MA 102: Engineering Mathematics II Odd Sem ’19–’20

Solution to the Tutorial Sheet 3: September 12, 2019

Lecturer: D. Ghosh Scribes: Abhishek

Tutorial-3

1. Let S = {(2, −3, 5), (8, −12, 20), (1, 0, −2), (0, 2, −1), (7, 2, 0)}. Show that S spans R3 . Find a subset of S
which is a basis for R3 .

Solution

Let
S = {u1 = (2, −3, 5), u2 = (8, −12, 20), u3 = (1, 0, −2), u4 = (0, 2, −1), u5 = (7, 2, 0)}.
Let(a, b, c) ∈ R3 . Consider the equation

(a, b, c) = a1 u1 + a2 u2 + a3 u3 + a4 u4 + a5 u5 ,

for a1 , a2 , a3 , a4 , a5 ∈ R.
Then,  
 a1
  
2 8 1 0 7  a2 
 a
−3 −12 0 2 2 a3
 = b ,
 
5 20 −2 −1 0 a4  c
a5
which can be represented by Ax = B, where
   
2 8 1 0 7 a
A = −3 −12 0 2 2 and B =  b  .
5 20 −2 −1 0 c

Use elementary column operations to reduce this matrix A:

   
2 0 1 0 0 2 0 1 0 0
C →C −4C1 C →C5 −C4
A −−2−−−2−−−→ −3 0 0 2 2  −−5−−−− −−→ −3 0 0 2 0 ,
5 0 −2 −1 14 5 0 −2 −1 15

 
1 0 0
which has a submatrix  0 2 0 , with nonzero determinant. Therefore, the rank of matrix A = rank of
−2 −1 15
the augmented matrix (A : B) = 3. So, the system Ax = b has a nontrivial solution.
Thus, every (a, b, c) ∈ R3 can be written as a linear combinations of vectors of set S. So, S spans R3 .

For the basis:

For this we need to find an L.I. subset of set S which spans the whole R3 .
Since u2 = 4u1 , so by deletion theorem we can remove u2 . Therefore, check L.I. of next 3 remaining vectors

3-1
3-2 Lecture 3: September 12, 2019

of S.
Let a, b, c ∈ R and consider  
0
au1 + bu3 + cu4 = 0 .

0
Which can be written as

2a + b = 0,
−3a + 2c = 0,
5a − 2b − c = 0.

Solving these equations, we get a = b = c = 0, which shows that u1 , u3 , u4 are L.I. vectors.
So the subset S1 = {u1 = (2, −3, 5), u3 = (1, 0, −2), u4 = (0, 2, −1)} is L.I. and since
number of L.I. vectors in S1 = 3 = Dimension of Space R3 . So this subset S1 forms a basis of R3 .
2. Consider the vector spaces V1 = C (over C), V2 = C2 (over C), V3 = C (over R), V4 = C2 (over R), and
V5 = R (over R). Find a basis for each of these vector spaces, and determine their dimensions.
Solution
• Given vector space V1 = C (over C).
Let S = {1}. Since 1 6= 0, so S is L.I. and for every scaler a ∈ C we have a × 1 = a. Thus every vector of
vector space C can be generated by the set S. So the basis of V1 is S = {1} and therefore the dimension
of V1 is 1.
• Given vector space V2 = C2 (over C) = {(x, y) : x, y ∈ C}.
Let S = {(1, 0), (0, 1)}.
(a) We need to check whether S sis L.I. or not?
Since (1, 0) and (0, 1) are not scalar multiple of each other, so they are L.I. vectors. Therefore this set
S ⊆ V is an L.I. subset of V2 .

(b) We need to show that span(S) = V2 .

Here S ⊆ V2 =⇒ span(S) ⊆ V2 .
Thus, for every (x, y) ∈ V2 , we have (x, y) = x × (1, 0) + y × (0, 1), where x, y ∈ C.
Therefore, (x, y) ∈ span(S) =⇒ V ⊆ span(S). So V2 = span(S).

