Journal of Mathematical Chemistry 19 (1996) 131-145

131

On the closed form ofWigner rotation matrix elements*

Shan- Tao Lai

Vitreous State Laboratory, The Catholic University of America, Washington, D. C. 20064, USA

P. Palting and Ying-Nan Chiu 1

Center for Chemical Dynamics and Energy Transfer, Department of Chemistry, The Catholic University of America, Washington, D. C. 20064, USA

Received 10 August 1995

The closed forms of some rotation matrix elements t!,n'm(1r/2) are presented. The closed forms of summation involved two binomials and some special hypergeometric functions are also obtained. The MAPLE V program which calculates ci,,"m(f3), t!,,,,m(1r/2) and the help file are given in appendix.

1. Introduction

An arbitrary rotation of a coordinate frame about its origin in the threedimensional space can be completely specified by three real parameters. The most I useful description of rotation in the literature and in textbooks is that in terms of the Euler angles o, (3, T' The matrix representations to the finite rotations in terms of Euler angles were first derived by Wigner [1,2] (therefore, they are often named

. Wigner rotation matrix elements), and subsequently their many properties and different derivations have been investigated by many authors [3-8]. Because the rotation matrix elements involve the sum of a product of two binomial coefficients and sines and cosines of a half-angles, any calculation involving them is tedious. Although the properties of the rotation matrix elements have been well investigated, the calculation of functional expressions is still not satisfactory to us. Even the results of numerical calculations are correct only to five significant figures as the angular momentum quantum number increases to j = 13 [9]. The reason for the inaccuracy may be lies in the operations dealing with large numbers (such as factorials). These cannot be handled by FORTRAN or BASIC programs, for example, unless they have good arithmetic methods [10]. Fortunately, symbolic calculation

• Supported in part by the Chiu Feng-Chia research Fund. Thanks are due to the Board of Trustees Chairman Dr. Ying-Ming Liao as well as the Trustee Dr. Charles Chiu-Hsiong Huang and President Dr. Ted C. Yang of Feng-Chia University, Taichung, Taiwan, ROC.

1 To whom requests for reprints should be addressed.

©J.C. Baltzer AG, Science Publishers

132

S. T. Lai et al. / Wigner rotation matrix elements

systems, such as, MAPLE V, MATHEMATICA, REDUCE, MacSyma and DERIVE, etc., can overcome the pro blem of operations involving large numbers to obtain both exact functional expressions and correct numerical numbers with the desired significant figures.

In section 2 of this paper, we simply summarize the notation and properties of the rotation matrix elements for convenient reference. In section 3, we mainly discuss the special properties of the rotation matrix elements with (3 = 1T' /2. The similarity transformation for any rotation angle (3 around the Y axis can be converted to a combination of a rotation of 1T' /2 around the Y axis and a rotation with respect to the Z axis. This similarity transformation will convert the powers of sines and cosines of half-angles to sines and cosines of multiples of (3. A method has been proposed to find closed forms of rotation matrix elements with (3 = 1T' /2. These closed forms allow calculations up toj greater than 200 (which are still very fast!). Many closed forms of summations involving two binomials are also obtained. Since the rotation matrix elements are related to the Jacobi polynomials and hypergeometrical functions, many closed forms of hyper geometrical functions and Jacobi polynomials can be obtained (we leave the Jacobi polynomial evaluation to the interested reader). The rotation matrix elements play an important role in atomic and molecular physics, chemical physics, molecular spectroscopy, angular momentum theory, group theoretical applications and nuclear physics.

2. Expressions for Wigner rotation matrix elements

A 2j + l-dimensional irreducible vector space, V (J), of the group of proper rotations in the 3-dimensional space 0+(3) is spanned by a set of2j + 1 basis functions { '1/Jjm, m = - j to j}. Since an irreducible vector space of a group is an invariant space with respect to any operator of the group, we may then express an Eulerian rotation upon one of the basis functions of V (J) as

R'Itjm = 2:dm'm(a(3,)'Itjm" (1)

m'

where the rotation ma trix elements are defined as dm'm(a(3,) = Um' I e-iaJze-i(3JYe-i'YJz Ijm).

