本文主要内容如下：
1. 度量张量
1.1 度量张量协（逆）变分量的坐标转换关系
由度量张量协（逆）变分量的定义及基矢量的坐标变换关系可得到如下的度量张量协（逆）变分量的坐标转换关系： g i ′ j ′ = ( g ⃗ i ′ , g ⃗ j ′ ) = ( β m i ′ g ⃗ m , β n j ′ g ⃗ n ) = β m i ′ β n j ′ ( g ⃗ m , g ⃗ n ) = β m i ′ β n j ′ g m n g i ′ j ′ = ( g ⃗ i ′ , g ⃗ j ′ ) = ( β i ′ m g ⃗ m , β j ′ n g ⃗ n ) = β i ′ m β j ′ n ( g ⃗ m , g ⃗ n ) = β i ′ m β j ′ n g m n g i j = ( g ⃗ i , g ⃗ j ) = ( β m ′ i g ⃗ m ′ , β n ′ j g ⃗ n ′ ) = β m ′ i β n ′ j ( g ⃗ m ′ , g ⃗ n ′ ) = β m ′ i β n ′ j g m ′ n ′ g i j = ( g ⃗ i , g ⃗ j ) = ( β i m ′ g ⃗ m ′ , β j n ′ g ⃗ n ′ ) = β i m ′ β j n ′ ( g ⃗ m ′ , g ⃗ n ′ ) = β i m ′ β j n ′ g m ′ n ′ g^{i'j'} =(\vec{g}^{i'},\vec{g}^{j'}) =(\beta_{m}^{i'}\vec{g}^{m},\beta_{n}^{j'}\vec{g}^{n}) =\beta_{m}^{i'}\beta_{n}^{j'}(\vec{g}^{m},\vec{g}^{n}) =\beta_{m}^{i'}\beta_{n}^{j'}g^{mn} \\\ \\g_{i'j'} =(\vec{g}_{i'},\vec{g}_{j'}) =(\beta^{m}_{i'}\vec{g}_{m},\beta^{n}_{j'}\vec{g}_{n}) =\beta^{m}_{i'}\beta^{n}_{j'}(\vec{g}_{m},\vec{g}_{n}) =\beta^{m}_{i'}\beta^{n}_{j'}g_{mn} \\\ \\g^{ij} =(\vec{g}^{i},\vec{g}^{j}) =(\beta_{m'}^{i}\vec{g}^{m'},\beta_{n'}^{j}\vec{g}^{n'}) =\beta_{m'}^{i}\beta_{n'}^{j}(\vec{g}^{m'},\vec{g}^{n'}) =\beta_{m'}^{i}\beta_{n'}^{j}g^{m'n'} \\\ \\ g_{ij} =(\vec{g}_{i},\vec{g}_{j}) =(\beta^{m'}_{i}\vec{g}_{m'},\beta^{n'}_{j}\vec{g}_{n'}) =\beta^{m'}_{i}\beta^{n'}_{j}(\vec{g}_{m'},\vec{g}_{n'}) =\beta^{m'}_{i}\beta^{n'}_{j}g_{m'n'} gi′j′=(gi′,gj′)=(βmi′gm,βnj′gn)=βmi′βnj′(gm,gn)=βmi′βnj′gmn gi′j′=(gi′,gj′)=(βi′mgm,βj′ngn)=βi′mβj′n(gm,gn)=βi′mβj′ngmn gij=(gi,gj)=(βm′igm′,βn′jgn′)=βm′iβn′j(gm′,gn′)=βm′iβn′jgm′n′ gij=(gi,gj)=(βim′gm′,βjn′gn′)=βim′βjn′(gm′,gn′)=βim′βjn′gm′n′
1.2 度量张量
显然，度量张量的协变分量或逆变分量满足张量分量的坐标转换关系，因而该有序数组的确是一个张量，一般称作度量张量 G \bold G G。根据张量分量的指标升降关系与度量张量协变分量、逆变分量的互逆关系得知度量张量的混变分量： g i ∙ j = g k j g i k = g i k g k j = δ i j = δ j i g ∙ j i = g i k g k j = g k j g i k = δ j i g^{\bullet j}_i=g^{kj}g_{ik}=g_{ik}g^{kj}=\delta^{j}_{i}=\delta^{i}_{j}\\\ \\g_{\bullet j}^i=g^{ik}g_{kj}=g_{kj}g^{ik}=\delta^{i}_{j} gi∙j=gkjgik=gikgkj=δij=δji g∙ji=gikgkj=gkjgik=δji那么度量张量的实体记法可写作： G = g i j g ⃗ i g ⃗ j = g i j g ⃗ i g ⃗ j = δ j i g ⃗ i g ⃗ j = g ⃗ i g ⃗ i = δ j i g ⃗ i g ⃗ j = g ⃗ i g ⃗ i \bold G=g^{ij}\vec{g}_i\vec{g}_j=g_{ij}\vec{g}^i\vec{g}^j=\delta^{i}_j\vec{g}_i\vec{g}^j=\vec{g}_i\vec{g}^i=\delta^{i}_j\vec{g}^i\vec{g}_j=\vec{g}^i\vec{g}_i G=gijgigj=gijgigj=δjigigj=gigi=δjigigj=gigi
2. 置换张量
2.1 置换符号
定义如下三指标的符号为置换符号/Ricci符号：
e
i
j
k
=
e
i
j
k
=
{
1
(
i
,
j
,
k
为顺序排列：
123
、
231
