## Full graph and complete Lie algebra

### 2003.10

Abstract
We introduce a concept: $d$  -complete, and show that a Lie algebra is $d$  -complete if and only if its full graph is complete.
Key words and phrases $d$  -derivation, $d$  -complete Lie algebra, complete Lie algebra, full graph MR(2000) Subject Classification: 17B45, 17B55, 17B56 Let $\mathcal{G}$  be a Lie algebra. We will introduce two concepts: $d$  -derivation and $d$  -center. Let $Der\left(\mathcal{G}\right)$  be the derivation algebra of $\mathcal{G}$  .
A $d$  -derivation of $\mathcal{G}$  , means a linear operator $L$  from $Der\left(\mathcal{G}\right)$  to $\mathcal{G}$  such that
 $\begin{array}{c}L\left(\left[{D}_{1},{D}_{2}\right]\right)={D}_{1}\left(L\left({D}_{2}\right)\right)-{D}_{2}\left(L\left({D}_{1}\right)\right)\forall {D}_{1},{D}_{2}\in Der\left(\mathcal{G}\right)\end{array}$ (1).
We write $\mathcal{D}$  for the set of $d$  -derivations.
The $d$  -center of $\mathcal{G}$  , means the set
 $\begin{array}{c}{C}_{d}\left(\mathcal{G}\right)=.\left\{g\in \mathcal{G}:D\left(g\right)=0\forall D\in Der\left(\mathcal{G}\right)\right\}\end{array}$ (2).
For any $g\in \mathcal{G}$  , we can define a $d$  -derivation ${L}_{g}$  of $\mathcal{G}$  :
 $\begin{array}{c}{L}_{g}\left(D\right)=-D\left(g\right)\forall D\in Der\left(\mathcal{G}\right)\end{array}$ (3).
Such a $d$  -derivation is called inner $d$  -derivation.
For any two $d$  -derivations ${L}_{1}$  , ${L}_{2}$  , define $\left[{L}_{1},{L}_{2}\right]$  by
 $\begin{array}{c}\left[{L}_{1},{L}_{2}\right]\left(D\right)={L}_{1}\left(ad\left({L}_{2}\left(D\right)\right)\right)-{L}_{2}\left(ad\left({L}_{1}\left(D\right)\right)\right)\forall D\in Der\left(\mathcal{G}\right)\end{array}$ (4),
then $\left[{L}_{1},{L}_{2}\right]$  is a $d$  -derivation. And it is easy to check that ( $\mathcal{D}$  ,[,]) is a Lie algebra. If ${C}_{d}\left(\mathcal{G}\right)$  is trivial, $\mathcal{G}$  is a Lie subalgebra of $\mathcal{D}$  .
If ${C}_{d}\left(\mathcal{G}\right)$  is trivial and all $d$  -derivations are inner, $\mathcal{G}$  is called $d$  -complete. In this case, there is a isomorphic from $\mathcal{G}$  to $\mathcal{D}$  : $g↦{L}_{g}$  .
We have a homomorphic from $Der\left(\mathcal{G}\right)$  to the derivation algebra $Der\left(\mathcal{D}\right)$  of $\mathcal{D}$  by setting:
 $\begin{array}{c}D\left(L\right)=.D\circ L-L\circ ad\left(D\right)\forall L\in \mathcal{D},D\in Der\left(\mathcal{G}\right)\end{array}$ (5).
We can now define semidirect product $\mathcal{ℋ}=Der\left(\mathcal{G}\right){×}_{t}\mathcal{D}$  , a Lie algebra, by
 $\begin{array}{c}\left[\left({D}_{1},{L}_{1}\right),\left({D}_{2},{L}_{2}\right)\right]=\left(\left[{D}_{1},{D}_{2}\right],\left[{L}_{1},{L}_{2}\right]+{D}_{1}\left({L}_{2}\right)-{D}_{2}\left({L}_{1}\right)\right)\end{array}$ (6).
We can make $\mathcal{ℋ}$  act on $C\left(\mathcal{G}\right)=Der\left(\mathcal{G}\right){×}_{t}\mathcal{D}$  as derivations:
 $\begin{array}{c}{D}_{\left(D,L\right)}\left({D}_{1},g\right)=\left(\left[D,{D}_{1}\right],D\left(g\right)+L\left(ad\left(g\right)\right)+L\left({D}_{1}\right)\right)\end{array}$ (7).
In fact we have:
Theorem 1. $\mathcal{ℋ}$  is the derivation algebra $Der\left(C\left(\mathcal{G}\right)\right)$  of the full graph $C\left(\mathcal{G}\right)$  of $\mathcal{G}$  .