Since the matrix representative of Jz is diagonal in this base, eq. (2) becomes dm'm(a(3,) = e-im'aUm' I e-i(3Jy Ijm)e-im'Y

= e-im'aatn'm((3)e-im'Y .

(2)

(3)

Wigner [1,2] first derived the expression for the rotation matrix element atn'm((3) and Rose [5] modified it as follows:

S. T. Lai et al. / Wigner rotation matrix elements

d!, ((3) = [(J.+m')!~-m')!]1/2,",(_I)m'_m+s(j+m) (. j-m )

mm (J +m)!{J -m)! ~ s J -m'-s

( e: (. (3)m'-m+2s

X cos2 sm2 '

133

(4)

where (:)'s are binomial coefficients. The summation over s is restricted to the argument of any factorial which is non-negative. Let s = j - m' - a, then a:n'm((3) is related to other definitions [12] by

atn'm((3) (Rose [5]) = atn'm((3) (Brink & Satchler [6], Zare [8])

= (-I)m'-matn'm((3) (Edmonds [4], Wigner [1,2], Fano & Racah [3]). (5)

From eq. (4) we can obtain the symmetry properties of atn'm ((3), where j, m' and m may be either integral or half-integral:

atn'm((3) = (-I)m'-matnm,((3) ~ (-It'-mai__m'_m((3) = (-I)m'-matn'm( -(3), (6a)

atn'm (1T' + (3) = (-1 y-m' c.: ((3) , atn'm (1T' - (3) = (-1 v" atn,-m' ((3) ,

. '+m

a:n'm(1T') = (-1 Y 8m+m,,0,

atn'm ( -1T') = (-1 r: 8m+m, ,0 , atn'm(O) = 8m',m .

The atn'm ((3) satisfy the following differential equations:

{d2 (3.!!__m2+m'2-2mm'cos(3 'f; 1)}d! ((3)=0

d(32 + cot d(3 sin2 (3 + Jv + m'm

and [13]

{i cos (3 - mm - j sin (3 :(3 } atn'm ((3)

= [(J + m){J - m){J + m'){J - m')ll/2atn-;~((3)

and

{ {J + 1)2 cos (3 - mm + (J + 1) sin (3 :(3 } «: ((3)

= [(J + m + 1){J - m + 1){J + m' + 1){J - m' + l)ll/2atn~~(,6).

(6b) (6c) (6d) (6e) (6f)

(7)

"(8a)

(8b)

134

S. T. Lat et al. ! Wigner rotation matrix elements

The detailed descriptions of eqs. (8a) and (8b) can be found in [13]. Eqs. (5) to (8b) can also be translated to the rotational wave functions dmlm(a(3,) (not normalized).

3. On the rotation matrix elements of (3 = 11'/2

3.1. THE CALCULATION OF tPm'm(3)

The disadvantage of eq. (4) is that it expresses the rotation matrix elements in terms of powers of cos((3/2) and sin((3/2). Wigner [14] has shown the matrix representatives for an arbitrary (3 can be obtained by the following similarity transformation:

Therefore,

(jm' I R(O, (3, 0) I jm)

= (jm'l RG,O,O)R(O,~;O)R((3,O,O)R(O,-~,O)R( -~,0,0) Ijm) (10)

where N'm1m == ~'m(1I'/2), and this equation can be simplified to ~'m((3) = N'mloN'mo~(O) + 2 L N'mlm"N'mm,,~(m/l(3)

m">O

(lla)

for integralj, m', m, and

~'m((3) = 2 L N'mlm"N'mm,,~(m/l(3)

m"~1/2

(llb)

for half-integralj, m', m, where

(COS x if m' - m = 0 (mod 4) ,

~(x) = -sinx if m'-m= 1 (mod 4), - cos x if m' - m = 2 (mod 4) , sin x if m' - m = 3 (mod 4) .