、
312
)
−
1
(
i
,
j
,
k
为逆序排列：
132
、
213
、
321
)
0
(
i
,
j
,
k
为非序排列）
e^{ijk}=e_{ijk}= \begin{cases} 1&\quad(i,j,k为顺序排列：123、231、312)\\ -1&\quad(i,j,k为逆序排列：132、213、321)\\ 0&\quad(i,j,k为非序排列） \end{cases}
eijk=eijk=⎩
⎨
⎧1−10(i,j,k为顺序排列：123、231、312)(i,j,k为逆序排列：132、213、321)(i,j,k为非序排列）
根据置换符号的定义，知道置换符号具有反对称性，即：
e
i
j
k
=
−
e
i
k
j
=
−
e
k
j
i
=
−
e
j
i
k
e
i
j
k
=
−
e
i
k
j
=
−
e
k
j
i
=
−
e
j
i
k
e^{ijk}=-e^{ikj}=-e^{kji}=-e^{jik}\\ \ \\ e_{ijk}=-e_{ikj}=-e_{kji}=-e_{jik}
eijk=−eikj=−ekji=−ejik eijk=−eikj=−ekji=−ejik
2.2 行列式的展开式
行列式的展开式可借助置换符号进行简单地表示，如下：(注：上/前指标表示行编号，下/后指标为列编号)
d
e
t
(
a
m
∙
n
)
=
∣
a
1
∙
1
a
2
∙
1
a
3
∙
1
a
1
∙
2
a
2
∙
2
a
3
∙
2
a
1
∙
3
a
2
∙
3
a
3
∙
3
∣
=
a
1
∙
1
a
2
∙
2
a
3
∙
3
+
a
1
∙
2
a
2
∙
3
a
3
∙
1
+
a
1
∙
3
a
2
∙
1
a
3
∙
2
−
a
1
∙
1
a
2
∙
3
a
3
∙
2
−
a
1
∙
2
a
2
∙
1
a
3
∙
3
−
a
1
∙
3
a
2
∙
2
a
3
∙
1
=
a
1
∙
i
a
2
∙
j
a
3
∙
k
e
i
j
k
=
a
i
∙
1
a
j
∙
2
a
k
∙
3
e
i
j
k
d
e
t
(
a
∙
n
m
)
=
∣
a
∙
1
1
a
∙
2
1
a
∙
3
1
a
∙
1
2
a
∙
2
2
a
∙
3
2
a
∙
1
3
a
∙
2
3
a
∙
3
3
∣
=
a
∙
1
1
a
∙
2
2
a
∙
3
3
+
a
∙
1
2
a
∙
2
3
a
∙
3
1
+
a
∙
1
3
a
∙
2
1
a
∙
3
2
−
a
∙
1
1
a
∙
2
3
a
∙
3
2
−
a
∙
1
2
a
∙
2
1
a
∙
3
3
−
a
∙
1
3
a
∙
2
2
a
∙
3
1
=
a
∙
1
i
a
∙
2
j
a
∙
3
k
e
i
j
k
=
a
∙
i
1
a
∙
j
2
a
∙
k
3
e
i
j
k
d
e
t
(
a
m
n
)
=
∣
a
11
a
12
a
13
a
21
a
22
a
23
a
31
a
32
a
33
∣
=
a
11
a
22
a
33
+
a
13
a
21
a
32
+
a
12
a
23
a
31
−
a
11
a
23
a
32
−
a
13
a
22
a
31
−
a
12
a
21
a
33
=
a
1
i
a
2
j
a
3
k
e
i
j
k
=
a
i
1
a
j
2
a
k
3
e
i
j
k
d
e
t
(
a
m
n
)
=
∣
a
11
a
12
a
13
a
21
a
22
a
23
a
31
a
32
a
33
∣
=
a
11
a
22
a
33
+
a
13
a
21
a
32
+
a
12
a
23
a
31
−
a
11
a
23
a
32
−
a
13
a
22
a
31
−
a
12
a
21
a
33
=
a
1
i
a
2
j
a
3
k
e
i
j
k
=
a
i
1
a
j
2
a
k
3
e
i
j
k
det(a^{\bullet n}_m)= \begin{vmatrix} a^{\bullet 1}_1&a^{\bullet 1}_2&a^{\bullet 1}_3\\ \\ a^{\bullet 2}_1&a^{\bullet 2}_2&a^{\bullet 2}_3\\ \\ a^{\bullet 3}_1&a^{\bullet 3}_2&a^{\bullet 3}_3\\ \end{vmatrix} = a^{\bullet 1}_1a^{\bullet 2}_2a^{\bullet 3}_3+a^{\bullet 2}_1a^{\bullet 3}_2a^{\bullet 1}_3+a^{\bullet 3}_1a^{\bullet 1}_2a^{\bullet 2}_3-a^{\bullet 1}_1a^{\bullet 3}_2a^{\bullet 2}_3-a^{\bullet 2}_1a^{\bullet 1}_2a^{\bullet 3}_3-a^{\bullet 3}_1a^{\bullet 2}_2a^{\bullet 1}_3 =a_1^{\bullet i}a_2^{\bullet j}a_3^{\bullet k}e_{ijk} =a_i^{\bullet 1}a_j^{\bullet 2}a_k^{\bullet 3}e^{ijk}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%%%% det(a_{\bullet n}^m)= \begin{vmatrix} a_{\bullet 1}^1&a_{\bullet 2}^1&a_{\bullet 3}^1\\ \\ a_{\bullet 1}^2&a_{\bullet 