Lecture 5.

Quaternions
Oleg Viro

The lecture on quaternions was given by Professor Alexander Kirillov.

Below you can find a concise list of definitions and statements on this topic.

5.1 Quaternions as quartiples of real numbers

The set of quaternions is denoted by H. This is a very concrete mathematical
object. As a vector space over R, it has the standard basis 1, i, j, k. A
quaternion expanded in this standard basis is a+bi+cj+dk, where a, b, c, d ∈
R. The quaternion addition is component-wise.

(a+bi+cj +dk)+(a0 +b0 i+c0 j +d0 k) = (a+a0 )+(b+b0 )i+(c+c0 )j +(d+d0 )k.

Quaternions can be multiplied by each other. The multiplication is as-

sociative (i.e., (xy)z = x(yz) for any quaternions x, y, z ∈ H) and distribu-
tive (i.e., (x + y)z = xz + yz). The generators are subject to relations
i2 = j 2 = k 2 = ijk = −1.
The quaternion products of the generators are calculated according to
the formulas

ij = k, ji = −k, jk = i, kj = −i, ki = j and ik = −j (1)

These formulas can be deduced from the relations i2 = j 2 = k 2 = ijk = −1

and associativity of multiplication. For example, multiply the last relation
ijk = −1 by k from the right hand side: ijk 2 = −k. Since k 2 = −1, then
−ij = −k. Multiply both sides by −1. This results the first formula ij = k
that we wanted to prove.
Take the square of it: ijij = k 2 = −1. Multiply by i from the left and
by j from the right: i(ijij)j = −ij. The right hand side is −ij = −k. The
left hand side: i2 jij 2 = (−1)ji(−1) = ji. Hence ji = −k.
Exercise: prove the rest of formulas (1).
Notice that the quaternion multiplication is not commutative: ij 6= ji.

1
5.2 Scalars and vectors.
The field R of real numbers is contained in H as {a + 0i + 0j + 0k | a ∈ R}.
A quaternion of the form a + 0i + 0j + 0k, is called real. A quaternion of
the form 0 + bi + cj + dk, where b, c, d ∈ R is called pure imaginary .
If q = a + bi + cj + dk is any quaternion, then a is called its scalar part
or real part and denoted by Re q and bi + cj + dk is called its vector part
and denoted by Ve q. The set of pure imaginary quaternions bi + cj + dk is
identified with the real 3-space R3 .

5.3 Multiplication of quaternions and multiplications of vec-

tors.
A real quaternion commutes with any quaternion. Multiplication of quater-
nions is composed of all the standard multiplications of factors which are
real numbers and vectors: multiplications of real numbers, multiplication of
a vector by a real number and dot and cross products of vectors. It is not
accident: the very notion of vector and all the operations with vectors were
introduced by Hamilton after invention of quaternions. (Many mathemati-
cians nowadays are not aware about this.)
Quaternion product of vectors. Let p = ui+vj+wk and q = xi+yj+zk
be vector quaternions. Then pq = −p · q + p × q.

Proof. Indeed,
pq = (ui + vj + wk)(xi + yj + zk)
= −(ux + vy + wz) + (vz − wy)i + (wx − uz)j + (uy − vx)k
= −p · q + p × q.

Product of arbitrary quaternions via other products. For any

p, q ∈ H

pq = (Re p+Ve p)(Re q+Ve q) = Re p Re q+Re p Ve q+Ve p Re q+Ve p Ve q

= Re p Re q + Re p Ve q + Re q Ve p − Ve p · Ve q + Ve p × Ve q
= Re p Re q − Ve p · Ve q + Re p Ve q + Re q Ve p + Ve p × Ve q

5.4 Conjugation.
The map H → H : q 7→ q̄ = Re q−Ve q is called conjugation. The conjugation
is an antiautomorphism of H in the sense that

2
 • it is invertible (in fact, it coincides with its inverse: (q̄) = q),
• preserve addition, i.e., (p + q) = p̄ + q̄,
• and maps the product to the product of exchanged factors: (pq) = q̄ p̄.
The latter property is verified as follows:
(pq) = (Re p Re q − Ve p · Ve q + Re p Ve q + Re q Ve p + Ve p × Ve q)
= Re p Re q − Ve p · Ve q − Re p Ve q − Re q Ve p − Ve p × Ve q
= Re p Re q−(− Ve p)·(− Ve q)+Re p(− Ve q)+Re q(− Ve p)+(− Ve q)×(− Ve p)
= (Re q − Ve q)(Re p − Ve p) = q̄ p̄.
(Here we used well-known properties of dot and cross products.)