Therefore, in principle, to calculate ~'m((3) one only needs to know N'm1m' Both the exact and numerical values of Nm,m, have been calculated for 0 ~m', m,j ~20 in the present work. Note that Behkami [9] has tabulated blm'm> j ~ 13, numerically and Bradley [15] used a desk machine to construct the tables of Nm'm fromj = 0 up to j = 20, numerically. The tables of Nm'm have been deposited with the Royal Society and the Library of Congress. The general forms for different m' = j - /C,

(12)

S. T. Lai et al.! Wigner rotation matrix elements

135

m = j - k, /C, k = 0, 1, ... ,20, have also been calculated by a MAPLE program and the program is given in the appendix. Both procedures for calculating eqs. (4) and (11) have been written and run under the MAPLE V system. The results agree with each other. In terms of eqs. (6a) and (6c) (three properties), it is easy to show that the number of different matrix elements to be tabulated from (2l + 1)2 reduce to (j + 1)(j + 2)/2 for integer j, and to (2j + 1)(2j + ~)/8 whenj is a half-integer. Previously Buckmaster [16] used two properties of dJn'm((3) and reduced the tabulated numbers from (2l + 1)2 to (j + 1)2 for integer j and (2j + 1) (2j + 3) /4 for the half-integer j.

3.2. THE CLOSED FORMS OF SOME SPECIAL Dlm'm

When (3 = 11'/2, eq. (4) reduces to

N', = (!)j[(j.+m')!~-m')!]1/2L(_1)ml-m+s(j+m) (. j-m )

mm 2 (j+m)!(j-m)! s s J-m'-s

(13)

and eq. (6c) becomes

j _ '+m j

ls.m1m - (-1 Y ls.-m1m .

~t is easy to show that

I

(14)

~O=O if j = odd, and

~m = (-1 y+m ~m =1= 0

only ifj + m = even. It can also be shown that

(15)

(16)

!J.2j = !J.2j = (-lyr(j + 1/2) .

11 00 j!y'1r

Here P's arethegammafunctions,j = 1,2,3, ... , and from eq. (13) we have

(17)

/J.jj = (~y

By using our previous results [17] and combining them with eq. (17) we obtain the following closed forms:

(18)

!J.2k+1 _ (-l)kr(k+ 1/2)

11 - 2(k + 1)!y'1r ,

(19)

136

S.T. Laietal. / Wigner rotation matrix elements

A 2k+~ _ (-1l+1r(k + ~)

L.l.11 - . ~ ,

22 2(k+1)!v21l'

2k-H (-1lr(k +!)

t::..11 2 = .

22 k!v'21f

Here k = 0, 1,2, .... Also we have

t::..2k = (-1l+l [ k(2k + 1) ] 1/2 r(k + 1/2)

20 (k + 1)(2k - 1) k!..fo'

t::..2k+1 _ (-1l+1 [k(2k + 3)]1/2r(k + !)

21 - (k + 1)!v'21f '

2k k fi r(k + !)

t::..21 = (-1) Y; k![(2k+ 1)(k+ 1)]1/2 '

2k _ (_ )k (2P+k-4)r(k+!)

t::..22 - 1 (k + 1)!(2k - 1)..fo '

2k+l k 4r(k + 3/2)

t::..22 = (-1) (k+ 1)!(2k+ 1)..fo'

(20)

(21)

(22)

(23)

(24)

(25)

(26)

wherek = 1,2,3,. 0" Moreover,

t::..2k+1 = (-1l+1 [2k+ 1 ]1/2 r(k+ 1/2)

10 2(k + 1) k!..fo'

(27)

wherek = 0, 1,2,3,0' 0, and

t::..2k+l-(-1l[ k(2k+1)(2k+3) ]1/2 r(k+ 1/2)

30 - (k+1)(k+2)(2k-1) k!v'21f'

(28)

wherek = 1,2,3,. 0 0, and

2k k (2k2 + k - 19)r(k + !)

