Given non-zero elements and in a field , the quaternion algebra over a field is the 4-dimensional vector space with the relations
In characteristic , the quaternion algebra is defined by the relations
Any algebra generated by non-zero elements , satisfying these relations with and non-zero has to be the quaternion algebra, that is it will be 4-dimensional and generated by , and .
If or are zero, we wont get a quaternion algebra. For instance modulo the relations is just 3-dimensional.
We denote by , then we have when .
We assume from now, in characteristic 2 we have to make some modifications to all the statements and arguments. (Basically because quadratic forms over characteristic behave differently)
Examples: a) Hamiton’s quaternions are . They are closely related to the rotations , and the group .
b) The algebra is isomorphic to the algebra of matrices by the following identification
c) is a division algebra. But is equivalent to is the matrix algebra.
In general, can be viewed as a sub-algebra of a matrix algebra (possibly over a extension field) by the following identification
If , the algebra is isomorphic to
Note that we have the isomorphism
because we can simply interchange and .
We can also consider the basis or which gives us the equivalences
,
We can also rescale and to and to see that
The equivalence allows us to see that
So we can assume to be representatives of the square classes . Also if we extend the scalars to field which has square roots of at least one and , the algebra will be isomorphic to or both of which are equivalent to the matrix algebra Such a field is called splitting field of algebra.
For example, everything is a square in , so is the only quaternion algebra over . Over reals, we have to just consider , and gives the Hamilton’s algebra and every other choice is of the form , so equivalent to But both the algebra when we extend the class become isomorphic to . That is and
When , is square and hence
Conjugates, Norm, Trace:
We have the conjugate map
And the norm
If the norm doesn’t represent zero non-trivially, then every non-zero element can be inverted.
Therefore every element satisfies a quadratic equation with coefficient in given by
We call the trace. So pure quaternions with are the trace zero elements. Trace zero elements have squares which are in , in fact only non-scalar elements with squares in the field.
Thus we can think quaternions as an noncommutative analog of quadratic field extensions. In fact, the only quadratic division algebras (all elements satisfy degree at most equations) are the quaternions and quadratic field extensions. Division property is important because otherwise a lot degree 2 algebras exist.
Question: When do we get algebras which are not split, that is not isomorphic to ?
For instance if the norm form doesn’t represent zero, the algebra is a division algebra. So it cannot be the matrix algebra because the matrix algebra has zero divisors i.e., .
The quadratic form is the norm form.
So if the form doesn’t represent zero, the algebra cannot be split. And It can also be seen that if this form represents zero non-trivially, the algebra is the matrix algebra.
Proof: If is square, the algebra is split. So assume is not a square and the norm represent zero, that is we have a non-trivial solution to which implies that
Now check that is a basis for the algebra with . This is a basis because the determinant
So the algebra is equivalent to
So we have two cases: Either
a) the algebra is a division algebra, doesn’t represent zero non-trivially.
b) The algebra is the matrix algebra .
We also consider the forms and
represents zero non-trivially iff represents zero non-trivially. And both the forms represent non-trivially iff represent
Proof: One side is obvious. So assume has a non-trivial solution with . Let and consider orthogonal (with the bilinear form associated to the quadratic form) to the both and . We have traces of and both zero. Because , one of and is non-zero and we found a trace zero element with norm 0.
If we have a solution with for the second form , we get
, and we found a solution to . So we consider the case , so that we have non-trivial solution to . In this case and hence , which represents .
So either we have a split matrix algebra, or a division algebra. To check if it’s a division algebra, we have to understand the properties of these quadratic forms.
We finally see reason for the notation . The Hilbert symbol equals iff represent , which happens if and only if the algebra is split, that is equal to .
Example: Consider where is not a square . By the above discussion we need to check if has rational solutions. But a rational solution will imply that , which gives , and is not possible because For example if , then is not square which implies that
We saw the definining criterion for to be split matrix algebra. have solutions in . Some special cases where this happens are
1) Either of , is , or in general either of them is a square.
2)
3) If and .
4) (This is equivalent to , and hence the most general case)
Alternatively if we find with , then we have a division algebra.
Example: 1) The algebra is a division algebra iff there is not sums of squares.
2) Any quaternion algebra over for odd is split. This is because and as vary have to intersect to give a solution to . ( both of them are of size )
Question: When are the algebras different (or the same)?
We saw some conditions on for to be the matrix algebra, has to of the form . Similarly we and is the same quaternion algebra iff . The matrix algebra corresponds to taking .
Thus scaling by an element of the form doesn’t change the algebra. Also two elements and different modulo the subgroup of elements of the form represent different quaternion algebras. To prove this we need to show there is a conjugation map on the to send to the basic element of satisfying
For example , for represents different quaternion algebras as because cannot be written as . So we have infinitely many distinct division algebras over !
The case of , we also other isomorphism between and , and the quaternion algebra is determined by the discriminant.