2}^2&a_{\bullet 3}^2\\ \\ a_{\bullet 1}^3&a_{\bullet 2}^3&a_{\bullet 3}^3\\ \end{vmatrix} = a_{\bullet 1}^1a_{\bullet 2}^2a_{\bullet 3}^3+a_{\bullet 1}^2a_{\bullet 2}^3a_{\bullet 3}^1+a_{\bullet 1}^3a_{\bullet 2}^1a_{\bullet 3}^2-a_{\bullet 1}^1a_{\bullet 2}^3a_{\bullet 3}^2-a_{\bullet 1}^2a_{\bullet 2}^1a_{\bullet 3}^3-a_{\bullet 1}^3a_{\bullet 2}^2a_{\bullet 3}^1 =a_{\bullet 1}^ia_{\bullet 2}^ja_{\bullet 3}^ke_{ijk} =a_{\bullet i}^1a_{\bullet j}^2a_{\bullet k}^3e^{ijk}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%% det(a_{mn})= \begin{vmatrix} a_{11}&a_{12}&a_{13}\\ \\ a_{21}&a_{22}&a_{23}\\ \\ a_{31}&a_{32}&a_{33}\\ \end{vmatrix} =a_{11}a_{22}a_{33}+a_{13}a_{21}a_{32}+a_{12}a_{23}a_{31}-a_{11}a_{23}a_{32}-a_{13}a_{22}a_{31}-a_{12}a_{21}a_{33} =a_{1i}a_{2j}a_{3k}e^{ijk} =a_{i1}a_{j2}a_{k3}e^{ijk}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%% det(a^{mn})= \begin{vmatrix} a^{11}&a^{12}&a^{13}\\ \\ a^{21}&a^{22}&a^{23}\\ \\ a^{31}&a^{32}&a^{33}\\ \end{vmatrix} =a^{11}a^{22}a^{33}+a^{13}a^{21}a^{32}+a^{12}a^{23}a^{31}-a^{11}a^{23}a^{32}-a^{13}a^{22}a^{31}-a^{12}a^{21}a^{33} =a^{1i}a^{2j}a^{3k}e_{ijk} =a^{i1}a^{j2}a^{k3}e_{ijk}
det(am∙n)=∣
∣a1∙1a1∙2a1∙3a2∙1a2∙2a2∙3a3∙1a3∙2a3∙3∣
∣=a1∙1a2∙2a3∙3+a1∙2a2∙3a3∙1+a1∙3a2∙1a3∙2−a1∙1a2∙3a3∙2−a1∙2a2∙1a3∙3−a1∙3a2∙2a3∙1=a1∙ia2∙ja3∙keijk=ai∙1aj∙2ak∙3eijk det(a∙nm)=∣
∣a∙11a∙12a∙13a∙21a∙22a∙23a∙31a∙32a∙33∣
∣=a∙11a∙22a∙33+a∙12a∙23a∙31+a∙13a∙21a∙32−a∙11a∙23a∙32−a∙12a∙21a∙33−a∙13a∙22a∙31=a∙1ia∙2ja∙3keijk=a∙i1a∙j2a∙k3eijk det(amn)=∣
∣a11a21a31a12a22a32a13a23a33∣
∣=a11a22a33+a13a21a32+a12a23a31−a11a23a32−a13a22a31−a12a21a33=a1ia2ja3keijk=ai1aj2ak3eijk det(amn)=∣
∣a11a21a31a12a22a32a13a23a33∣
∣=a11a22a33+a13a21a32+a12a23a31−a11a23a32−a13a22a31−a12a21a33=a1ia2ja3keijk=ai1aj2ak3eijk可根据互换两行（列）次序则行列式变号的性质得到如下关系：
（1）任意更换行列式两行：
∣
a
1
∙
p
a
2
∙
p
a
3
∙
p
a
1
∙
q
a
2
∙
q
a
3
∙
q
a
1
∙
l
a
2
∙
l
a
3
∙
l
∣
=
a
i
∙
p
a
j
∙
q
a
k
∙
l
e
i
j
k
=
a
i
∙
1
a
j
∙
2
a
k
∙
3
e
i
j
k
e
p
q
l
=
d
e
t
(
a
m
∙
n
)
e
p
q
l
∣
a
∙
1
p
a
∙
2
p
a
∙
3
p
a
∙
1
q
a
∙
2
q
a
∙
3
q
a
∙
1
l
a
∙
2
l
a
∙
3
l
∣
=
a
∙
i
p
a
∙
j
q
a
∙
k
l
e
i
j
k
=
a
∙
i
1
a
∙
j
2
a
∙
k
3
e
i
j
k
e
p
q
l
=
d
e
t
(
a
∙
n
m
)
e
p
q
l
∣
a
p
1
a
p
2
a
p
3
a
q
1
a
q
2
a
q
3
a
l
1
a
l
2
a
l
3
∣
=
a
p
i
a
q
j
a
l
k
e
i
j
k
=
a
1
i
a
2
j
a
3
k
e
i
j
k
e
p
q
l
=
d
e
t
(
a
m
n
)
e
p
q
l
∣
a
p
1
a
p
2
a
p
3
a
q
1
a
q
2
a
q
3
a
l
1
a
l
2
a
l
3
∣
=
a
p
i
a
q
j
a
l
k
e
i
j
k
=
a
1
i
a
2
j
a
3
k
e
i
j
k
e
p
q
l
=
d
e
t
(
a
m
n
)
e
p
q
l