5.5 Norm.
The product q̄q is a real number for any quaternion q.
Indeed, (q̄q) = q̄(q̄) = q̄q.
If q = a + bi + cj + dk, then q ∗ q = a2 + b2 + c2 + d2 .
Indeed, q̄q = Re q 2 − (− Ve q) · Ve q + Re q Ve q + Re q(− Ve q) + (− Ve q) ×
Ve q = Re q 2 + Ve q · Ve q = a2 + b2 + c2 + d2 .
Corollary. For any quaternion q, the product q̄q is non-negative real number.
It is zero if and only if q is zero.
√
The number q ∗ q is called the norm of q and denoted by |q|. This is
the Euclidean distance from q to the origin in H = R4 .
The norm is a multiplicative homomorphism H → R. This means that |pq| =
|p||q| for any quaternions p and q.
√ √ √
q p
Proof. |pq| = pq(pq) = pq q̄ p̄ = p(q q̄)p̄ = pp̄ q q̄ = |p||q|.

5.6 Unit quaternions.

A quaternion q with |q| = 1 is called a unit quaternion. The set of all unit
quaternions {q ∈ H | |q| = 1} is a sphere of radius one in the 4-space H. It
¯
is denoted by S 3 .
The set of unit quaternions is closed under quaternion multiplication,
because the norm of the product of quaternions is the product of norms
of the factors.pThe inverse to a unit quaternion q ∈ S 3 coincides with q̄.
Indeed, |q| = (q q̄) = 1, hence q q̄ = 1 and q̄ = q −1 .
Unit vector quaternions form the unit 2-sphere S 2 in R3 . It is contained
in S 3 as an equator. The unit vectors are very special quaternions.

3
Theorem 1. Each unit quaternion can be presented as a product of two unit
vectors. Moreover, if q is a unit quaternion and v is a unit vector perpendicular
to qv , then there exist unit vectors w+ and w− such that q = vw+ = w− v.

Proof. Let q ∈ S 3 be a unit quaternion. Then q = Re q+Ve q with 1 = |q|2 =

Re q 2 + | Ve q|2 . Choose α ∈ [0, π] such that Re q = cos α and | Ve q| = sin α.
Then q = cos α + u sin α for some unit vector u.
Take any unit vector v perpendicular to u. Then w+ = −v cos α + (u ×
v) sin α and w− = −v cos α−(u×v) sin α are also unit vectors perpendiculars
to u, with the required properties: vw+ = q and w− v = q. Indeed, vw+ =
v(−v cos α+(u×v) sin α) = −v·(−v cos α)+v×(u×v) sin α = cos α+u sin α =
q and w− v = (−v cos α −(u×v) sin α)v = −(−v cos α)·v −(u×v)×v sin α =
cos α + u sin α = q

Remark. Any unit vector quaternion u has order four, its multiplicative
inverse coincides with the additive inverse: u−1 = −u.
Indeed, let u be unit vector. Then u2 = −u · u + u × u = −1, hence u3 = −u
and u4 = (u2 )2 = (−1)2 = 1.
By Theorem 1 any unit quaternion q admits presentation as product of
two unit vector quaternions: q = vw.
A unit quaternion can be presented as a sort of quotient of two unit
vectors: first, present −q as product of two unit vector quaternions: −q =
vw, then re-write this as q = −vw = (−v)w = v −1 w. This presentation
goes back to W.R.Hamilton, the inventor of quaternions. In his book [1],
Hamilton introduced quaternions as quotients of vectors.

