LECTURE 3:
MORE ON VECTOR PRODUCTS
Prof. N. Harnew
University of Oxford
MT 2012
1
Outline: 3. VECTOR PRODUCTS AND GEOMETRY
3.1 Scalar Triple Product
3.1.1 Properties of scalar triple product
3.1.2 Geometrical interpretation
3.2 Vector Triple Product
3.2.1 Lagrange’s identity
3.3 Generating orthogonal axes
2
3.1 Scalar Triple Product
I The scalar triple product, a.(b × c), is the scalar product of
the vector a with the cross products of vectors (b × c).
I The result is a scalar.
I Scalar triple product is also written [a, b, c].
I Scalar triple product in component form :
i j k
a.(b × c) = a. bx by bz
(1)
cx cy cz
= (ax i+ay j+az k).((by cz −bz cy )i−(bx cz −bz cx )j+(bx cy −by cx )k)
= ax (by cz − bz cy ) − ay (bx cz − bz cx ) + az (bx cy − by cx )
ax ay az
In matrix (determinant) form : a.(b × c) = bx by bz
cx cy cz
(2)
3
3.1.1 Properties of scalar triple product
I It is obvious that a.(b × c) = (b × c).a
I Cyclic permutations of a, b and c leaves the triple scalar
product unaltered:
a.(b × c) = c.(a × b) = b.(c × a)
(easily derived by working in components).
I Non-cyclic permutations change sign:
[a, b, c] = [c, a, b] = [b, c, a] =
−[a, c, b] = −[c, b, a] = −[b, a, c]
I The scalar triple product is zero if any two vectors are
parallel.
I The scalar triple product is zero if the three vectors are
coplanar (lie in the same plane).
4
3.1.2 Geometrical interpretation
The triple scalar product can be interpreted as the volume of a
parallelepiped:
I [Volume] = [Area of base] ×
[Vertical height of parallelepiped]
I [Area of base] = |a × b|
(vector direction is
perpendicular to the base)
I [Vertical height]
a×b
= |c| cos φ = c · ( |a×b| )
a×b
I Hence [Volume] = |a × b| c · ( |a×b| ) = c · (a × b)
I Obviously if a, b and c are coplanar, volume = 0.
5
Example
Calculate the volume of a parallelepiped defined by vectors
(1, 1, 2), (1, 3, 2), (−2, 1, 1) from the origin :
I Solution:
−2 1 1
Volume = c.(a × b) = 1 1 2
(3)
1 3 2
= −2(1 × 2 − 3 × 2) − 1(2 × 1 − 2 × 1) + 1(1 × 3 − 1 × 1)
=8−0+2
= 10
6
3.2 Vector Triple Product
I The vector triple vector product, a × (b × c), is the vector
product of the vector a with the cross products of vectors
(b × c).
I The result is a vector.
I This is not associative. i.e. a × (b × c) 6= (a × b) × c.
I Clearly for a × (b × c), the vector lies in the plane of b and
c and can be expressed in terms of them.
It can be shown:
a × (b × c) = (a.c) b − (a.b) c
(partial proof, see over ...).
7
Partial proof (x-component only):
i j k
(b × c) = bx by bz
(4)
cx cy cz
i j k
a × (b × c) = ax ay az
(5)
by cz − bz cy −(bx cz − bz cx ) bx cy − by cx
I x-component only
i: ay (bx cy − by cx ) + az (bx cz − bz cx )
= (ay cy + az cz ).bx − (ay by + az bz ).cx +
+ ((ax cx )bx − (ax bx )cx ) ← [note, add this extra term, sum = 0]
= i ((a.c) bx − (a.b) cx ) Similarly for j and k components.
I Also easy to show:
(a × b) × c = (a.c) b − (b.c) a.
I Can also show from above expressions:
a × (b × c) + b × (c × a) + c × (a × b) = 0
8
3.2.1 Lagrange’s identity
Another useful identity (can be proved using components)
(a × b).(c × d) = (a.c)(b.d) − (a.d)(b.c)
Or alternatively: can be proved using identities of scalar and
vector triple products:
I (a × b) · (c × d) = d · ((a × b) × c) = c · (d × (a × b))
(Using properties of scalar triple product)
I = c · ((d · b) a − (d · a) b)
(Using identity of vector product)
I = (a · c)(b · d) − (a · d)(b · c)
(Rearranging)
9
3.3 Generating orthogonal axes
Orthogonal axes can be constructed from cross product of two
general vectors
Prescription:
I i) Start from general vectors a
and b,
I ii) Choose vector a as the
direction of the x-axis
I iii) The direction of the y-axis is
then given by a × b
I iv) The direction of the z-axis is
then simply given by a × (a × b).
10
