,
0 if
or or
n m q n m q
H n n a
m m b q q c
' ' '
' =
' '


(27)
when numbering the matrix elements of H by the ordi-
nary indices ( ) , i j each of which is varying from 1 to
(
3
2 1 c ) + , we should establish for each of them a
one-to-one correspondence to three numbers by the prin-
ciple: ( ) q , ,
i i
i n m
i
' ' ' ;
( )
, ,
j j j
j n m q
3 i =
. Specifically,
in the case given by (24) for we have
3
1 n' = ;
3
1 m' = ;
3
1 q' = , and for we have
19
19 j = 1 n = ;
19
1 m = ;
19
1 q = . Now an algorithm to form the ma-
trix H takes the form:
2 2 2
000 ii i i i
H n A m B q C u E ' ' ' = + + + (28)
, ,
if ;
and
i j i j i j
ij n n m m q q i j
i j i j
H u n n
m m b q q c
' ' '
' a = s
' ' s s
(29)
0 if
or or
ij i j
i j i j
H n n a
m m b q q c
' =
' '

(30)
3. Using of Orthogonal Polynomials
Let us consider the Schrdinger equation with one vari-
able (31), taking orthogonal Hermitian polynomials
n
H
as an example.
( )
( ) ( ) ( )
2
2
d
d
x
R U x x E
x
+
+ + = + x (31)
Let the potential energy be given as:
( ) ( )
0
m
k k
k
U x u H x
=
=

(32)
Copyright 2012 SciRes. OPJ
G. . PITSEVICH, A. E. MALEVICH 335
And we are looking for a wave function of the form:
( ) ( )
2
1
2
0
e
x
n n
n
x b H x

=
+ =

(33)
We substitute (32) and (33) into (31):
( ) ( )
( )
( ) ( )
( )
( ) ( )
0
2
0
2
0
0 0
1
2
1
4
0
n n
n
n n
n
n n
n
m
k n n k
n k
R n E b H x
Rn n b H x
R
b H x
u b H x H x

+
=

= =
+

=
(34)
Using the orthogonality of Hermitian polynomials, we
can write:
( ) ( ) ( )
( ) ( ) ( )
2
, ,
,2
1
2
, ,
;
e d
n k
n k l n k l
l n k
x
l n k n k l
H x H x c H x
c H x H x H x x
+
=

=
=

}
(35)
As a result, Equation (34) takes the form:
( ) ( )
( ) ( ) ( )
( ) ( )
, ,
2
0 0
2
0 0 0 ,2
1
1
2
0
4
l n k k n l
n n n n
n n
m n k
n n
n n k l n k
c u b H
R n E b H x Rn n b H x
R
b H x x

= =
+
+
= = = =
+
+


=
(36)
Taking the coefficients for
n
H
'
, we construct a matrix
with the dimensions ( ) ( ) 1 h +1
,
h + . In the second term
of Equation (36) we must assume
2 2 n n n n ' ' = = +
in the third term we assume ,
and in the fourth . Instead of (36), we get:
2 2 n n n n ' ' + = =
l n' =
( ) ( )
( )
( )( ) ( )
( ) ( )
2
2
0
1
2
2 1
0
4
n n
n n
h
n n n nk n n k
n k
R n E b H x
R n n b H x
R
b H x c H x b u
' '
' ' +
' ' ' '
=
' +
' ' + +
+

=
(37)
As previously, the index numbers rows of the
Hamiltonian matrix, whereas the index n numbers its
columns. Let an order of indices in the column vector of
the wave function be so that a form of the transposed
vector is given by:
n'
0 1
; ;
h
b b b b =

(38)
In a similar way we will number rows of the Hamilto-
nian matrix from top to bottom from 0 to h per unity.
According to Equation (37), at the first stage we can fill
the Hamiltonian matrix with the elements existing for
representation of the potential energy in the form
by the following principle:
n nk k
c u
'
n n n nk k
k
H c u
' '
=