\begin{vmatrix} a^{\bullet p}_1&a^{\bullet p}_2&a^{\bullet p}_3\\ \\ a^{\bullet q}_1&a^{\bullet q}_2&a^{\bullet q}_3\\ \\ a^{\bullet l}_1&a^{\bullet l}_2&a^{\bullet l}_3 \end{vmatrix} =a_i^{\bullet p}a_j^{\bullet q}a_k^{\bullet l}e^{ijk} =a_i^{\bullet 1}a_j^{\bullet 2}a_k^{\bullet 3}e^{ijk}e^{pql} =det(a_m^{\bullet n})e^{pql}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a_{\bullet 1}^p&a_{\bullet 2}^p&a_{\bullet 3}^p\\ \\ a_{\bullet 1}^q&a_{\bullet 2}^q&a_{\bullet 3}^q\\ \\ a_{\bullet 1}^l&a_{\bullet 2}^l&a_{\bullet 3}^l\\ \end{vmatrix} =a_{\bullet i}^pa_{\bullet j}^qa_{\bullet k}^le^{ijk} =a_{\bullet i}^1a_{\bullet j}^2a_{\bullet k}^3e^{ijk}e^{pql} =det(a_{\bullet n}^m)e^{pql}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a_{p1}&a_{p2}&a_{p3}\\ \\ a_{q1}&a_{q2}&a_{q3}\\ \\ a_{l1}&a_{l2}&a_{l3}\\ \end{vmatrix} =a_{pi}a_{qj}a_{lk}e^{ijk} =a_{1i}a_{2j}a_{3k}e^{ijk}e_{pql} =det(a_{mn})e_{pql}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a^{p1}&a^{p2}&a^{p3}\\ \\ a^{q1}&a^{q2}&a^{q3}\\ \\ a^{l1}&a^{l2}&a^{l3}\\ \end{vmatrix} =a^{pi}a^{qj}a^{lk}e_{ijk} =a^{1i}a^{2j}a^{3k}e_{ijk}e^{pql} =det(a^{mn})e^{pql}
∣
∣a1∙pa1∙qa1∙la2∙pa2∙qa2∙la3∙pa3∙qa3∙l∣
∣=ai∙paj∙qak∙leijk=ai∙1aj∙2ak∙3eijkepql=det(am∙n)epql ∣
∣a∙1pa∙1qa∙1la∙2pa∙2qa∙2la∙3pa∙3qa∙3l∣
∣=a∙ipa∙jqa∙kleijk=a∙i1a∙j2a∙k3eijkepql=det(a∙nm)epql ∣
∣ap1aq1al1ap2aq2al2ap3aq3al3∣
∣=apiaqjalkeijk=a1ia2ja3keijkepql=det(amn)epql ∣
∣ap1aq1al1ap2aq2al2ap3aq3al3∣
∣=apiaqjalkeijk=a1ia2ja3keijkepql=det(amn)epql（2）任意更换行列式两列：
∣
a
p
∙
1
a
q
∙
1
a
l
∙
1
a
p
∙
2
a
q
∙
2
a
l
∙
2
a
p
∙
3
a
q
∙
3
a
l
∙
3
∣
=
a
p
∙
i
a
q
∙
j
a
l
∙
k
e
i
j
k
=
a
1
∙
i
a
2
∙
j
a
3
∙
k
e
i
j
k
e
p
q
l
=
d
e
t
(
a
m
∙
n
)
e
p
q
l
∣
a
∙
p
1
a
∙
q
1
a
∙
l
1
a
∙
p
2
a
∙
q
2
a
∙
l
2
a
∙
p
3
a
∙
q
3
a
∙
l
3
∣
=
a
∙
p
i
a
∙
q
j
a
∙
l
k
e
i
j
k
=
a
∙
1
i
a
∙
2
j
a
∙
3
k
e
i
j
k
e
p
q
l
=
d
e
t
(
a
∙
n
m
)
e
p
q
l
∣
a
1
p
a
1
q
a
1
l
a
2
p
a
2
q
a
2
l
a
3
p
a
3
q
a
3
l
∣
=
a
i
p
a
j
q
a
k
l
e
i
j
k
=
a
i
1
a
j
2
a
k
3
e
i
j
k
e
p
q
l
=
d
e
t
(
a
m
n
)
e
p
q
l
∣
a
1
p
a
1
q
a
1
l
a
2
p
a
2
q
a
2
l
a
3
p
a
3
q
a
3
l
∣
=
a
i
p
a
j
q
a
k
l
e
i
j
k
=
a
i
1
a
j
2
a
k
3
e
i
j
k
e
p
q
l
=
d
e
t
(
a
m
n
)
e
p
q
l
