1

Exercise: 7 (Linearly dependent, Basis and dimension)

1. Show that the sets of vectors are linearly dependent in ℝ3 .
(i) {(2, 3,1), (2, 1,3), (1 , 1, 1)}
(ii) {(0, −1, 3), (3,4,3), (1, 1, 2)}.
Answer: (i) Consider the equation 𝑐1 (2, 3,1) + 𝑐2 (2, 1,3) + 𝑐3 (1 , 1, 1) = 𝜃 = (0,0,0)
2𝑐1 + 2𝑐2 + 𝑐3 = 0
⇒ 3𝑐1 + 𝑐2 + 𝑐3 = 0 … … . . … . (1)
𝑐1 + 3𝑐2 + 𝑐3 = 0
2 2 1
Now |3 1 1| = 2 × (1 − 3) − 2(3 − 1) + 1(9 − 1) = −4 − 4 + 8 = 0.
1 3 1
So the system of homogeneous linear equation (1) has non-zero solution.
Therefore {(2, 3,1), (2, 1,3), (1 , 1, 1)} is linearly dependent.
0 3 1
(ii) [Shortcut] Since |−1 4 1| = 0 − 3(−2 − 3) + 1(−3 − 12) = 15 − 15 = 0.
3 3 2
So {(0, −1, 3), (3,4,3), (1, 1, 2)} is linearly dependent.

2. Show that the sets of vectors are linearly dependent in ℝ4 .

(i) {(1, 1, 1,0), (1,0,1,1), (1, 2,1,2), (1, 1, 1, 1)},
(ii) {(2, 3,1,4), (3,2,4,1), (1, 1, 1, 1)}.

1 1 1 1
1 0 2 1
Answer: (i) [Shortcut] | | = 0 (since first and third rows are identical).
1 1 1 1
0 1 2 1
So {(1, 1, 1,0), (1,0,1,1), (1, 2,1,2), (1, 1, 1, 1)} is linearly dependent.

(ii) [Here we cannot use the previous shortcut technique. This is a general method]
Consider the equation
𝑐1 (2, 3,1,4) + 𝑐2 (3,2,4,1) + 𝑐3 (1,1 , 1, 1) = 𝜃 = (0,0,0,0)
2𝑐1 + 3𝑐2 + 𝑐3 = 0 2 3 1 0
𝑐1
3𝑐 + 2𝑐2 + 𝑐3 = 0 3 2 1 0
⇒ 1 or ( ) (𝑐2 ) = ( ) … … (1)
𝑐1 + 4𝑐2 + 𝑐3 = 0 1 4 1 𝑐3 0
4𝑐1 + 𝑐2 + 𝑐3 = 0 4 1 1 0
 2

𝑅1′ =𝑅1 +𝑅2 5 5 2 𝑐1 0 5 5 2 𝑐1 0

′ =𝑅 −𝑅
𝑅4′ =𝑅4 +𝑅3 3 2 1 0 𝑅4 4 1 3 2 1 0
→ ( ) (𝑐2 ) = ( ) → ( ) (𝑐2 ) = ( ).
1 4 1 𝑐3 0 1 4 1 𝑐3 0
5 5 2 0 0 0 0 0
5 5 2
Now |3 2 1| = 5(2 − 4) − 5(3 − 1) + 2(12 − 2) = −10 − 10 + 20 = 0.
1 4 1
So the system (1) has non-zero solution and thus {(2, 3,1,4), (3,2,4,1), (1, 1, 1, 1)} is L.D.

Another method: This problem can also be solve by simply inspection:

Let 𝛼 = (2, 3,1,4), 𝛽 = (3,2,4,1) & 𝛾 = (1, 1, 1, 1)}.
Then 𝛼 + 𝛽 = (5,5,5,5) = 5𝛾. So obviously {𝛼, 𝛽, 𝛾} is L.D.

3. Determine 𝑘 so that the set 𝑆 is linearly dependent in ℝ3 , where

(i) 𝑆 = {(1, 2,1), (𝑘, 3,1), (2, 𝑘, 0)},
(ii) 𝑆 = {(𝑘, 1, 1), (1, 𝑘, 1), (1, 1, 𝑘)}.