(39)
Summation in (39) is over all the existing indices k for
the specified index pair ( ) , n n ' . Next, to every diagonal
element
n n
H
' '
we add
( )
1
2
n R ' + and to every element
of the diagonal, parallel to the main diagonal and posi-
tioned above it as a next nearest
( )
, 2 n n
H
' '+
, we add
( )( ) 2 1 n n ' ' + + R
)
. And in the case of a similar diagonal
positioned as a next nearest below
(
we add
, 2 n n
H
' '
1
4
R . So, diagonal elements take the form:
( )
1
2
n n n n k k
k
H n R c u
' ' ' '
' = + +

(40)
Nondiagonal elements are of the form:
( )( )
2 ,
2 1
n n n n k k
k
2,
H n n R c
' ' ' ' + +
' ' = + + +

u (41)
2 ,
1
4
n n n n k k
k
2,
H R c u
' ' ' '
= +

(42)
The remaining nondiagonal elements are as (39). Us-
ing the ordinary indices varying from 1 to , i j 1 h + ,
we can rewrite this algorithm as:
( ) 1, 1,
1
2
ii i i k k
k
H i R c u

= +

(43)
( )
, 2 1, 1,
1
i i i i k k
k
H i iR c u
+
= + +
+
(44)
, 2 1, 3,
1
4
i i i i k k
k
H R c u

= +

(45)
1, 1, ij i j k k
k
H c u

=

(46)
Finally, we consider Equation (3), trying to construct
the Hamiltonian matrix with the use of Hermitian poly-
nomials as basis functions. Let the potential energy be
represented as:
( ) ( ) ( ) ( )
, ,
, , 0
, , ;
, ,
a b c
klm k l m
k l m
U x y z u H x H y H z
a b c N
=
=
e

(47)
A wave function is derived as follows:
( ) ( ) ( ) ( )
2
2
0
2 2 2 2
, , e ;
r
nts n t s
nts
x y z b H x H y H z
r x y z

=
+ =
= + +

(48)
Substituting (47) and (48) into (3), we obtain:
Copyright 2012 SciRes. OPJ
G. . PITSEVICH, A. E. MALEVICH 336
( ) ( )
( ) ( ) ( )
( ) ( ) ( )
( ) ( ) ( ) ( )
( ) ( ) ( ) ( )
( ) ( ) ( )
( ) ( ) ( )
( ) ( ) ( )
0
2
0
2
0
2
0
2
0
2
0
2
0
3
2
( 1)
1
1
4
4
4
nts n t s
nts
nts n t s
nts
nts n t s
nts
nts n t s
nts
nts n t s
nts
nts n t s
nts
nts n t s
nts
fnk ht
R n t s E b H x H y H z
Rn n b H x H y H z
Rt t b H x H y H z
Rs s b H x H y H z
R
b H x H y H z
R
b H x H y H z
R
b H x H y H z
c c

+
=

+
=

+
=
+ + +

( ) ( ) ( )
0 0
0
abc
l rsm nts klm f h r
nts klm fhr
c b u H x H y H z

= =
=


(49)
Suppose that we need to construct a matrix with the
dimensions . We determine coeffi- ( ) ( )
3
1 d d + +
( )
3
1
cients for the factor ( ) ( ) :
n t s
H x H y H z
' ' '

( ) ( )
( ) ( ) ( )
( )( ) ( ) ( ) ( )
( )( ) ( ) ( ) ( )
( )( ) ( ) ( ) ( )
( ) ( ) ( )
( ) ( ) ( )
2, ,
, 2,
, , 2
2, ,
, 2,
, , 2
3
2
2 1
2 1
2 1 )
4
4
4
n t s n t s
n t s n t s
n t s n t s
n t s n t s
n t s n t s
n t s n t s
n t s
R n t s E b H x H y H z
R n n b H x H y H z
R t t b H x H y H z
R s s b H x H y H z
R
b H x H y H z
R
b H x H y H z
R
b H
' ' ' ' ' '
' ' ' ' ' ' +
' ' ' ' ' ' +
' ' ' ' ' ' +
' ' ' ' ' '
' ' ' ' ' '
' ' '
' ' ' + + +
' ' + +
' ' + +
' ' + +