Thus, S is L.I. and spans the whole space V , so S is a basis of vector space V2 and therefore, the dimension
of V2 is 2.
• Given vector space V3 = C(over R) = {(x + iy) : x, y ∈ R}.
Let S = {1, i}.
(a) Since we have 1 6= a × i, for any a ∈ R. So 1 and i are L.I. vectors in C. Thus S is an L.I. subset of
V3 .
(b) We need to check that span(S) = V3 .
Since S ⊆ V3 , so span(S) ⊆ V3 .
For every z = x + iy ∈ C, we have x × 1 + y × i, where x, y ∈ R.
Here, x + iy ∈ V3 =⇒ x + iy ∈ span(S) =⇒ V ⊆ span(S). Thus V3 = span(S).
Thus S spans the whole vector space V3 . So S is a basis of V3 and therefore, the dimension of vector space
V3 is 2.
Lecture 3: September 12, 2019 3-3

• Given vector space V4 = C2 (R) = {(z1 , z2 ) : z1 , z2 ∈ C} over the field R.

ℂ (ℝ) = {("# , " ): "# , " ∈ ℂ}

&# + '*# & + '*

First place Second place
in the tuple in the tuple
{1, '} {1, '}

{(1,0), (', 0)} {(0,1), (0, ')}

Consider the set S = {(1, 0), (0, 1), (i, 0), (0, i)}.

(a) We need to check whether S is L.I. or not?

Take a, b, c, d ∈ R and consider the linear combination
a × (1, 0) + b × (0, 1) + c × (i, 0) + d × (0, i) = (0, 0),
=⇒ (a + ic, b + id) = (0, 0),
=⇒ a + ic = 0, b + id = 0,
=⇒ a = 0, b = 0, c = 0, d = 0.
{ Since a + ib = 0 = 0 + i × 0 =⇒ a = 0, b = 0. }
So S is an L.I. set.
(b) Since S ⊆ V4 =⇒ span (S) ⊆ V4 .
Here, every vector z1 = a1 + ib1 , z2 = a2 + ib2 ∈ C and (z1 , z2 ) ∈ V4 ,
we have (z1 , z2 ) = a1 (1, 0) + b1 (i, 0) + a2 (0, 1) + b2 (0, i), so V4 ⊆ span(S).
Thus span (S) = V .
Thus S spans the whole vector space V4 . So S is a basis of V4 and dimension of vector space V4 is 4.

3. Suppose {α1 , α2 , α3 } is a basis for a vector space V over R. Let β1 = α1 + α3 , β2 = 2α1 + 3α2 + 4α3 and
β3 = α1 + 2α2 + 3α3 . Prove that {β1 , β2 , β3 } is a basis for V.
Solution:
Given that S = {α1 , α2 , α3 } a basis of vector space V .
Consider the set B = {β1 , β2 , β3 }.
To show B an L.I. set:
Consider the linear combination
aβ1 + bβ2 + cβ3 = 0,
=⇒ a(α1 + α3 ) + b(2α1 + 3α2 + 4α3 ) + c(α1 + 2α2 + 3α3 ) = 0,
=⇒ (a + 2b + c)α1 + (3b + 2c)α2 + (a + 4b + 3c)α3 = 0.
Since α1 , α2 , α3 are L.I. vectors, so
(a + 2b + c) = 0 (3b + 2c) = 0 and (a + 4b + 3c) = 0.
Solving above equations we get a = b = c = 0.
So, B = {β1 , β2 , β3 } is an L.I. set.
Since the number of L.I. vectors in set B = dimension of vector space V = 3. Therefore B is a basis of the
vector space V .

4. Show that W = {(x, y, z) ∈ R3 |x + 2y = z, 2x + 3y = z} is a subspace of R3 . Find a basis for W . What is

the dimension of W ?
3-4 Lecture 3: September 12, 2019

Solution
Observe that W = {(x, y, z) ∈ R3 |x + 2y = z, 2x + 3y = z} is nonempty because (0, 0, 0) ∈ W .
Note that (x1 , y1 , z1 ) ∈ W =⇒ x1 + 2y1 = z1 , 2x1 + 3y1 = z1
and (x2 , y2 , z2 ) ∈ W =⇒ x2 + 2y2 = z2 , 2x2 + 3y2 = z2 .

To show that W is a subspace of V:

(a) To show that (x1 , y1 , z1 ) + (x2 , y2 , z2 ) ∈ W .