t::..33=(-1) (k+1)!(2k-1)..fo'

(29)

wherek = 2,3,4,0'" and

2k+l _ (_ )k 3(6P + 9k - 16)r(k + !)

t::..33 - 1 2(k + 2)!(2k - 1)..fo '

(30)

wherek = 1,2,3, ... , and

S. T. Lai et al. / Wigner rotation matrix elements

137

t::..2k=(-1l[ k(2k+1)(2k+3)(k-1) ]1/2r(k+1/2)

40 (k + l)(k + 2)(2k - 1)(2k - 3) k!..fo'

t::..2k = (-1l (576 - 67k-133k2 +4k3 +4k4)r(k+!)

44 (2k-3)(k+2)!(2k-1)..fo'

t::..2k+1 = (_1)k+l 16(17 - 3k - 2k2)r(k + ~)

44 (2k+ 1)(k+2)!(2k-1)..fo'

where k = 2, 3,4, 0 ••• Somewhat more general forms are

tY _ (-1)m..fo [(j +m)!] 1/2

mO - 2mr(1 + (j + m)/2)r(1/2 - (j - m)/2) (j - m)!

(31 )

(32)

(33)

(34)

and

tY = (_1)m+l..for_ (j+m)! ]1/2

ml . 2m [}(j + 1)(j - m)!

X {r(1 + (j + m)/2)01/2 - (j - m)/2)

j+m+1 }

+ r(3/2 + (j + m)/2)r( -(j - m)/2)

(35)

and

tY _(-1)m..for_ (j+m)! ]1/2

m2 - 2m [}(j + 1)(;2 + j - 2)(j - m)!

{ 2m2 -j(j + 1)

(36)

. (37)

(38)

in which 1 + a - b =1= 0, -1, -2, . 0 0, and the hypergeometric function is defined

138

S. T. Lai et al. / Wigner rotation matrix elements

2Fl(a,b;c;z) = f (a)n(b)nz" n=O (c)nn!

and the Pochhammer symbol is

(39)

()=r(a+n)

an r(a)'

i.e.

(a)o=l, (a)n=a(a+1) ... (a+n-1), n=1,2,3, ....

In principle, we can use the same method to find closed forms for t:! t:!

j m3' m4'

~m5"'" etc.

3.3. THE CLOSED FORMS FOR SUMMATIONS OF TWO BINOMIAL COEFFICIENTS

Summations involving two binomial coefficients are often used in angular momentum theory and combination theory. We can rewrite eq. (13) as follows:

2:)-lt(J+m)(. J-m ) ='}}[~-m)!~+m)!]l/\_l)ml-mt:!1 .

8 S } - m' - s U - m')! U + m')! m m

(40) By comparing the results of the previous section with eq. (40), we then obtain the following expressions:

(41)

8(2k) ( 2k) (_1)k+122kk

~(-1) s 2+s =(k+1)!y0rr(k+1/2).

(42)

Herek= 1,2,3, ... ,and

(43)

(44)

S.T. Lai et al. / Wigner rotation matrix elements

139

I) _1)8 (2k + 2) (2k - 2) = (-1)k22k(2k2 + k - 4)r(k + 1/2)

8 s s (2k - l)(k + 1)!y0r ,

2::(_1)8(2k+3).(2k-1) = (-1)k22k+3r(k+3/2)

8 s s (2k + l)(k + 1)!y0r ,

2::( _It(2k + 3) (2k - 3) = (_1)k22k(2k2 + k - 19)r(k + 1/2)

8 S S (2k - l)(k + 1)!y0r ,

(45)

(46)

(47)

where k = 2,3,4, ... , and

2::(_lt(2k+4) (2k-2) = (-ll3 x 22k(6k2+9k-16)r(k+ 1/2)

8 S S (k + 2)!(2k - 1)y0r ,

(48)

}Vherek = 1,2,3, ... , and

2::( _1)8 (2k + 1) (2k + 1) = (-1)k22k(2k + l)r(k + 1/2)