1 𝑘 2 1 1 0
Answer: (i) |2 3 𝑘 | = |1 𝑘 2| = (𝑘 2 − 6) − (𝑘 − 4) = 𝑘 2 − 𝑘 − 2 = (𝑘 − 2)(𝑘 + 1).
1 1 0 2 3 𝑘
So {(1, 2,1), (𝑘, 3,1), (2, 𝑘, 0)} is L.D. if (𝑘 − 2)(𝑘 + 1) = 0, i.e., 𝑘 = 2, −1.

𝑘 1 1
(ii) |1 𝑘 1| = 𝑘 (𝑘 2 − 1) − (𝑘 − 1) + (1 − 𝑘 ) = 𝑘 3 − 3𝑘 + 2 = (𝑘 + 2)(𝑘 − 1)2
1 1 𝑘
So {(𝑘, 1, 1), (1, 𝑘, 1), (1, 1, 𝑘)} is L.D. if (𝑘 + 2)(𝑘 − 1)2 = 0, i.e., 𝑘 = −2,1.

4. Find the conditions on x, y so that the set of vectors is linearly dependent in ℝ3 .

(i) {(𝑥, 𝑦, 𝑦), (𝑦, 𝑥, 𝑦), (𝑦, 𝑦, 𝑥)}
(ii) {(𝑥, 𝑦, 1), (𝑦, 1, 𝑥), (1, 𝑥, 𝑦)}.
𝑥 𝑦 𝑦
Answer: (i) |𝑦 𝑥 𝑦| = 𝑥 3 − 3𝑥𝑦 2 + 2𝑦 3 = (𝑥 − 𝑦)2 (𝑥 + 2𝑦)
𝑦 𝑦 𝑥
So {(𝑥, 𝑦, 𝑦), (𝑦, 𝑥, 𝑦), (𝑦, 𝑦, 𝑥)} is L.D. if (𝑥 − 𝑦)2 (𝑥 + 2𝑦) = 0, i.e., 𝑥 = 𝑦 or 𝑥 = −2𝑦.
 3

𝑥 𝑦 1
(ii) |𝑦 1 𝑥 | = −1 − 𝑥 3 + 3𝑥𝑦 − 𝑦 3 = −(1 + 𝑥 + 𝑦)(1 − 𝑥 + 𝑥 2 − 𝑦 − 𝑥𝑦 + 𝑦 2 )
1 𝑥 𝑦
1
= − (1 + 𝑥 + 𝑦){(𝑥 − 1)2 + (𝑦 − 1)2 + (𝑥 − 𝑦)2 }.
2
Therefore {(𝑥, 𝑦, 1), (𝑦, 1, 𝑥), (1, 𝑥, 𝑦)} is L.D. if (1 + 𝑥 + 𝑦) = 0 or 𝑥 = 𝑦 = 1.

5. Show that set of vectors 𝑆 = {(1,2,0), (2,1,3), {1,1,1), (2,3,1)} is linearly dependent in ℝ3 .
Find a linearly independent subset 𝑆1 of 𝑆 such that 𝐿(𝑆1 ) = 𝐿(𝑆).
Answer: [General procedure] Consider the equation
𝑐1 (1, 2, 0) + 𝑐2 (2,1,3) + 𝑐3 (1, 1, 1) + 𝑐4 (2,3,1) = 𝜃 = (0,0,0)
𝑐1
1 2 1 2 𝑐2 0 Our target is to find out a
⇒ (2 1 1 3) (𝑐 ) = (0) … … … (1) non-zero solution of (1).
3
0 3 1 1 𝑐4 0

𝑅1′ =𝑅1 −𝑅3 𝑐1 Here two equations

𝑅2′ =𝑅2 −𝑅3
1 −1 0 1 𝑐2 0 𝑐1 − 𝑐2 + 𝑐4 = 0
with four unknowns, so
→ (2 −2 0 2) (𝑐 ) = (0) i. e., . we can chose two
3 3𝑐2 + 𝑐3 + 𝑐4 = 0
0 3 1 1 𝑐4 0 unknowns as arbitrary.