We have (x1 , y1 , z1 ) + (x2 , y2 , z2 ) = (x1 + x2 , y1 + y2 , z1 + z2 ).
Consider (x1 + x2 ) + 2(y1 + y2 ) = x1 + 2y1 + x2 + 2y2 = z1 + z2 ,
and 2(x1 + x2 ) + 3(y1 + y2 ) = 2x1 + 3y1 + 2x2 + 3y2 = z1 + z2 .
Thus (x1 , y1 , z1 ) + (x2 , y2 , z2 ) ∈ W .
(b) To show that a(x1 , y1 , z1 ) ∈ W .
For scalar a ∈ R, we have a(x1 , y1 , z1 ) = (ax1 , ay1 , az1 ).
Consider ax1 + 2ay1 = a(x1 + 2y1 ) = az1 and 2ax1 + 3ay1 = a(2x1 + 3y1 ) = az1 .
Thus a(x1 , y1 , z1 ) ∈ W.

Therefore, W is a subspace of R3 .
To find a basis of subspace W:
For every vector (x, y, z) ∈ W, we have x + 2y = z, 2x + 3y = z
=⇒ x + 2y = z = 2x + 3y
=⇒ x + y = 0, z = 2x + 3y
=⇒ y = −x and z = x − 2x = −x.
Therefore, W = {x(1, −1, −1) : x ∈ R}. Here, the vector subspace W is spanned by vector (1, −1, −1).
Therefore a basis of W is {(1, −1, −1)} and the dimension of subspace W is 1.

5. Show that W = {(x, y, z, w) ∈ R4 |x + 2y − z = 0, 2x + y + w = 0} is a subspace of R4 . Find a basis for W .

What is the dimension of W ?
Solution:
To show that W is a subspace of R4
Consider vectors (x1 , y1 , z1 , w1 ), (x2 , y2 , z2 , w2 ) ∈ W.
Then x1 + 2y1 − z1 = 0, 2x1 + y1 + w1 = 0
and x2 + 2y2 − z2 = 0, 2x2 + y2 + w2 = 0.

(a) We will show (x1 , y1 , z1 , w1 ) + (x2 , y2 , z2 , w2 ) ∈ W .

Therefore, we have (x1 , y1 , z1 , w1 ) + (x2 , y2 , z2 , w2 ) = (x1 + x2 , y1 + y2 , z1 + z2 , w1 + w2 ).
Since

x + 2y − z = (x1 + x2 ) + 2(y1 + y2 ) − (z1 + z2 )

= (x1 + 2y1 − z1 ) + (x2 + 2y2 − z2 ) = 0 + 0 = 0

and

2x + y + w = 2(x1 + x2 ) + (y1 + y2 ) + (w1 + w2 )

= (2x1 + y1 + w1 ) + (2x2 + y2 + w2 ) = 0 + 0 = 0,

(x1 , y1 , z1 , w1 ) + (x2 , y2 , z2 , w2 ) ∈ W .
(b) We will show that for a ∈ R, a(x1 , y1 , z1 , w1 ) ∈ W .
Lecture 3: September 12, 2019 3-5

We have a(x1 , y1 , z1 , w1 ) = (ax1 , ay1 , az1 , aw1 ).

Since

x + 2y − z = ax1 + 2ay1 − az1 = a(x1 + 2y1 − z1 ) = 0,

2x + y + w = 2ax1 + ay1 + aw1 = a(2x1 + y1 + w1 ) = 0,

a(x1 , y1 , z1 , w1 ) ∈ W for every scalar a ∈ R.

Thus, W is a vector subspace of R4 .

To find a basis of subspace W :
For any vector (a, b, c, d) ∈ W , we have a + 2b − c = 0 and 2a + b + d = 0 =⇒ c = a + 2b, d = −2a − b.
Therefore, we have

(a, b, c, d) = (a, b, a + 2b, −2a, −b) = a(1, 0, 1, −2) + b(0, 1, 2, −1).

Thus, W = {a(1, 0, 1, −2) + b(0, 1, 2, −1) : a, b ∈ R}. Note that W is spanned by the vectors (1, 0, 1, −2) and
(0, 1, 2, −1) and these two vectors are L.I. because they are not scalar multiple of each other.
Therefore, a basis of W is B = {(1, 0, 1, −2), (0, 1, 2, −1)} and the dimension of W is 2.