8 S l+s (k+1)!y0r'

}Vherek = 0, 1,2, ... , and

i

(49)

2::( -It (2k + 1) (2k + 1) = (-1)k+l22kk(2k + l)r(k + 1/2)

8 S 3+s (k+2)!y0r ,

where z = 0, 1,2, ... , and

(50)

2::(-lt(k) ( k ) _ 2k-mk!y0r

8 S m + s - (k - m)!r(l + (k + m)/2)r(1/2 - (k -1r!)/2) ,

(51 )

2::(-1 8(k+m) (k-m) _ (-lt2k-m(k+ m)!y0r

8 ) S S - m - (k)!r(l + (k + m)/2)r(1/2 - (k - m)/2) ,

(52)

wherem = 0, ±1, ±2, ... , and

2::( -It (2k + 5) (2k - 3) = (_1)k+122k+5(17 - 3k - 2k2)r(k + 3/2)

8' s s (2k+1)(2k-1)(k+2)!y0r'

(53)

wherek = 2,3, ... , and

140

S. T. Lai et al. / Wigner rotation matrix elements

~(_I)s(2k: 4) (2k; 4)

(-I)k22k(576 - 67k - 133k2 + 4k3 + 4k4)r(k + !)

-

(2k - 3)(2k - 1)(k + 2)!y0T

(54)

,",(_lr(k+m) ( k-m ) = (-It+l2k-m(k+ m)!y0T

~ s k - 1- s (k + I)!

X {r(1 + (k + m)/2)~1/2 - (k - m)/2)

k+m+l }

+ r(3/2 + (k + m)/2)r( -(k - m)/2) ,

(55)

where I'(O) = 00, and

,",(_lr(k+m) ( k-m ) = (_I)m2k-my0T(k+m)!

~ s k-2-s (k+2)!

{ 2m2-k(k+l)

(56)

(57)

,",(_lr(2k+2) ( 2k ) = (-ll22k+1r(k+ 1/2) .

~ s l+s· (k+l)!y0T

(58)

3.4. SPECIAL VALUES OF HYPER GEOMETRIC FUNCTIONS

When,6 = 7r /2, eq. (37) becomes

2Fl(m' - j, -m - jim' - m + 1; -1)

[U + )'U ')'] 1/2

= m. - m. (_I)m'-m(, _ )'LY

U - m)!U + m')! m m. m'm'

where m' ~m, and, using eqs. (17), (19), (35), (36) and (59), we obtain some special

(59)

values of the hypergeometric functions:

S. T. Lai et al. / Wigner rotation matrix elements

141

2Fl( -2j, -2j; 1; -1) = 2Fl(1- 2j, -1 - 2j; 1; -1) (-1!22jru + 1/2)

=

j!y0T

(60)

. . (-1!22jrU+l/2)

2Fl( -2J, -2 - 2J; 1; -1) = U + 1)!y0T ,

wherem = 1,2,3, ... ,and 2Fl(m-j,-I-j;m;-I) ='2!-m(m-l)!y'1r

X {r(1 + U + m)/2)~1/2 - U - m)/2)

j+m+l }

+ r(3/2 + U + m)/2)r( -U - m)/2) ,

(61 )

(62)

2Fl(m - j, -2 - jim -1; -1) = '2!-my'1r(m - 2)!

{ 2m2 - jU + 1)

(63)

where m = 2,3,4, .... The interested reader can use the same technique to obtain lfany other special values of the hypergeometric functions. Therefore, it is noteworthy that the relation between the sum over binomial coefficients and hypergeometric function are also obtainable:

L( -1 Y (j + m) ( . j -,m )

s s J-m-s

U - m)! F (" ., 1 1)

= (m'-m)!U-m')! 21m -J,-m-J;m -m+ ;- ,

(64)

I

wherem'~m.

