A) Epsilon delta relation; (ϵ−δ) relation: Assigment no: 1 Roll no: MSCEQ TH 2020 - 10 Name: Pradej Badal
A) Epsilon delta relation; (ϵ−δ) relation: Assigment no: 1 Roll no: MSCEQ TH 2020 - 10 Name: Pradej Badal
δ 11 δ 12 δ 13
[
det δ 21 δ 22 δ 23 =
δ 31 δ 32 δ 33 ]
δ 11 ( δ 22 δ 33−δ 23 δ 32 )−δ 12 ( δ 21 δ 33−δ 23 δ 31 ) + δ 13 ( δ 21 δ 32−δ 22δ 31 )
From above we can work out the value of determinant to be 1.
Now, if we allow the first index in the above Kronecher to vary i.e.
δi 1 δi 2 δi 3
[ ]
det δ j 1 δj 2 δ j3 And put i=1, j =3, k=2 which results in the answer -1.
δ k1 δ k 2 δ k 3
Similarly, we obtain 1 when we put i=1, j=2, k=3 and 0 with some other values of ijk. Therefore, we can
be sure from the above conditions that whatever the permutations the end value of the determinant is
either 1 or -1 or 0.
Now, from Levi-Civita permutations we know that the value of ϵ ijk is either of 1,-1 or 0. So, from this
relation we can write;
δi 1 δi 2 δi 3
[
det δ j 1 δj 2 δ j3 =ϵ ijk
δ k1 δ k 2 δ k 3 ] …………………………………………………………………… (1)
Similarly, if we allow the second index of the above Kronecher determinant to vary i.e.
δ 1 l δ 1m δ 1 n
[ ]
det δ 2 l δ 2 m δ 2 n ; we will get the value of determinant either of 1,-1 or 0 for the different
δ3l δ3m δ3n
combination of values of {l,m,n}={(1,2,3),(1,3,2),….}. Therefore, similarly to (1) we can write;
δ 1 l δ 1m δ 1 n
[
det δ 2 l δ 2 m δ 2 n =ϵ lmn
δ3l δ3m δ3n ] …….…………………………………………………………... (2)
δi 1 δi 2 δi 3 δ 1l δ 1 m δ 1n
[ ][ ]
det( δ j 1 δj 2 δ j3 × δ 2l δ 2 m δ 2n ¿=ϵ ijk × ϵ lmn
δk1 δk2 δk3 δ3l δ3m δ3n
δil δi m δi n
[ ]
det δ j l δj m δ jn =ϵ ijk × ϵ lmn
δkl δkm δ kn
Now, on to determine the epsilon delta relationship. We know this relationship requires that there be a
repeated index the e terms, and that the repeated index must occupy the same slot in the permutation
tensor. Therefore, we can meet this criterion by setting l = i and obtaining the determinant:
We can evaluate the terms on the right by applying summation notation to Kronecker deltas:
Using these rules, the right hand side of eq. (3) becomes:
3 ( δjmδkn−δjnδkm )−( δjm δkn−δkm δjn ) +(δ jn δkm−δjm δkn)}=ϵ ijk × ϵ lmn
A × ( B ×C )= A × D=K
Now, K m =ϵ mni A n D i i.e.
Assigment no: 1
Roll no: MSCEQ TH 2020_10
Name : Pradej Badal
K m =ϵ mni A n ( ϵ ijk B j C k )
K m =ϵ mni ϵ ijk An B j Ck
K m =ϵ imn ϵ ijk A n B j C k
K m =( δ mj δ nk −δ mk δ nj ) A n B j C k
K m =( δ mj δ nk A n B j C k −δ mk δ nj A n B j C k )
Now, The above expression has two parts in the right hand side and its is a scalar product for now so it
can be rearranged. The left part will be non zero only if m=j and n=k and right part will be non zero only
if m=k and n=j. Since A, B,C is non zero for the equation (i) to make any sense we can write;
K m =( B m A n C n−C m A n Bn )
Now since A and C has the same index we can assume it is a dot product. Similarly A and B has the same
index it also can be assumed as a dot product.
K m =B m ( A . C )−C m ( A . B )
K e m=B e m ( A . C ) −C e m ( A . B )
K=B ( A . C )−C( A . B)
Finally