6. Find a basis for R3 which contains the vector (3, 2, 6).

Solution:
Replacement Theorem:
If {u1 , u2 , · · · , un } is a basis of a vector space V over a field F and a non-zero vector β of V is expressed as

β = c1 u1 + c2 u2 + · · · + cn un , ci ∈ F,

then, if cj 6= 0, =⇒ {u1 , u2 , · · · , uj−1 , β, uj+1 , · · · , un } is a basis of V .

The standard basis of R3 is B = {(1, 0, 0), (0, 1, 0), (0, 0, 1)}.

So the vector (3, 2, 6) can be written as a linear combination of vectors of basis B.
Thus, (3, 2, 6) = 3 × (1, 0, 0) + 2 × (0, 1, 0) + 6 × (0, 0, 1).
Since the coefficient of vector (1, 0, 0) is non-zero, so (1, 0, 0) can be replaced by the vector (3, 2, 6)
and therefore we have a new basis B1 = {(3, 2, 6), (0, 1, 0), (0, 0, 1)}.

or

Let (3, 2, 6) ∈ R3 .
Therefore, other L.I. vectors in R3 can be find as follows.
Choose (a, b, c) that is not a nonzero scalar multiple of (3, 2, 6). Take (3, 2, 0). Then, the nonzero vectors
(3, 2, 6) and (3, 2, 0) are not scalar multiple of each other. So, these two vectors are L.I.
Find the 3rd vector which is L.I. to the span of these two L.I. vectors.
Next, the spanning set S = {a(3, 2, 6) + b(3, 2, 0) : a, b ∈ R}, i.e.,

S = {(3a + 3b, 2a + 2b, 6a) : a, b ∈ R}.

If (x, y, z) = (3a + 3b, 2a + 2b, 6a) =⇒ x = 3a + 3b, y = 2a + 2b, z = 6a.

Take x = 0 =⇒ b = −a =⇒ y = 0. For (x, y, z) to be out of span(S), take x = 0 and y 6= 0.
Take (x, y, z) = (0, 1, 1).
Finally, we will have an L.I. set S = {(3, 2, 6), (3, 2, 0), (0, 1, 1)}.
Since it is an L.I. set and has 3-L.I. vectors, so S forms a basis of R3 (because dim(R3 ) = 3.)
3-6 Lecture 3: September 12, 2019

7. Find a basis for R4 which contains the vector (1, 1, 1, 0) .

Solution:
The standard basis of R4 is B = {(1, 0, 0, 0), (0, 1, 0, 0), (0, 0, 1, 0), (0, 0, 0, 1)}.
So, the vector (1, 1, 1, 0) can be written as a linear combination of vectors of basis B.
Note that (1, 1, 1, 0) = 1 × (1, 0, 0, 0) + 1 × (0, 1, 0, 0) + 1 × (0, 0, 1, 0) + 1 × (0, 0, 0, 1).
Since the coefficient of vector (1, 0, 0, 0) is non-zero, (1, 0, 0, 0) can be replaced by the vector (1, 1, 1, 0)
and therefore, we have a new basis B1 = {(1, 1, 1, 0), (0, 1, 0, 0), (0, 0, 1, 0), (0, 0, 0, 1)}.

8. Let S = {(1, 2, 1), (2, 1, 1)}. Show that S is linearly independent. Extend S to a basis for R3 .
Solution:

Note: Two non-zero vectors are L.D. if and only if one is scalar multiple of other vector.

Here, (1, 2, 1) = a × (2, 1, 1) = (2a, a, a) =⇒ 2a = 1, a = 2, a = 1

which is absurd. So the above equation have no solution.
Therefore, the vectors (1, 2, 1) and (2, 1, 1) are not L.D. =⇒ these vectors are L.I..
Consider the spanning set

L(S) = {a × (1, 2, 1) + b × (2, 1, 1) : a, b ∈ R},

= {(a + 2b, 2a + b, a + b) : a, b ∈ R}.