4. Conclusions

In this work we have presented many relations between LYm'm and gamma functions, and summations involving two binomial coefficients, and hypergeometric functions, Even though a general closed form for LYm'm has not been found yet, a useful way to determine closed forms of &m'm> with special values of m' or m has been explored. Since d'm'm (,6) is related to the Jacobi polynomials [17], many relations between the Jacobi polynomials and summations involving two binomial

142

S. T. Lai et al. / Wigner rotation matrix elements

coefficients, and hypergeometric functions can be deduced. Some of the applications of t::!m1m can be found in refs. [11,15,18,19].

Acknowledgements

The author (STL) would like to thank Professor R.N. Zare for many helpful discussions and suggestions. Thanks are also due to Professor Macedo and Professor I.L. Pegg for constant support and encouragement.

Appendix

MAPLE PROGRAM FOR ROTATION MATRIX ELEMENTS

# The following program calculates Wigner rotation matrix elements # which runs on MAPLE V system.

#

'helpltext/rotationmatrix' :=TEXT(

'FUN CTI ON: d, Rot - calculate Wigner rotation matrix elements with', , angle x.',

"

,

'CALLING SEQUENCE: " 'dtj, n, m, x)'

'Rottj, n, m, x)',

'd90G, n, m);',

"

,

'PARAMETERS: "

'j - rotational angular momentum quantum number.', 'n, m - magnetic quantum number.',

'x - Euler angle.',

"

,

'SYNOPSIS: - A typical call to the d and Rot functions are',

, dfj, n, m, x), d90G, n, m) and Rottj, n, m,x)wherej >= 0, it can be integer', or half-integer. n, m = -j to j, x is an angle (radian).',

- Definition of rotation matrix element is the same as',

Rose [5], Brink & Satchler [6], Zare [8] and Lai [12].',

The results of dtj, n, m, x) and Rottj, n, m, x) should be the same.',

"

"

,

'EXAMPLES: " 'd(I, 1, l ;x);', 'd90(I, 1, 1);', 'Rot(1, 1, l,x);', 'd(I, 0, 0, x);',

S. T. Lai et at. / Wigner rotation matrix elements

143

'Rot(I, 0, 0, x);',

'd(I, 0, -I,x);',

'Rot(I, 0, -1, x);',

'd(I, -1, -1, x);',

'Rot(I, -1, -I,x);',

'd(2, 0, 0, x);',

'Rot(2, 0, 0, x);',

'd(l/2, 112, 112, x);', 'Rot(1l2, 112, 112, x);', 'd(3/2, 112, -3/2, x);', 'd90(312, 1/2, -3/2);', 'Rot(3/2, 112, -3/2,x);',

"

,

"):

#

# Based on eq. (4)

#

d:= proc G, n, m, x) local d l , d2, ds, s: i if x = Pi then

dl:= (-ItU + m)*Kronecker(n+ m,O); else

if x = -Pi then

i dI:= (-1 tU - m)*Kronecker(n + m, 0); else

if x = 2*Pi then dI:= (-1) A(2*j)*Kronecker(n, m); else

if x = 0 then dl := Kronecker(n, m);

else

dI:= U + n)!A(1I2)*G - n)r(1I2)/G - m)r(I/2)/G + m)r(1I2); ds:=O;

for s from 0 to 2*j do

if j + m - s >= 0 andj - n - s >= 0 and n + s - m >= 0 then

ds:= ds +( -It(n - m + s)*binomiaIG + m, s)*binomiaIG - m,j - n - s) *cos(1 12*xf(2*j + m - n - 2*s)

*sin(II2*xf(n - m + 2*s):

fi; od;

d2 := factor(simplify( ds*dI »; fi:

fi;fi;fi

end;

#

144

S. T. Lai et al. / Wigner rotation matrix elements

d90 := proctj, n, m) localdl, s;

dl := simplify(sqrtG +n)!*G - n)!/G - m)!/G +m)!)*

sumf -1)**(n -m + s)*binomialG +m, s)*binomialG - m,j - n - s) *(ll2fj, s = 0 .. 2*j));end:

k:= proc(n, m, x)

localkl;

if typerjn - m) mod 4,0) then kl := cos(x); elseiftype(n-mmod4, l)thenkl:= -sin(x); else iftype(n - m mod 4,2) then kl := -cos(x); elseiftype(n - mmod4, 3) thenkl := sin(x); fi;fi;fi;fi; end:

