Hamiltonian quaternions
Definition of H
We define a unital associative algebra ℍ over ℝ, of dimension 4,
by the basis {𝟏,𝐢,𝐣,𝐤} and
the multiplication table
1 | i | j | k |
i | -1 | k | -j |
j | -k | -1 | i |
k | j | -i | -1 |
(where the element in row x and column y is xy, not yx). Thus an arbitrary element of ℍ is of the form
a𝟏+b𝐢+c𝐣+d𝐤,a,b,c,d∈ℝ |
(sometimes denoted by ⟨a,b,c,d⟩
or by a+⟨b,c,d⟩) and the product of two elements
⟨a,b,c,d⟩ and ⟨α,β,γ,δ⟩ (order matters)
is ⟨w,x,y,z⟩ where
w | = | aα-bβ-cγ-dδ | ||
x | = | aβ+bα+cδ-dγ | ||
y | = | aγ-bδ+cα+dβ | ||
z | = | aδ+bγ-cβ+dα |
The elements of ℍ are known as Hamiltonian quaternions.
Clearly the subspaces of ℍ generated by {𝟏}
and by {𝟏,𝐢} are subalgebras
isomorphic
to ℝ and ℂ respectively. ℝ is customarily identified with
the corresponding subalgebra of ℍ.
(We shall see in a moment that there are other and less obvious
embeddings
of ℂ in ℍ.)
The real numbers commute with all the elements of ℍ, and we have
λ⋅⟨a,b,c,d⟩=⟨λa,λb,λc,λd⟩ |
for λ∈ℝ and ⟨a,b,c,d⟩∈ℍ.
Like the complex numbers (ℂ), the quaternions have a
natural involution called the quaternion conjugate. If q=a𝟏+b𝐢+c𝐣+d𝐤, then the quaternion
conjugate of q, denoted ˉq, is simply ˉq=a𝟏-b𝐢-c𝐣-d𝐤.
One can readily verify that if q=a𝟏+b𝐢+c𝐣+d𝐤, then qˉq=(a2+b2+c2+d2)𝟏. (See Euler four-square identity.) This product is used to form a norm ∥⋅∥ on the algebra (or the ring) ℍ: We define ∥q∥=√s where qˉq=s𝟏.
If v,w∈ℍ and λ∈ℝ, then
-
1.
∥v∥≥0 with equality only if v=⟨0,0,0,0⟩=0
-
2.
∥λv∥=|λ|∥v∥
-
3.
∥v+w∥≤∥v∥+∥w∥
-
4.
∥v⋅w∥=∥v∥⋅∥w∥
which means that ℍ qualifies as a normed algebra when we give it the norm ∥⋅∥.
Because the norm of any nonzero quaternion q is real and nonzero, we have
qˉq∥q∥2=ˉqq∥q∥2=⟨1,0,0,0⟩ |
which shows that any nonzero quaternion has an inverse:
q-1=ˉq∥q∥2. |
Other embeddings of C into H
One can use any non-zero q to define an embedding of ℂ into ℍ. If 𝐧(z) is a natural embedding of z∈ℂ into ℍ, then the embedding:
z→q𝐧(z)q-1 |
is also an embedding into ℍ. Because ℍ is an associative algebra, it is obvious that:
(q𝐧(a)q-1)(q𝐧(b)q-1)=q(𝐧(a)𝐧(b))q-1 |
and with the distributive laws, it is easy to check that
(q𝐧(a)q-1)+(q𝐧(b)q-1)=q(𝐧(a)+𝐧(b))q-1 |
Rotations in 3-space
Let us write
U={q∈ℍ:|| |
With multiplication, is a group.
Let us briefly sketch the relation between and the group
of rotations (about the origin) in 3-space.
An arbitrary element of can be expressed
,
for some real numbers such that .
The permutation of thus gives rise to a permutation
of the real sphere. It turns out that that permutation is a rotation.
Its axis is the line through and , and the angle
through which it rotates the sphere is .
If rotations and correspond to quaternions and
respectively, then clearly the permutation corresponds
to the composite rotation .
Thus this mapping of onto is a group homomorphism.
Its kernel is the subset of , and thus it comprises
a double cover of . The kernel has a geometric interpretation
as well: two unit vectors
in opposite directions determine the same
axis of rotation.
On the algebraic side, the quaternions provide an example of a division ring that is not a field.
Title | Hamiltonian quaternions |
Canonical name | HamiltonianQuaternions |
Date of creation | 2013-03-22 12:35:42 |
Last modified on | 2013-03-22 12:35:42 |
Owner | mathcam (2727) |
Last modified by | mathcam (2727) |
Numerical id | 10 |
Author | mathcam (2727) |
Entry type | Definition |
Classification | msc 16W99 |
Synonym | quaternion |
Related topic | EulerFourSquareIdentity |
Related topic | QuaternionGroup |
Related topic | HyperkahlerManifold |
Related topic | MathematicalBiology |
Defines | quaternion algebra |