• Proof. A direct calculation shows that 1. ${D}_{\left(D,L\right)}$  is a derivation, 2. $\left[{D}_{\left({D}_{1},{L}_{1}\right)},{D}_{\left({D}_{2},{L}_{2}\right)}\right]={D}_{\left[\left({D}_{1},{L}_{1}\right),\left({D}_{2},{L}_{2}\right)\right]}$  , 3. ${D}_{\left(D,L\right)}=0$  if and only if $D=0$  and $L=0$  .
So $\mathcal{ℋ}$  is a Lie subalgebra of $Der\left(C\left(\mathcal{G}\right)\right)$  .
Now , we will show that every derivation $D$  of $Der\left(C\left(\mathcal{G}\right)\right)$  is in $\mathcal{ℋ}$  . We write $Der\left(\mathcal{G}\right)$  for $\left(Der\left(\mathcal{G}\right),0\right)$  and $\mathcal{G}$  for $\left(0,\mathcal{G}\right)$  . Write $D=\left({T}_{1},{T}_{2}\right)$  . First we show that ${T}_{1}|\mathcal{G}=0$  .
By  $\begin{array}{c}D\left(\left[\left(0,{g}_{1}\right),\left(D,{g}_{2}\right)\right]\right)=\left[D\left(0,{g}_{1}\right),\left(D,{g}_{2}\right)\right]+\left[\left(0,{g}_{1}\right),D\left(D,{g}_{2}\right)\right]\end{array}$ (8),
we obtain  $\begin{array}{c}-{T}_{1}\left(D\left({g}_{1}\right)\right)+{T}_{1}\left(\left[{g}_{1},{g}_{2}\right]\right)=\left[{T}_{1}\left({g}_{1}\right),D\right]\end{array}$ (9).
Formula (9) shows that  $\begin{array}{c}{T}_{1}|\left[\mathcal{G},\mathcal{G}\right]=0\end{array}$ (10)
and that  $\begin{array}{c}-{T}_{1}\left(D\left(g\right)\right)=\left[{T}_{1}\left(g\right),D\right]\end{array}$ (11).
Therefore $\left[{T}_{1}\left(g\right),D\right]=0\forall D\in ad\left(\mathcal{G}\right).$  So ${T}_{1}\left(g\right)=0$  and ${T}_{1}|\mathcal{G}=0$  .
We have ${T}_{2}\left(\left[{D}_{1},{D}_{2}\right]\right)={D}_{1}\left({T}_{2}\left({D}_{2}\right)\right)-{D}_{2}\left({T}_{2}\left({D}_{1}\right)\right).$  So ${T}_{2}|Der\left(\mathcal{G}\right)=L\in \mathcal{D}$  .
By setting $\stackrel{~}{D}=D-{D}_{\left(0,L\right)}$  , we may assume that ${T}_{2}|Der\left(\mathcal{G}\right)=0$  . We know that ${T}_{1}|Der\left(\mathcal{G}\right)$  is a derivation of $Der\left(\mathcal{G}\right)$  .
Since ${T}_{1}|\mathcal{G}=0$  , ${T}_{2}|\mathcal{G}$  is a derivation ${D}^{{}^{\prime }}$  of $\mathcal{G}$  . Now $D\left[\left(D,0\right),\left(0,g\right)\right]=\left[\left(D,0\right),D\left(0,g\right)\right]+\left[D\left(D,0\right),\left(0,g\right)\right],$  shows that ${T}_{2}\left(D\left(g\right)\right)=D\left({T}_{2}\left(g\right)\right)+\left({T}_{1}\left(D\right)\right)\left(g\right),$  that is ${T}_{1}\left(D\right)=\left[{D}^{{}^{\prime }},D\right].$  So $D={D}_{\left({D}^{{}^{\prime }},0\right)}$  . We have now showed that $D\in \mathcal{ℋ}$  .
We have a easy lemma:
Lemma. The center of $C\left(\mathcal{G}\right)$  is $\left(0,{C}_{d}\left(\mathcal{G}\right)\right)$  We have the following corollary of theorem 1 and the above lemma:
Theorem 2. A Lie algebra is $d$  -complete if and only if its full graph is complete.
References: [1] Hsie B.Y., On complete Lie algebra, Science in China, in appear.
[2] Meng D.J., Complete lie algebra(in Chinese), BeiJing: Press of science,2001.
DEPARTMENT OF MATHEMATICS, PEKING UNIVERSITY, BEIJING, P.R. CHINA. E-mail address: xiebingyong1@sina.com.cn