#

# Based on eqs. (1Ia) to (12).

#

Rot=proctj, n, m, x)

local d2, g2;

if x =Pi theng2:= (-lrG +m)*Kronecker(n+m, 0);

else if x =-Pi then g2 := (-lrG - m)*Kronecker(n + m, 0); else if x = 0 then g2 i= Kronecker(n, m);

else if x = 2*Pi then g2 :=( -1r<2*j)*Kronecker(n, m); else

if type G, fraction) then d2:=0, form2from 1 12 toj

do

d2:= d2 + 2*d90G, n, m2)*d90G, m, m2)*k(n, m, m2*x); od;

else

d2 := d90G, n, 0)*d90G, m, O)*k(n, m, 0);

form2from 1 toj .

do

d2:= d2 + 2*d90G, n, m2)*d90G, m, m2)*k(n, m, m2*x);

od; fi;fi;fi;fi;fi;

factor( simplify( d2)); end:

Kronecker := proc(n, m) localkl;

ifn=m thenkl:= 1; elsekl :=0;

fi;

end:

S. T. Lai et al. / Wigner rotation matrix elements

145

References

[1] E.P. Wigner, Gruppentheorie und ihre Anwendung auf die Quantenmechanik der Atomspektren, (Vieweg, Braunschweig, Germany, 1931).

[2] E.P. Wigner, Group Theory and Its Application to the Quantum Mechanics of Atomic Spectra (Academic Press, New York, 1959).

[3] U. Fano and G. Racah, Irreducible Tensorial Sets (Academic Press, New York, 1957).

[4] A.R. Edmonds, Angular Momentum in Quantum Mechanics (Princeton,U.P., Princeton, NJ,

1960).

[5] M.E. Rose, Elementary Theory of Angular Momentum (Wiley, New York, 1957). [6] D .M. Brink and Satchler, Angular Momentum (Oxford University Press, 1962).

[7] L.C. Biedenharn and J.D. Louck, Angular Momentum in Quantum Physics (Addison-Wesley, Reading, MA, 1981).

[8] R.N. Zare, Angular Momentum (Wiley, 1988). [9] A.N. Behkami, Nuclear Data Tables 10 (1971) 1.

[10] D.A. Varshalovich, A.N. Moskalev and V.K. Khersonskii, Quantum Theory of Angular Momentum (W orld Scientific, 1988).

[11] Q.E. Zhang, L.T. Lin, N.Q. Wang, Y.G. Wang, S.T. Lai and v.x. Yu, Polyhedral Molecular

Orbitals (Chinese Academic Press, 1987).

[12] S.T. Lai, Ph.D. Thesis, The Catholic University of America (1991). [P] S.T. Lai, P. Palting and Y.N. Chiu, J. Molec. Sci. 10 (1994) 1.

04] E.P. Wigner, On the matrices which reduce the Kronecker products of representations of S.R. i Groups (1940).

05] C.J. Bradley, The matrix representatives d~'~(7r/2) for the rotation group, London: The Royal Society Depository of Unpublished Tables, File No. 79, Washington: Library of Congress

i Collection, No. 61-18944.

[16] H.A. Buckmaster, Can. J. Phys. 42 (1964) 386; 44(1966) 2525. n 7] S.T. Lai and Y.N. Chiu, J. Math. Phys. 30 (1989) 844.

[i8] S.L. Altmann and C.J. Bradley, Phil. Trans. Roy. Soc. A255 (1963) 25.

[19] S.T. Lai, Q.E. Zhang, P. Palting and Y.N. Chiu, The transformation coefficients between the irreducible bases of SO(3) and 0 group: A MAPLE V symbolic calculation package, presented at the 5th National Quantum Chemistry Symposium of China, Xiamen University (1993).