\begin{vmatrix} a^{\bullet 1}_p&a^{\bullet 1}_q&a^{\bullet 1}_l\\ \\ a^{\bullet 2}_p&a^{\bullet 2}_q&a^{\bullet 2}_l\\ \\ a^{\bullet 3}_p&a^{\bullet 3}_q&a^{\bullet3}_l \end{vmatrix} =a_p^{\bullet i}a_q^{\bullet j}a_l^{\bullet k}e_{ijk} =a_1^{\bullet i}a_2^{\bullet j}a_3^{\bullet k}e_{ijk}e_{pql} =det(a_m^{\bullet n})e_{pql}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a_{\bullet p}^1&a_{\bullet q}^1&a_{\bullet l}^1\\ \\ a_{\bullet p}^2&a_{\bullet q}^2&a_{\bullet l}^2\\ \\ a_{\bullet p}^3&a_{\bullet q}^3&a_{\bullet l}^3 \end{vmatrix} =a_{\bullet p}^ia_{\bullet q}^ja_{\bullet l}^ke_{ijk} =a_{\bullet 1}^ia_{\bullet 2}^ja_{\bullet 3}^ke_{ijk}e_{pql} =det(a_{\bullet n}^m)e_{pql}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a_{1p}&a_{1q}&a_{1l}\\ \\ a_{2p}&a_{2q}&a_{2l}\\ \\ a_{3p}&a_{3q}&a_{3l}\\ \end{vmatrix} =a_{ip}a_{jq}a_{kl}e^{ijk} =a_{i1}a_{j2}a_{k3}e^{ijk}e_{pql} =det(a_{mn})e_{pql}\\\ \\ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a^{1p}&a^{1q}&a^{1l}\\ \\ a^{2p}&a^{2q}&a^{2l}\\ \\ a^{3p}&a^{3q}&a^{3l}\\ \end{vmatrix} =a^{ip}a^{jq}a^{kl}e_{ijk} =a^{i1}a^{j2}a^{k3}e_{ijk}e^{pql} =det(a^{mn})e^{pql}
∣
∣ap∙1ap∙2ap∙3aq∙1aq∙2aq∙3al∙1al∙2al∙3∣
∣=ap∙iaq∙jal∙keijk=a1∙ia2∙ja3∙keijkepql=det(am∙n)epql ∣
∣a∙p1a∙p2a∙p3a∙q1a∙q2a∙q3a∙l1a∙l2a∙l3∣
∣=a∙pia∙qja∙lkeijk=a∙1ia∙2ja∙3keijkepql=det(a∙nm)epql ∣
∣a1pa2pa3pa1qa2qa3qa1la2la3l∣
∣=aipajqakleijk=ai1aj2ak3eijkepql=det(amn)epql ∣
∣a1pa2pa3pa1qa2qa3qa1la2la3l∣
∣=aipajqakleijk=ai1aj2ak3eijkepql=det(amn)epql（3）任意同时更换行列式两行两列：
∣
a
p
∙
i
a
q
∙
i
a
l
∙
i
a
p
∙
j
a
q
∙
j
a
l
∙
j
a
p
∙
k
a
q
∙
k
a
l
∙
k
∣
=
d
e
t
(
a
m
∙
n
)
e
i
j
k
e
p
q
l
∣
a
∙
p
i
a
∙
q
i
a
∙
l
i
a
∙
p
j
a
∙
q
j
a
∙
l
j
a
∙
p
k
a
∙
q
k
a
∙
l
k
∣
=
d
e
t
(
a
∙
n
m
)
e
i
j
k
e
p
q
l
∣
a
i
p
a
i
q
a
i
l
a
j
p
a
j
q
a
j
l
a
k
p
a
k
q
a
k
l
∣
=
d
e
t
(
a
m
n
)
e
i
j
k
e
p
q
l
∣
a
i
p
a
i
q
a
i
l
a
j
p
a
j
q
a
j
l
a
k
p
a
k
q
a
k
l
∣
=
d
e
t
(
a
m
n
)
e
i
j
k
e
p
q
l
\begin{vmatrix} a^{\bullet i}_p&a^{\bullet i}_q&a^{\bullet i}_l\\ \\ a^{\bullet j}_p&a^{\bullet j}_q&a^{\bullet j}_l\\ \\ a^{\bullet k}_p&a^{\bullet k}_q&a^{\bullet k}_l\\ \end{vmatrix}= det(a^{\bullet n}_m)e^{ijk}e_{pql}\\\ \\%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a_{\bullet p}^i&a_{\bullet q}^i&a_{\bullet l}^i\\ \\ a_{\bullet p}^j&a_{\bullet q}^j&a_{\bullet l}^j\\ \\ a_{\bullet p}^k&a_{\bullet q}^k&a_{\bullet l}^k \end{vmatrix} =det(a_{\bullet n}^m)e^{ijk}e_{pql}\\\ \\%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a_{ip}&a_{iq}&a_{il}\\ \\ a_{jp}&a_{jq}&a_{jl}\\ \\ a_{kp}&a_{kq}&a_{kl}\\ \end{vmatrix} =det(a_{mn})e_{ijk}e_{pql} \\\ \\ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{vmatrix} a^{ip}&a^{iq}&a^{il}\\ \\ a^{jp}&a^{jq}&a^{jl}\\ \\ a^{kp}&a^{kq}&a^{kl}\\ \end{vmatrix} =det(a^{mn})e^{ijk}e^{pql}
