3.1.5. Differential
Let .
We define a linear map
by
|
|
|
0P4P
Proposition 3.1.3. The map defines a structure of differential graded algebra on
.
0P4Q
Proof. Let and with .
We have
. We have
[Hu, Theorem 5.10]
|
|
|
It follows that .
Consider now and with .
We have by the result above.
It follows that .
We deduce that for all
.
Since for , it follows that by induction that .
β
The following corollary shows that the computation of can be done
using the Leibniz rule, given a reduced decomposition of . The terms that do not
vanish are exactly the terms given in the original definition of .
0P4R
Corollary 3.1.4. Let be a reduced expression of .
We have
|
|
|
We have if and only if is reduced,
i.e., if and only if .
Given with reduced, we have
.
0P4S
Proof. The first statement follows from Proposition 3.1.3.
The second statement is a property of the multiplication of βs.
For the third statement, let us assume . We have
reduced, hence
is not reduced, a contradiction.
β
The specialization over
at of the bimodules of Β§3.1.3 acquire a
structure of differential graded bimodules, using the differential graded structure
of .
We keep
the same notation for those differential graded specialized bimodules and for the maps and
.
0P4U
Proposition 3.1.6. If is finite, then
|
|
|
is a morphism of differential graded -modules and
Corollary 3.1.2 provides an isomorphism of differential graded
-bimodules
|
|
|
0P4V
Proof. Let . There is a unique decomposition where
, and
.
We have . If and
, then . It follows that
|
|
|
β