( ) ( ) ( )
( ) ( ) ( )
0
0
n t s
d
n nk t tl s sm nts klm n t s
nts klm
x H y H z
c c c b u H x H y H z
' ' '
' ' ' ' ' '
=
+ =

(50)
This expression may be rewritten as follows:
( ) ( )
( )
( )( ) ( )( )
( )( )
2, , , 2,
, , 2 2, , , 2,
, , 2
0
3
2
2 1 2 1
2 1
4 4
0
4
n t s n
n t s n t s
n t s n t s n t s
d
n t s n nk t tl s sm nts klm
nts klm
R n t s E b H x
R n n b R t t b
R R
R s s b b b
R
b c c c b u
' ' ' '
' ' ' ' ' ' + +
' ' ' ' ' ' ' ' ' +
' ' ' ' ' '
=
' ' ' + + +
' ' ' ' + + + +
' ' + +
+ =

(51)
We fix an order of the coefficients
nts
in the column
vector of the wave function defined by Equation (48).
For example, if , the transposed column vector
may be of the form:
b
2 d =
0,0,0 0,0,1 0,0,2 0,1,0 0,1,1 0,1,2 0,2,0 0,2,1
0,2,2 1,0,0 1,0,1 1,0,2 1,1,0 1,1,1 1,1,2 1,2,0 1,2,1 1,2,2
2,0,0 2,0,1 2,0,2 2,1,0 2,1,1 2,1,2 2,2,0 2,2,1 2,2,2
; ; ; ; ; ; ; ;
; ; ; ; ; ; ; ; ;
; ; ; ; ; ; ; ;
b b b b b b b b b
b b b b b b b b b b
b b b b b b b b b
=

; (52)
Let us assume that a change in three indices ( ) , , n t s ' ' '
numbering rows of the Hamiltonian matrix is in the same
order from top to bottom. The matrix element H is
numbered by a pair of three indices
( ) ( ) , , , , , n t s n t s
. As
earlier, first we can fill the Hamiltonian matrix with the
existing elements representing the potential energy of the
form by the following principle:
H
' ' '
n nk t tl s sm klm
c c c u
' ' '
( ) ( ) , , , , , n nk t tl s sm klm n t s n t s
klm
H c c c u
' ' ' ' ' '
=

(53)
Summation in Equation (53) is performed over all the
existing triples ( ) , , k l m for the pair of the specified
triples ( ) , , n t s ' ' ' and ( ) , , n t s . For the main diagonal
( ) ( , , , , , n t s n t s
H
) ' ' ' ' ' '
we must add
( )
3
2
n t s R ' ' ' + + +
)
. To the
nondiagonal elements of the form
( ) ( ) , , , 2, ,
;
n t s n t s
H
' ' ' ' ' ' +
( ) ( , , , 2, , n t s n t s
H
' ' ' ' ' ' +
, and we must add
( ) ( , , , , , 2 n t s n t s
H
' ' ' ' ' '+ )
( )( ) 2 1 R n n ' ' ; + + ( )( ) 2 1 R t t ' ' + + , and
( )( )
' 2 R s s 1 ' + + . Finally, to the diagonal elements of
the form
( ) , , , n t s n
H
( ) 2, ,
;
t s ' ' '
H
' ' ' ( ) ( , , , , 2, n t s n t s
H
' ' ' ' ' ' )
)
, and
( ) ( , , , , , 2 n t s n t s ' ' ' ' ' '
we must add
1
R . Thus, we have:
4
( ) ( ) ( ) , , , , ,
3
2
n t s n t s
n n k t t l s s m klm
klm
H n t s R
c c c u
' ' ' ' ' '
' ' ' ' ' '
' ' ' = + + +
+