∣
∣ap∙iap∙jap∙kaq∙iaq∙jaq∙kal∙ial∙jal∙k∣
∣=det(am∙n)eijkepql ∣
∣a∙pia∙pja∙pka∙qia∙qja∙qka∙lia∙lja∙lk∣
∣=det(a∙nm)eijkepql ∣
∣aipajpakpaiqajqakqailajlakl∣
∣=det(amn)eijkepql ∣
∣aipajpakpaiqajqakqailajlakl∣
∣=det(amn)eijkepql
2.3 置换张量
借助置换符号次序任意排列的基矢量混合积为：
ϵ
i
j
k
=
g
⃗
i
∙
(
g
⃗
j
×
g
⃗
k
)
=
g
e
i
j
k
ϵ
i
j
k
=
g
⃗
i
∙
(
g
⃗
j
×
g
⃗
k
)
=
1
g
e
i
j
k
\epsilon_{ijk}=\vec{g}_i\bullet(\vec{g}_j\times\vec{g}_k)=\sqrt{g}e_{ijk}\\\ \\ \epsilon^{ijk}=\vec{g}^i\bullet(\vec{g}^j\times\vec{g}^k)=\frac{1}{\sqrt{g}}e^{ijk}
ϵijk=gi∙(gj×gk)=geijk ϵijk=gi∙(gj×gk)=g1eijk由于
ϵ
i
′
j
′
k
′
=
g
⃗
i
′
∙
(
g
⃗
j
′
×
g
⃗
k
′
)
=
β
i
′
p
g
⃗
p
∙
(
β
j
′
q
g
⃗
q
×
β
k
′
l
g
⃗
l
)
=
β
i
′
p
β
j
′
q
β
k
′
l
[
g
⃗
p
∙
(
g
⃗
q
×
g
⃗
l
)
]
=
β
i
′
p
β
j
′
q
β
k
′
l
ϵ
p
q
l
ϵ
i
′
j
′
k
′
=
g
⃗
i
′
∙
(
g
⃗
j
′
×
g
⃗
k
′
)
=
β
p
i
′
g
⃗
p
∙
(
β
q
j
′
g
⃗
q
×
β
l
k
′
g
⃗
l
)
=
β
p
i
′
β
q
j
′
β
l
k
′
[
g
⃗
p
∙
(
g
⃗
q
×
g
⃗
l
)
]
=
β
p
i
′
β
q
j
′
β
l
k
′
ϵ
p
q
l
\epsilon_{i'j'k'} =\vec{g}_{i'}\bullet(\vec{g}_{j'}\times\vec{g}_{k'}) =\beta_{i'}^p\vec{g}_p\bullet(\beta_{j'}^q\vec{g}_{q}\times\beta_{k'}^l\vec{g}_{l}) =\beta_{i'}^p\beta_{j'}^q\beta_{k'}^l[\vec{g}_{p}\bullet(\vec{g}_{q}\times\vec{g}_{l})] =\beta_{i'}^p\beta_{j'}^q\beta_{k'}^l\epsilon_{pql}\\\ \\ \epsilon^{i'j'k'} =\vec{g}^{i'}\bullet(\vec{g}^{j'}\times\vec{g}^{k'}) =\beta^{i'}_p\vec{g}^p\bullet(\beta^{j'}_q\vec{g}^{q}\times\beta^{k'}_l\vec{g}^{l}) =\beta^{i'}_p\beta^{j'}_q\beta^{k'}_l[\vec{g}^{p}\bullet(\vec{g}^{q}\times\vec{g}^{l})] =\beta^{i'}_p\beta^{j'}_q\beta^{k'}_l\epsilon^{pql}
ϵi′j′k′=gi′∙(gj′×gk′)=βi′pgp∙(βj′qgq×βk′lgl)=βi′pβj′qβk′l[gp∙(gq×gl)]=βi′pβj′qβk′lϵpql ϵi′j′k′=gi′∙(gj′×gk′)=βpi′gp∙(βqj′gq×βlk′gl)=βpi′βqj′βlk′[gp∙(gq×gl)]=βpi′βqj′βlk′ϵpql知
ϵ
i
j
k
、
ϵ
i
j
k
\epsilon_{ijk}、\epsilon^{ijk}
ϵijk、ϵijk 分别为某三阶张量的协变分量、逆变分量，将之称作置换张量/Eddington张量
ϵ
\epsilon
ϵ。采用实体记法有：
ϵ
=
ϵ
i
j
k
g
⃗
i
g
⃗
j
g
⃗
k
=
ϵ
i
j
k
g
⃗
i
g
⃗
j
g
⃗
k
=
⋯
\epsilon=\epsilon_{ijk}\vec{g}^i\vec{g}^j\vec{g}^k=\epsilon^{ijk}\vec{g}_i\vec{g}_j\vec{g}_k=\cdots
ϵ=ϵijkgigjgk=ϵijkgigjgk=⋯此外，需要指出如下两点：
（1）置换符号不是三阶张量；
（2） g \sqrt{g} g 或 1 g \frac{1}{\sqrt{g}} g1 不是标量。
理由如下：
e
i
′
j
′
k
′
≠
β
i
′
p
β
j
′
q
β
k
′
l
e
p
q
l
g
′
=
g
⃗
1
′
∙
(
g
⃗
2
′
×
g
⃗
3
′
)
=
β
1
′
p
g
⃗
p
∙
(
β
2
′
q
g
⃗
q
×
β
3
′
l
g
⃗
l
)
=
β
1
′
p
β
2
′
q
β
3
′
l
g
e
p
q
l
≠
g
e_{i'j'k'} \ne\beta_{i'}^p\beta_{j'}^q\beta_{k'}^le_{pql} \\\ \\ \sqrt{g'} =\vec{g}_{1'}\bullet(\vec{g}_{2'}\times\vec{g}_{3'}) =\beta_{1'}^p\vec{g}_p\bullet(\beta_{2'}^q\vec{g}_{q}\times\beta_{3'}^l\vec{g}_{l}) =\beta_{1'}^p\beta_{2'}^q\beta_{3'}^l\sqrt{g}e_{pql} \ne\sqrt{g}