So we have a particular solution (non-zero) of (1) as 𝑐4 = 0, 𝑐1 = 1 and then 𝑐2 = 1, 𝑐3 = −3.

Thus (1, 2, 0) + (2,1,3) − 3(1, 1, 1) = 𝜃 and we can say 𝑆 is L.D. ------- (*)
Again according to deletion theorem, we have 𝐿(𝑆1 ) = 𝐿(𝑆), where 𝑆1 = 𝑆 − {(1,1,1)}
[∵ in the dependence relation (*) the coefficient of (1,1,1) is −3 ≠ 0]

Another method [Shortcut method by inspection]: By inspection we observed that

(1, 2, 0) + (1, 1, 1) = (2,3,1)
So we can conclude that 𝑆 = {(1,2,0), (2,1,3), {1,1,1), (2,3,1)} is L.D. and by deletion theorem
{(1,2,0), (2,1,3), {1,1,1)} ⊂ 𝑆 spans 𝐿(𝑆).
≠

6. Show that the set of vectors 𝑆 = {(1, 2, 3, 0), (2, 1,0,3), (1, 1, 1, 1), (2, 3,4, 1)} is linearly
dependent in ℝ4 . Find a linearly independent subset 𝑆1 of 𝑆 such that 𝐿(𝑆1 ) = 𝐿(𝑆).
Answer: [General procedure] Consider the equation
𝑐1 (1, 2,3, 0) + 𝑐2 (2,1,0,3) + 𝑐3 (1, 1, 1,1) + 𝑐4 (2,3,4,1) = 𝜃 = (0,0,0,0) Our target is to find out a
non-zero solution of (1).
 4

1 2 1 2 𝑐1 0
2 1 1 3 𝑐2 0
⇒( ) (𝑐 ) = ( ) … … … (1)
3 0 1 4 3 0
0 3 1 1 𝑐4 0
𝑅1′ =𝑅1 +𝑅2 3 3 2 5 𝑐1 0 𝑅1′ =𝑅1 −𝑅3 0 0 0 0 𝑐1 0
𝑅3′ =𝑅3 +𝑅4 2 1 1 3 𝑐2 0 𝑅2′ =𝑅2 −𝑅4 2 −2 0 2 𝑐2 0
→ ( ) (𝑐 ) = ( ) → ( ) (𝑐 ) = ( )
3 3 2 5 3 0 3 3 2 5 3 0
0 3 1 1 𝑐4 0 0 3 1 1 𝑐4 0
2𝑐1 − 2𝑐2 + 2𝑐4 = 0
i. e., 3𝑐1 + 3𝑐2 + 2𝑐3 + 5𝑐4 = 0.
3𝑐2 + 2𝑐3 + 5𝑐4 = 0
So we have a particular solution (non-zero) of (1) as
𝑐1 = 0, then 𝑐2 = 𝑐4 and 3𝑐2 + 2𝑐3 + 5𝑐4 = 8𝑐2 + 2𝑐3 = 0.
So moreover if 𝑐3 = −4, then 𝑐2 = 1 = 𝑐4 .

Thus (1, 2,3, 0) + (2,1,0,3) − 4(1, 1, 1,1) = 𝜃 and we can say 𝑆 is L.D.
Again according to deletion theorem, we have 𝐿(𝑆1 ) = 𝐿(𝑆), where 𝑆1 = 𝑆 − {(1,1,1,1)}.
[∵ in the dependence relation the coefficient of (1,1,1,1) is −4 ≠ 0]

Another method [Shortcut method by inspection]: By inspection we observed that

(1, 2,3, 0) + (2,1,0,3) = 3(1,1,1,1).
So we can conclude that 𝑆 = {(1,2,0), (2,1,3), {1,1,1), (2,3,1)} is L.D. and by deletion theorem
{(1,2,3,0), (2,1,0,3), {2,3,4,1)} ⊂ 𝑆 spans 𝐿(𝑆).
≠