(54)
( ) ( )
( )( )
, 2, , , , 2, ,
2 1
n n k t t l s s m klm n t s n t s
klm
H c c c u
R n n
' ' ' ' ' ' + ' ' ' ' ' ' +
=
' ' + +

(55)
( ) ( )
( )( )
, 2, , , , , 2,
2 1
n n k t t l s s m klm n t s n t s
klm
H c c c u
R t t
' ' ' ' ' ' + ' ' ' ' ' ' +
=
' ' + +

(56)
( ) ( )
( )( )
, 2, , , , , , 2
2 1
n n k t t l s s m klm n t s n t s
klm
H c c c u
R s s
' ' ' ' ' '+ ' ' ' ' ' '+
=
' ' + +

(57)
( ) ( ) 2, , , , , 2, ,
1
4
n n k t t l s s m klm n t s n t s
klm
H c c c u
' ' ' ' ' ' ' ' ' ' ' '
=

R (58)
( ) ( ) , 2, , , , , 2,
1
4
n n k t t l s s m klm n t s n t s
klm
H c c c u R
' ' ' ' ' ' ' ' ' ' ' '
=

(59)
( ) ( ) , , , , , 2
1
4
n n k t t l s s m klm n t s n t s
klm
H c c c u R
' ' ' ' ' ' ' ' ' ' ' '
=

(60)
In other cases, we have (53). Now numbering the ma-
trix elements of H by the ordinary indices ( ) , i j each of
Copyright 2012 SciRes. OPJ
G. . PITSEVICH, A. E. MALEVICH
Copyright 2012 SciRes. OPJ
337
) which is varying from 1 to , we have to estab-
lish for them a one-to-one correspondence to three num-
bers according to the principle:
(
3
2 1 d +
( ) , ,
i i i
i n t s ' ' ' ;
(
, ,
j j j
)
j n t s
5 i =
17 j =
. Specifically, in the case given by (52)
for we have ; ; , and for
we have
17
;
17
5
0 n' =
1 = t
5
1 t' =
2
5
1 s' =
n = ;
17
. Then an al-
gorithm to construct the matrix under condition of
(52) takes the following form:
1 = s
ij
H
( ii i i
H n t ' ' = +
, 18 ,
1 9;
i i
i i n n
klm
H c
i
+
=

, 6
1 3; 10 1
i i
i i n n k
klm
H c
i
' ' +
=

, 2
1, 4, 7,10,13
i i
i i n n k
klm
H c
i
' ' +
=

)
3
2
i
s' + +
18
,
i i
k t t l s
+
' '
6
, ,
2; 19 21
i i
t t l s s
c
+
' ' '

2
,
16,19, 22
i i i i
t t l s s
c c
+
' ' ' '
;
i i
s s m klm
c u
' '
( )( ) 2 1 ;
i
n n ' ' +
( )( ) 2 1 ; t t ' ' + +
( )( ) 2 1 ; s s ' ' +
i i i i
n n k t t l
klm
c c
' ' ' '
+

m klm i
c u =
klm
u R =
25;
klm
u R =
R
i i
s ' '
;
m
c
,
,
m
(61)
c
' '
R +
+
(62)
'
(63)
,
(64)
18
,
i i
n k t t l
c

' '
=
18, ,
i i
i i n
klm
H c
' ' +
1
4
klm
u ; 1 9; R i =
i
s'
i
s m
c
'
(65)
6
, ,
2; 19 21
i i i
k t t l s 6,
1 3; 10 1
i i
i i n n
klm
1
4
;
i
s m klm
; H c c c

' ' '

u
'
= R
i
' ' +
=

(66)
2
16,19, 22
i i i i
t t l s s 2,
1, 4, 7,10,13
i i
i i n n k
klm
1
4
, , 25;
m klm
; H c c c