ei′j′k′=βi′pβj′qβk′lepql g′=g1′∙(g2′×g3′)=β1′pgp∙(β2′qgq×β3′lgl)=β1′pβ2′qβ3′lgepql=g
2.4 基矢量的叉积
根据协变基矢、逆变基矢互相转换的关系式可知基矢量的叉积可通过其对偶基矢与置换张量来表示，即：
g
⃗
i
×
g
⃗
j
=
(
g
e
i
j
k
)
g
⃗
k
=
ϵ
i
j
k
g
⃗
k
=
g
⃗
i
g
⃗
j
:
ϵ
=
ϵ
:
g
⃗
i
g
⃗
j
g
⃗
i
×
g
⃗
j
=
(
1
g
e
i
j
k
)
g
⃗
k
=
ϵ
i
j
k
g
⃗
k
=
g
⃗
i
g
⃗
j
:
ϵ
=
ϵ
:
g
⃗
i
g
⃗
j
\vec{g}_i\times\vec{g}_j=(\sqrt{g}e_{ijk})\vec{g}^k=\epsilon_{ijk}\vec{g}^k=\vec{g}_i\vec{g}_j:\epsilon=\epsilon:\vec{g}_i\vec{g}_j\\\ \\ \vec{g}^i\times\vec{g}^j=(\frac{1}{\sqrt{g}}e^{ijk})\vec{g}_k=\epsilon^{ijk}\vec{g}_k=\vec{g}^i\vec{g}^j:\epsilon=\epsilon:\vec{g}^i\vec{g}^j
gi×gj=(geijk)gk=ϵijkgk=gigj:ϵ=ϵ:gigj gi×gj=(g1eijk)gk=ϵijkgk=gigj:ϵ=ϵ:gigj
2.5 ϵ ∼ δ \epsilon \sim \delta ϵ∼δ恒等式
∣
δ
p
i
δ
q
i
δ
l
i
δ
p
j
δ
q
j
δ
l
j
δ
p
k
δ
q
k
δ
l
k
∣
=
d
e
t
(
δ
n
m
)
e
i
j
k
e
p
q
l
=
e
i
j
k
e
p
q
l
=
ϵ
i
j
k
ϵ
p
q
l
=
δ
p
q
l
i
j
k
=
δ
p
i
δ
q
j
δ
l
k
+
δ
l
i
δ
p
j
δ
q
k
+
δ
q
i
δ
l
j
δ
p
k
−
δ
p
i
δ
l
j
δ
q
k
−
δ
l
i
δ
q
j
δ
p
k
−
δ
q
i
δ
p
j
δ
l
k
\begin{vmatrix} \delta^{i}_p&\delta^{i}_q&\delta^{i}_l\\ \\ \delta^{j}_p&\delta^{j}_q&\delta^{j}_l\\ \\ \delta^{k}_p&\delta^{k}_q&\delta^{k}_l\\ \end{vmatrix} =det(\delta^m_n)e^{ijk}e_{pql} =e^{ijk}e_{pql} =\epsilon^{ijk}\epsilon_{pql} =\delta^{ijk}_{pql} =\delta^{i}_{p}\delta^{j}_{q}\delta^{k}_{l}+\delta^{i}_{l}\delta^{j}_{p}\delta^{k}_{q}+\delta^{i}_{q}\delta^{j}_{l}\delta^{k}_{p}-\delta^{i}_{p}\delta^{j}_{l}\delta^{k}_{q}-\delta^{i}_{l}\delta^{j}_{q}\delta^{k}_{p}-\delta^{i}_{q}\delta^{j}_{p}\delta^{k}_{l}
∣
∣δpiδpjδpkδqiδqjδqkδliδljδlk∣
∣=det(δnm)eijkepql=eijkepql=ϵijkϵpql=δpqlijk=δpiδqjδlk+δliδpjδqk+δqiδljδpk−δpiδljδqk−δliδqjδpk−δqiδpjδlk将
δ
p
q
l
i
j
k
\delta^{ijk}_{pql}
δpqlijk 称作广义Kronecker
δ
\delta
δ，显然 广义Kronecker
δ
\delta
δ 为六阶张量，因其满足坐标转换关系：
δ
p
′
q
′
l
′
i
′
j
′
k
′
=
ϵ
i
′
j
′
k
′
ϵ
p
′
q
′
l
′
=
β
a
i
′
β
b
j
′
β
c
k
′
ϵ
a
b
c
β
p
′
d
β
q
′
e
β
l
′
f
ϵ
d
e
f
\delta^{i'j'k'}_{p'q'l'} =\epsilon^{i'j'k'}\epsilon_{p'q'l'} =\beta^{i'}_{a}\beta^{j'}_{b}\beta^{k'}_{c}\epsilon^{abc}\beta_{p'}^{d}\beta_{q'}^{e}\beta_{l'}^{f}\epsilon_{def}
δp′q′l′i′j′k′=ϵi′j′k′ϵp′q′l′=βai′βbj′βck′ϵabcβp′dβq′eβl′fϵdef我们还可进一步得到：
e
i
j
k
e
p
q
k
=
ϵ
i
j
k
ϵ
p
q
k
=
δ
p
q
k
i
j
k
=
δ
p
q
i
j
=
δ
p
i
δ
q
j
−
δ
q
i
δ
p
j
e
i
j
k
e
p
j
k
=
ϵ
i
j
k
ϵ
p
j
k
=
δ
p
j
k
i
j
k
=
2
δ
p
i
e
i
j
k
e
i
j
k
=
ϵ
i
j
k
ϵ
i
j
k
=
δ
i
j
k
i
j
k
=
3
!