7. Show that the sets of vectors are linearly independent in ℝ3 .

(i) {(1,2,3), (2,3,1), (3,1,2)}, (ii) {(0, 1, 1), (1,0,1), (1, 1, 0)}.
1 2 3
Answer: (i) [Shortcut] |2 3 1| = −18 ≠ 0.
3 1 2
So the set of vectors {(1,2,3), (2,3,1), (3,1,2)} is L.I.
0 1 1
(ii) |1 0 1| = 2 ≠ 0.
1 1 0
So the set of vectors {(0, 1, 1), (1,0,1), (1, 1, 0)} is L.I.
 5

8. Show that the sets of vectors are linearly independent in ℝ4 .

(i) {(1,2, 3,0), (2,3,0,1) , (3,0,1,2)},
(ii) {(1,1,1,0), (1,1,0,1), (1,0,1,1), (0,1,1,1)}.
1 1 2 3
2 3 0
0 2 3 0
Answer: (i) [Shortcut] | | = 1 |3 0 1| = −20 ≠ 0.
0 3 0 1
0 1 2
0 0 1 2
So the set of vectors {(1,0,0,0), (1,2, 3,0), (2,3,0,1) , (3,0,1,2)} is L.I.
Hence its subset {(1,2, 3,0), (2,3,0,1) , (3,0,1,2)} is L.I.

1 1 1 0
1 1 0 1
(ii) | | = −3 ≠ 0.
1 0 1 1
0 1 1 1
So the set of vectors {(1,1,1,0), (1,1,0,1), (1,0,1,1), (0,1,1,1)} is L.I.

9. For what real values of 𝑘 does the set 𝑆 form a basis of ℝ3 ?

(i) 𝑆 = {(𝑘, 1, 𝑘), (0, 𝑘, 1), (1,1,1)},
(ii) 𝑆 = {(𝑘, 0,1), (1, 𝑘 + 1,1), (1, 1,1)}.

Answer: (i) We know 𝑆 is a basis if and only if (a) 𝑆 is L.I. and (ii) 𝐿(𝑆) = ℝ3 .
We know dimension of ℝ3 is 3, so 𝑆 is a basis if and only if 𝑆 is L.I.
𝑘 0 1
Now |1 𝑘 1| = 1 − 𝑘. So 𝑆 is a basis iff 𝑘 − 1 ≠ 0 or 𝑘 ≠ 1.
𝑘 1 1
(ii) We know 𝑆 is a basis if and only if (a) 𝑆 is L.I. and (ii) 𝐿(𝑆) = ℝ3 .
We know dimension of ℝ3 is 3, so 𝑆 is a basis if and only if 𝑆 is L.I.
𝑘 1 1
Now |0 𝑘 + 1 1| = 𝑘 2 − 𝑘. So 𝑆 is a basis iff 𝑘 2 − 𝑘 ≠ 0 or 𝑘 ≠ 0,1.
1 1 1

10. Let {𝛼, 𝛽, 𝛾} be a basis of a real vector space 𝑉 and 𝑐 be a non-zero real number. Prove that
(i) {𝑐𝛼, 𝑐𝛽, 𝑐𝛾} is a basis of 𝑉,
(ii) {𝛼 + 𝑐𝛽, 𝛽, 𝛾} is a basis of 𝑉,
(iii) {𝛼 + 𝑐𝛽, 𝛽 + 𝑐𝛾, 𝛾 + 𝑐𝛼} may not be a basis of 𝑉.
 6

Answer: (i) [General/simple technique] Consider the equation 𝑐1 (𝑐𝛼 ) + 𝑐2 (𝑐𝛽) + 𝑐3 (𝑐𝛾) = 𝜃
Then 𝑐𝑐1 = 𝑐𝑐2 = 𝑐𝑐3 = 0 [as {𝛼, 𝛽, 𝛾} is a basis and hence L.I.]
⇒ 𝑐1 = 𝑐2 = 𝑐3 = 0 [as 𝑐 ≠ 0]
Therefore {𝑐𝛼, 𝑐𝛽, 𝑐𝛾} is L.I.
Again as dim(𝑉 ) = 𝑜({𝛼, 𝛽, 𝛾}) = 3, so {𝑐𝛼, 𝑐𝛽, 𝑐𝛾} is a basis of 𝑉.