5.7 The action of unit quaternions in the 3-space.

A unit quaternion q defines a map ρq : H → H by formula ρq (p) = qpq −1 =
qpq̄. We say that the group S 3 of unit quaternions acts in H.
This action commutes with the conjgation p 7→ p̄.
Indeed,
ρq (p̄) = q p̄q̄ = ((q̄)(p̄)q̄) = (qpq̄) = (ρq (p)).
Therefore the action of S 3 in H preserves all the structures defined by
the congugation. In particular, it preserves the norm and the decomposition
into scalar and vector parts. Indeed,
 
p − p̄ ρq (p) − ρq (p̄) ρq (p) − ρq (p)
ρq (Ve p) = ρq = = = Ve(ρq (p)),
2 2 2

4
  
p + p̄ ρq (p) + ρq (p̄) ρq (p) + ρq (p)
ρq (Re p) = ρq = = = Re(ρq (p)),
2 2 2
q q
|ρq (p)| = ρq (p)(ρq (p)) = ρq (p)ρq (p̄)
p √
= (qpq̄)(q p̄q̄) = qpp̄q̄ = |p|.

In particular, the space R3 of vector quaternions is invariant, and S 3 acts

on R3 by isometries.
Theorem 2. A unit vector quaternion v acts in R3 as the symmetry about
the line generated by v.
Proof. The statement that we are going to prove admits the following re-
formulation: for the linear operator R3 → R3 : u 7→ vuv ∗ , the vector v is
mapped to itself and each unit vector u orthogonal to v is mapped to the
opposite vector −u.
Let us verify the first statement. Since v is a unit vector, vv̄ = |v|2 = 1.
Therefore vvv̄ = v.
Now let us verify the second statement. Since u is a unit vector or-
thogonal to v, vu = v × u − v · u = v × u. Therefore, vuv̄ = −vuv =
−(v × u − v · u)v = −(v × u)v. Vector v × u is orthogonal to v. Therefore
−(v × u)v = −(v × u) × v + (v × u) · v = −(v × u) × v = −u. The latter
equality holds true, because (a × b) × a = b for any orthogonal unit vectors
a, b (e.g., (i × j) × i = k × i = j).

Theorem 3 (Euler-Rodrigues-Hamilton). Let q be any unit quaternion.

Represent it as q = cos 2θ + u sin 2θ . Then the map R3 → R3 : p 7→ qpq̄ is the
rotation of R3 about the axis generated by a unit vector u by the angle θ.
Proof. By Theorem 1 any unit quaternion q can be presented as a product
of unit vectors v and w. In this proof it will be more convenient to use a
modification of this presentation, the fraction presentation q = v −1 w = −vw
discussed above.
By Theorem 2 a unit vector acts as a symmetry about the line generated
by this vector. Thus, ρq is the composition of the symmetries ρ−v and ρw .
The composition of symmetries about lines is a rotation by the angle equal
the half of the angle between the lines. On the other hand, qs = (v(−w))s =
−v · (−w) = v · w = cos α, where α is the angle between the vectors v and
w. Thus qs = cos 2θ , where θ is the rotation angle.
The vector part qv of the product of two unit vectors v and −w is collinear
to v × (−w). The cross product of vectors is perpendicular to the vectors.

5
On the other hand, we know that composition of symmetries about lines is
a rotation about the axis perpendicular to the lines. Thus the vector qv is
collinear to the axis of the rotation ρq . The length |qv | is | sin 2θ |, because
|q| = 1 and qs = cos 2θ . Therefore qv = u sin 2θ for some unit vector u collinear
to the axis of rotation.

The quaternion q can be written down as a + bi + cj + dk. It is defined

by the rotation up to multiplication by −1. Its components a, b, c, d are
called the Euler parameters for this rotation. They are calculated as follows:
a = cos 2θ , b = ux sin 2θ , c = uy sin 2θ and d = uz sin 2θ , where ux , uy and uz
are coordinates of the unit vector u directed along the rotation axis.

References
[1] Elements of quaternions, by the late Sir William Rowan Hamilton, LL.
P., M.R.I.A., edited by his son, William Edwin Hamilton, A.B.T.C.D.,
C.E. London : Longmans, Green, & Co. 1866.