e^{ijk}e_{pqk} =\epsilon^{ijk}\epsilon_{pqk} =\delta^{ijk}_{pqk} =\delta^{ij}_{pq} =\delta^{i}_{p}\delta^{j}_{q}-\delta^{i}_{q}\delta^{j}_{p}\\\ \\ e^{ijk}e_{pjk} =\epsilon^{ijk}\epsilon_{pjk} =\delta^{ijk}_{pjk} =2\delta^{i}_{p}\\\ \\ e^{ijk}e_{ijk} =\epsilon^{ijk}\epsilon_{ijk} =\delta^{ijk}_{ijk} =3!
eijkepqk=ϵijkϵpqk=δpqkijk=δpqij=δpiδqj−δqiδpj eijkepjk=ϵijkϵpjk=δpjkijk=2δpi eijkeijk=ϵijkϵijk=δijkijk=3!
2.6 二维置换张量
二维空间为三维空间的子空间，可在二维的基矢
g
⃗
1
,
g
⃗
2
\vec{g}_1,\vec{g}_2
g1,g2 上引入与之垂直的第三个单位基矢
i
⃗
3
\vec{i}_3
i3，则
ϵ
α
β
=
ϵ
α
β
3
=
g
e
α
β
3
(
α
,
β
=
1
,
2
)
ϵ
α
β
=
ϵ
α
β
3
=
1
g
e
α
β
3
(
α
,
β
=
1
,
2
)
其中，
g
=
(
g
⃗
1
×
g
⃗
2
)
∙
i
⃗
3
=
∣
g
⃗
1
×
g
⃗
2
∣
\epsilon_{\alpha\beta}=\epsilon_{\alpha\beta3}=\sqrt{g}e_{\alpha\beta3}\quad(\alpha,\beta=1,2)\\\ \\ \epsilon^{\alpha\beta}=\epsilon^{\alpha\beta3}=\frac{1}{\sqrt{g}}e^{\alpha\beta3}\quad(\alpha,\beta=1,2)\\\ \\ 其中，\sqrt{g}=(\vec{g}_1\times\vec{g}_2)\bullet\vec{i}_3=|\vec{g}_1\times\vec{g}_2|
ϵαβ=ϵαβ3=geαβ3(α,β=1,2) ϵαβ=ϵαβ3=g1eαβ3(α,β=1,2) 其中，g=(g1×g2)∙i3=∣g1×g2∣写作矩阵形式可有：
[
ϵ
α
β
]
=
[
0
g
−
g
0
]
[
ϵ
α
β
]
=
[
0
1
g
−
1
g
0
]
[\epsilon_{\alpha\beta}] =\begin{bmatrix}0&\sqrt{g}\\-\sqrt{g}&0\end{bmatrix} \qquad [\epsilon^{\alpha\beta}] =\begin{bmatrix}0&\frac{1}{\sqrt{g}}\\-\frac{1}{\sqrt{g}}&0\end{bmatrix}
[ϵαβ]=[0−gg0][ϵαβ]=[0−g1g10]对应可得到二维的置换符号：
e
α
β
=
e
α
β
3
=
e
α
β
=
e
α
β
3
=
{
1
(
α
=
1
,
β
=
2
)
−
1
(
α
=
2
,
β
=
1
)
0
(
α
=
β
)
e_{\alpha\beta}=e_{\alpha\beta3}=e^{\alpha\beta}=e^{\alpha\beta3}= \begin{cases}1&(\alpha=1,\beta=2)\\-1&(\alpha=2,\beta=1)\\0&(\alpha=\beta)\end{cases}
eαβ=eαβ3=eαβ=eαβ3=⎩
⎨
⎧1−10(α=1,β=2)(α=2,β=1)(α=β)同样得到二维
ϵ
∼
δ
\epsilon \sim \delta
ϵ∼δ恒等式：
∣
δ
p
i
δ
q
i
δ
p
j
δ
q
j
∣
=
d
e
t
(
δ
β
α
)
e
i
j
e
p
q
=
e
i
j
e
p
q
=
ϵ
i
j
ϵ
p
q
=
δ
p
i
δ
q
j
−
δ
q
i
δ
p
j
e
i
j
e
p
j
=
ϵ
i
j
ϵ
p
j
=
2
δ
p
i
e
i
j
e
i
j
=
ϵ
i
j
ϵ
i
j
=
3
!
\begin{vmatrix} \delta^{i}_p&\delta^{i}_q\\ \\ \delta^{j}_p&\delta^{j}_q \end{vmatrix} =det(\delta^{\alpha}_{\beta})e^{ij}e_{pq} =e^{ij}e_{pq} =\epsilon^{ij}\epsilon_{pq} =\delta^{i}_p\delta^{j}_q-\delta^{i}_q\delta^{j}_p \\\ \\ e^{ij}e_{pj} =\epsilon^{ij}\epsilon_{pj} =2\delta^{i}_p \\\ \\ e^{ij}e_{ij} =\epsilon^{ij}\epsilon_{ij} =3!
∣
∣δpiδpjδqiδqj∣
∣=det(δβα)eijepq=eijepq=ϵijϵpq=δpiδqj−δqiδpj eijepj=ϵijϵpj=2δpi eijeij=ϵijϵij=3!
1322
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