[Replacement theory] Given that {𝛼, 𝛽, 𝛾} is a basis of 𝑉.

Now 𝑐𝛼 = 𝑐(𝛼) + 0 𝛽 + 0 𝛾. Then by replacement theory, we can replace 𝛼 in {𝛼, 𝛽, 𝛾} by 𝑐𝛼.
So {𝑐𝛼, 𝛽, 𝛾} is a basis of 𝑉.
Again 𝑐𝛽 = 0(𝑐𝛼) + 𝑐(𝛽) + 0 𝛾. So we can replace 𝛽 in {𝑐𝛼, 𝛽, 𝛾} by 𝑐𝛽.
So {𝑐𝛼, 𝑐𝛽, 𝛾} is a basis.
Again 𝑐𝛾 = 0(𝑐𝛼) + 0(𝑐𝛽) + 𝑐(𝛾). So we can replace 𝛾 in {𝑐𝛼, 𝑐𝛽, 𝛾} by 𝑐𝛾.
Thus {𝑐𝛼, 𝑐𝛽, 𝑐𝛾} is a basis of 𝑉.

[Change of basis technique] Given that {𝛼, 𝛽, 𝛾} is a basis of 𝑉.

𝑐𝛼 𝑐 0 0 𝛼 𝑐 0 0
Now (𝑐𝛽 ) = (0 𝑐 0 ) ( 𝛽 ). That is the change of basis matrix is ℬ = ( 0 𝑐 0).
𝑐𝛾 0 0 𝑐 𝛾 0 0 𝑐
Since det(ℬ ) = 𝑐 3 ≠ 0 [as 𝑐 ≠ 0], so {𝑐𝛼, 𝑐𝛽, 𝑐𝛾} is a basis of 𝑉.

(ii) [General/simple technique] Left for the reader.

[Replacement theory (here this is more effective)] Given that {𝛼, 𝛽, 𝛾} is a basis of 𝑉.
Now 𝛼 + 𝑐𝛽 = 1(𝛼) + 𝑐(𝛽) + 0 𝛾.
As coefficient of 𝛼 is 1, i.e. non-zero, so by replacement theorem, we can replace 𝛼 in
{𝛼, 𝛽, 𝛾} by 𝛼 + 𝑐𝛽. So {𝛼 + 𝑐𝛽, 𝛽, 𝛾} is a basis of 𝑉.
[Change of basis technique] Given that {𝛼, 𝛽, 𝛾} is a basis of 𝑉.
𝛼 + 𝑐𝛽 1 𝑐 0 𝛼 1 𝑐 0
Now ( 𝛽 ) = (0 𝑐 0) (𝛽). That is the change of basis matrix is ℬ = (0 𝑐 0).
𝛾 0 0 𝑐 𝛾 0 0 𝑐
Since det(ℬ ) = 𝑐 3 ≠ 0 [as 𝑐 ≠ 0], so {𝛼 + 𝑐𝛽, 𝛽, 𝛾} is a basis of 𝑉.

Note: The “Change of basis technique” may be not well-known. [Beware from it]
 7

(iii) [Change of basis technique (I take it first as it gives me to predict the answer)]
Given that {𝛼, 𝛽, 𝛾} is a basis of 𝑉.
𝛼 + 𝑐𝛽 1 𝑐 0 𝛼 1 𝑐 0
Then ( 𝛽 + 𝑐𝛾 ) = (0 1 𝑐 ) (𝛽). That is the change of basis matrix is ℬ = (0 1 𝑐 ).
𝛾 + 𝑐𝛼 𝑐 0 1 𝛾 𝑐 0 1
Now det(ℬ ) = 1 + 𝑐 3 . So det(ℬ) = 0 iff 1 + 𝑐 3 = 0, i.e., 𝑐 = −1
So {𝛼 + 𝑐𝛽, 𝛽 + 𝑐𝛾, 𝛾 + 𝑐𝛼} is a basis of 𝑉 except 𝑐 = −1.