Choose a vector from space V such that it belongs to the set V \ L(S).
Any vector (x, y, z) in L(S) is of form

x = a + 2b, y = 2a + b, z = a + b.
Take z = 0 =⇒ b = −a =⇒ y = a =⇒ x = −a, =⇒ (x, y, z) = (−a, a, 0).
Therefore, for (x, y, z) ∈
/ L(S), take a vector that does not have these conditions.
Take (x, y, z) = (1, 0, 0). Therefore, the extended set S1 = {(1, 2, 1), (2, 1, 1), (1, 0, 0)} is an L.I. set with 3
vectors and it forms a basis of space R3 .

Or

We have set S = {(1, 2, 1), (2, 1, 1)}.

We will show that S is an L.I. set.
Let a(1, 2, 1) + b(2, 1, 1) = (0, 0, 0) =⇒ a + 2b = 0, 2a + b = 0, a + b = 0 =⇒ b = −a, and a = 0 = b.
Therefore, these two vectors are L.I. vectors.
Next, we will extend this set to form a basis of R3 .
Choose arbitrary non-zero vector from R3 . Here, we take (1, 0, 0).
Consider a(1, 2, 1) + b(2, 1, 1) + c(1, 0, 0) = (0, 0, 0),
=⇒ a + 2b + c = 0, 2a + b = 0, a + b = 0,
=⇒ b = −a, c = a, a = 0,
=⇒ a = b = c = 0.
So the new set with this vector S1 = {(1, 2, 1), (2, 1, 1), (1, 0, 0)} is an L.I. set with 3 vectors. Therefore, S1
forms a basis of R3 since dim(R3 ) = 3.

Or

We have set S = {(1, 2, 1), (2, 1, 1)}.

We will show that S is L.I.:
Let a(1, 2, 1) + b(2, 1, 1) = (0, 0, 0) =⇒ a + 2b = 0, 2a + b = 0, a + b = 0 =⇒ b = −a, and a = 0 = b.
Lecture 3: September 12, 2019 3-7

Thus, S is an L.I. set. Therefore, it can be extended to form a basis of R3 .

We will extend S to form a basis of R3 :
Choose arbitrary any nonzero vector (x, y, z) ∈ R3 .
To check it’s L.I. with set S.
Consider the linear combination a(1, 2, 1) + b(2, 1, 1) + c(x, y, z) = (0, 0, 0).
Which gives
a + 2b + cx = 0,
2a + b + cy = 0,
a + b + cz = 0.

The above system of equations can be written as AX = B, i.e.

    
1 2 x a 0
2 1 y   b  = 0 .
1 1 z c 0

Note: The system of equations AX = 0 will have unique solution X = 0, if and only if rank of matrix(A)=
number of columns of matrix A (provided number of rows ≥ number of columns)
For (x, y, z) to be L.I. with set S, this system AX = B must have unique(trivial) solution a = b = c = 0 and
rank of matrix should be 3. Therefore, the determinant of matrix A 6= 0.
Consider
   
1 2 x 1 2 x
R2 →R2 −2×R1 and R3 →R3 −R1
A = 2 1 y  −−−−−−−−−−−−−−−−−−−−−→ 0 −3 y − 2x .
1 1 z 0 −1 z − x
For this −3(z − x) + (y − 2x) 6= 0 =⇒ x + y − 3z 6= 0. Therefore, take x = 1, y = 1, z = 0.
(Here, infinite number of such vector (x, y, z) will exist and we can take take any combination that satisfies
x + y − 3z 6= 0).
Therefore, the new set S1 = {(1, 2, 1), (2, 1, 1), (1, 1, 0)} is an L.I. set with 3 vectors and forms a basis of R3 .
9. Find the coordinate matrix of the vector α = (5, 2, 3) with respect to the (ordered) basis B = {(2, 1, 1), (1, 2, 1), (1, 1, 2)}
of R3 .

Solution:

Note: Write (5, 2, 3) as a linear combinations of vectors of basis B.

Consider (5, 2, 3) = a(2, 1, 1) + b(1, 2, 1) + c(1, 1, 2).

Which gives
2a + b + c = 5,
a + 2b + c = 2,
a + b + 2c = 3.
Solving these equations, we get a = 5/2, b = −1/2, c = 1/2.
Therefore, the coordinate of the vector (5, 2, 3) is (5/2, −1/2, 1/2) and the coordinate matrix of vector (5, 2, 6)
is  
5/2
X = −1/2 .
1/2