' ' ' '
i j i j
t t l s s m
u = R
i
' ' +
=

,
i j
i j n n k
klm
(67)
klm
H c c =

c
' ' '
u (68)
4. Conclusion
In this way we have derived analytical expressions for
elements of the Hamiltonian matrix describing the mole-
cules characterized by motions with a large amplitude.
The cases when the wave functions and potential energy
are represented by Fourier series and orthogonal poly-
nomials have been considered in detail taking Hermitian
polynomials as an example. Some specific types of
Schrdinger equations with a single variable or several
variables have been treated.
REFERENCES
[1] J. T. Hougen, A Group-Theoretical Treatment of Elec-
tronic, Vibrational, Torsional and Rotational Motions in
the Dimethylacetylene Molecule, Canadian Journal of
Physics, Vol. 42, No. 10, 1964, pp. 1920-1937.
doi:10.1139/p64-182
[2] N. Moazzen-Ahmadi, H. P. Gush, M. Halpern, H. Jagan-
nath, A. Leung and I. Ozier, The Torsional Spectrum of
CH
3
CH
3
, Journal of Chemical Physics, Vol. 88, No. 2,
1988, pp. 563-577. doi:10.1063/1.454183
[3] J. M. Fernandez-Sanchez, A. G. Valdenebro and S. Mon-
tero, The Torsional Raman Spectra of C
2
H
6
and C
2
D
6
,
Journal of Chemical Physics, Vol. 91, No. 6, 1989, pp.
3327-3334. doi:10.1063/1.456908
[4] N. Moazzen-Ahmadi, A. R. W. McKellar, J. W. C. Johns
and I. Ozier, Intensity Analysis of the Torsional Spec-
trum of CH
3
CH
3
, Journal of Chemical Physics, Vol. 97,
No. 6, 1992, pp. 3981-3988. doi:10.1063/1.462937
[5] K. S. Pitzer and W. D. Gwinn, Energy Levels and Ther-
modynamic Functions for Molecules with Internal Rota-
tion I. Rigid Frame with Attached Tops, Journal of Che-
mical Physics, Vol. 10, No. 7, 1942, pp. 428-440.
doi:10.1063/1.1723744
[6] J. D. Swalen and D. R. Herschbach, Internal Barrier of
Propylene Oxide from the Microwave Spectrum, Jour-
nal of Chemical Physics, Vol. 27, No. 1, 1957, pp. 100-
108. doi:10.1063/1.1743645
[7] R. W. Kilb, C. C. Lin and E. B. Wilson, Calculation of
Energy Levels for Internal Torsion and Over-All Rotation.
II. CH
3
CHO Type Molecules; Acetaldehyde Spectra,
Journal of Chemical Physics, Vol. 26, No. 6, 1957, pp.
1695-1703. doi:10.1063/1.1743607
[8] R. Subramanian, S. E. Novick and R. K. Bohn, Tor-
sional Analysis of 2-Butynol, Journal of Molecular Spec-
troscopy, Vol. 222, No. 1, 2003, pp. 57-62.
doi:10.1016/S0022-2852(03)00170-X
[9] D. A. Shea, L. Goodman and M. G. White, Acetone
n-Radical Cation Internal Rotation Spectrum: The Tor-
sional Potential Surface, Journal of Chemical Physics,
Vol. 112, No. 6, 2000, pp. 2762-2768.
doi:10.1063/1.480850
[10] E. V. Ivash and D. M. Dennison, The Methyl Alcohol
Molecule and Its Microwave Spectrum, Journal of Che-
mical Physics, Vol. 21, No. 10, 1953, pp. 1804-1816.
doi:10.1063/1.1698668
[11] K. T. Hecht and D. M. Dennison, Hindered Rotation in
Molecules with Relatively High Potential Barriers Jour-
nal of Chemical Physics, Vol. 26, No. 1, 1957, pp. 31-47.
doi:10.1063/1.1743262
[12] G. . Pitsevich and M. Shundalau, Computer Simulation
of the Effect Exerted by Argon Matrix on the Internal
Rotation Barriers and Torsional States of Methanol
Molecule Journal of Spectroscopy and Dynamics, Vol. 2,
No. 3, 2012, p. 15
http://www.simplex-academic-publishers.com/jsd.aspx