[General/simple technique] Consider 𝑐1 (𝛼 + 𝑐𝛽) + 𝑐2 (𝛽 + 𝑐𝛾) + 𝑐3 (𝛾 + 𝑐𝛼 ) = 𝜃

⇒ (𝑐1 + 𝑐 𝑐3 )𝛼 + (𝑐2 + 𝑐 𝑐1 )𝛽 + (𝑐3 + 𝑐 𝑐2 )𝛾 = 𝜃
Then 𝑐1 + 𝑐 𝑐3 = 𝑐2 + 𝑐 𝑐1 = 𝑐3 + 𝑐 𝑐2 = 0 -----(*) [as {𝛼, 𝛽, 𝛾} is a basis and hence L.I.]
1 0 𝑐 𝑐1 0
⇒ (𝑐 1 0) (𝑐2 ) = (0)
0 𝑐 1 𝑐3 0
1 0 𝑐
Therefore (*) has non-zero solution if | 𝑐 1 0| = 0, i.e., 1 + 𝑐 3 = 0, i.e., 𝑐 = −1.
0 𝑐 1
Hence (*) gives only zero solution, i.e., 𝑐1 = 𝑐2 = 𝑐3 = 0 if 𝑐 ≠ 1. this is our need for L.I.

So we can say that {𝛼 + 𝑐𝛽, 𝛽 + 𝑐𝛾, 𝛾 + 𝑐𝛼} is L.I. and hence a basis iff 𝑐 ≠ 1.

[Replacement theory (here this process complicated)] Given that {𝛼, 𝛽, 𝛾} is a basis of 𝑉.
Now 𝛼 + 𝑐𝛽 = 1(𝛼) + 𝑐(𝛽) + 0 𝛾.
As coefficient of 𝛼 is 1, i.e. non-zero, so by replacement theorem, we can replace 𝛼 in
{𝛼, 𝛽, 𝛾} by 𝛼 + 𝑐𝛽. So {𝛼 + 𝑐𝛽, 𝛽, 𝛾} is a basis of 𝑉.
Again 𝛽 + 𝑐𝛾 = 0(𝛼 + 𝑐𝛽) + 1(𝛽) + 𝑐(𝛾)
As coefficient of 𝛽 is 1, i.e. non-zero, so by replacement theorem, we can replace 𝛽 in
{𝛼 + 𝑐𝛽, 𝛽, 𝛾} by 𝛽 + 𝑐𝛾. So {𝛼 + 𝑐𝛽, 𝛽 + 𝑐𝛾, 𝛾} is a basis of 𝑉.
chose coefficient step by step
Now 𝛾 + 𝑐𝛼 = 𝑐 (𝛼 + 𝑐𝛽) − 𝑐 2 (𝛽 + 𝑐𝛾) + (𝑐 3 + 1) 𝛾 (something complicated)
So to replace 𝛾 from {𝛼 + 𝑐𝛽, 𝛽 + 𝑐𝛾, 𝛾} by 𝛾 + 𝑐𝛼, we have to (𝑐 3 + 1) ≠ 0, i.e., 𝑐 ≠ −1.
Therefore {𝛼 + 𝑐𝛽, 𝛽 + 𝑐𝛾, 𝛾 + 𝑐𝛼} is a basis iff 𝑐 ≠ 1.

[By inspection (only to show not always possible)] Given that {𝛼, 𝛽, 𝛾} is a basis of 𝑉.
Take 𝑐 = −1, and we easily observed that (𝛼 − 𝛽 ) + (𝛽 − 𝛾) + ( 𝛾 − 𝛼 ) = 𝜃.
So we can say that {𝛼 + 𝑐𝛽, 𝛽 + 𝑐𝛾, 𝛾 + 𝑐𝛼} is not always a basis of 𝑉.
 8

1.0

Geometry:
0.5
0.0
0.5
0.0
0.5
1.0 1.0

𝛼 − 𝛽, 𝛽 − 𝛾, 𝛾 − 𝛼
1.5
2.0 1.0

1.0

0.5

𝛼, 𝛽, 𝛾
0.5
0.0

0.5

0.0
0.0 1.0

1.0
0.5
0.5
1.0 0.0

0.5

Observe that 𝛼 − 𝛽, 𝛽 − 𝛾, 𝛾 − 𝛼 lie on a plane. 1.0

11. If {𝛼1 , 𝛼2 , 𝛼3 } be a basis of a real vector space 𝑉 and

𝛽1 = 𝛼1 + 𝛼3 , 𝛽2 = 2𝛼1 + 3𝛼2 + 4𝛼3 , 𝛽3 = 𝛼1 + 2𝛼2 + 3𝛼3
Prove that {𝛽1 , 𝛽2 , 𝛽3 } is also a basis of 𝑉.
Answer: [General/simple technique] Left to the reader.

[Change of basis technique] Given that {𝛼1 , 𝛼2 , 𝛼3 } is a basis of 𝑉. Then

𝛽1 𝛼1 + 𝛼3 1 0 1 𝛼1 1 0 1
(𝛽2 ) = (2𝛼1 + 3𝛼2 + 4𝛼3 ) = (2 3 4) (𝛼2 ). So change-matrix ℬ = (2 3 4).
𝛽3 𝛼1 + 2𝛼2 + 3𝛼3 1 2 3 𝛼3 1 2 3
Now det(ℬ ) = 2 ≠ 0. So {𝛽1 , 𝛽2 , 𝛽3 } is a basis.

12. Prove that the set 𝑆 = {(1, 1, 0), (1,0,1), (0, 1,1)} is a basis of the vector space ℝ3 . Show that
the vector (1, 1, 1) may replace anyone of the vectors of the set 𝑆 to form a new basis for ℝ3 , but
the same is not true for the vector (3, 1, 2).
1 1 0
Answer: Now |1 0 1| = −2, so 𝑆 = {(1, 1, 0), (1,0,1), (0, 1,1)} is L.I. Again since 𝑜(𝑆) = 3,
0 1 1
i.e., dimension of ℝ3 . Therefore 𝑆 is a basis of ℝ3 .
Let us consider (1,1,1) = 𝑐1 (1, 1, 0) + 𝑐2 (1,0,1) + 𝑐3 (0, 1,1)
1
⇒ 𝑐1 + 𝑐2 = 1, 𝑐1 + 𝑐3 = 1, 𝑐2 + 𝑐3 = 1 ⇒ 𝑐1 = 𝑐2 = 𝑐3 =
2
1 1 1
So we can write (1,1,1) = (1, 1, 0) + (1,0,1) + (0, 1,1).
2 2 2

Now since coefficient of each of (1, 1, 0), (1,0,1) and (0, 1,1) are non-zero, so we can replace
any one of them by (1,1,1) in 𝑆 = {(1, 1, 0), (1,0,1), (0, 1,1)}.
 9

Again suppose (3,1,2) = 𝑑1 (1, 1, 0) + 𝑑2 (1,0,1) + 𝑑3 (0, 1,1)

⇒ 𝑑1 + 𝑑2 = 3, 𝑑1 + 𝑑3 = 1, 𝑑2 + 𝑑3 = 2 ⇒ 𝑑1 = 1, 𝑑2 = 2, 𝑑3 = 0
So we can write (3,1,2) = (1, 1, 0) + 2 (1,0,1) + 0 (0, 1,1) ------(*)
Now since coefficient of each of (1, 1, 0) and (1,0,1) are non-zero, so we can replace any one of
them by (3,1,2) in 𝑆 = {(1, 1, 0), (1,0,1), (0, 1,1)}.
But we cannot replace (0,1,1) by (3,1,2) as the coefficient of (0,1,1) in (*) is zero.

13. Find a basis for the vector space ℝ3 , that contains the vectors
(i) (1, 2, 1) and (3, 6, 2), (ii) (1, 0, 1) and (1, 1, 1), (iii) (1,1,0) and (0,1,0)
[Tricks: In this type of extension problem we want to insert the standard basis elements (such as
(1,0), (1,0,0), (0,1,0), (1,0,0,0) etc.) in the given basis. But we have keep in mind that the
corresponding part is nonsingular. e.g.,
1 0 0
To extend {(1,2,3), (1,4,0)} we first chose (1,0,0) (1 2 3)
as the other basis vector, so we have to check 1 4 0 Non-singular
This is first to check by inspection. If we fail for
𝑒1 = (1,0, … ), then we chose 𝑒2 = (0,1,0, … ) to enter in the extended basis.]
1 1 3 1 0 0
2 1
Answer: (i) Let us consider the vector (1,0,0). Now |0 2 6| = |1 2 1| = | | = −2.
6 2
0 1 2 3 6 2
So {(1, 2, 1), (3, 6, 2), (1,0,0)} is an extended basis.
1 1 1 1 0 0
0 1
(ii) Let us consider the vector (1,0,0). Now |0 0 1| = |1 0 1| = | | = −1.
1 1
0 1 1 1 1 1
So {(1, 0, 1), (1,1,1), (1,0,0)} is an extended basis.
1 0
(iii) [Look here that for (1,0,0), corresponding matrix is ( ) is singular and 𝑒2 is already in]
1 0
1 0 0
Chose the vector (0,0,1). Now |1 1 0| = 1 ≠ 0.
0 0 1
So {(1,1,0), (0,1,0), (0,0,1)} is an extended basis of ℝ3 .

14. Find a basis for the vector pace ℝ4 , that contains the vectors
(i) (1, 0, 1, 0) and (0, 1, 0, 1); (ii) (1, 1,0,0) and (1, 1, 1,0).
 10

[Tricks: Here we have to insert 2 vectors. As similar to above we try to insert 𝑒1 , 𝑒2 , 𝑒3 … in basis.
We chose those 𝑒𝑖 ’s such that the remaining position part is non-singular.

(1,0, 1,0) (1,1, 0,0) (1, 1, 0, 0)

(0,1, 0,1) (1,1, 1,0) Singular (1, 1, 1, 0) Non-singular

Non-singular

So here we can chose 𝑒1 and 𝑒2 . Here we can’t chose 𝑒1 and 𝑒2 but we can chose 𝑒1 & 𝑒4

1 0 0 0
0 1 0 0 here we use transpose
Answer: (i) Consider (1,0,0,0) & (0,1,0,0). Then | | = 1.
1 0 1 0 directly for easy
0 1 0 1 calculation.
So {(1,0,0,0), (0,1,0,0), (1, 0, 1, 0), (0, 1, 0, 1)} is an extended basis.
1 0 0 0 Since 1st and 2nd row are
𝑒1 and 𝑒4 so determinant is
0 0 0 1 1 0
(ii) Consider (1,0,0,0) & (0,0,0,1). Then | |=| | = 1. equal to the determinant of
1 1 0 0 1 1 2nd and 3rd column of 3rd
1 1 1 0 and 4th row.
So {(1, 1,0,0), (1, 1, 1,0), (1,0,0,0), (0,0,0,1)} is an extended basis.

15. Extend the set 𝑆 to obtain a basis of the vector space ℝ3 .

(i) 𝑆 = {(0, 2,1), (2, 1, 1)}; (ii) 𝑆 = {(1, 1, 0), (1, 1, 1)}.
1 0 0
Answer: (i) Chose the vector (1,0,0). Now |0 2 1| = 1. (0, 2,1)
2 1 1 (2,1,1)
Non-singular
So {(1,0,0), (0, 2,1), (2, 1, 1)} is an extended basis of ℝ3 .
1 0 0 Here we use (1, 1,0)
(ii) Chose the vector (1,0,0). Now |1 1 0| = 1. the transpose
(1,1,1)
(why!!) Non-singular
1 1 1
So {(1,0,0), (1, 1, 0), (1, 1, 1)} is an extended basis of ℝ3 .

16. Extend the set 𝑆 to obtain a basis of the vector space ℝ4 .

(i) 𝑆 = {(1, 2, 1,2), (0, 1, 0, 1)}, (ii) 𝑆 = {(1, 0, 1, 1) , (1, 1, 1, 1)}
Answer: Left to the reader.
[Hint: (i) {𝑒1 , 𝑒2 } ∪ 𝑆 and (ii) {𝑒1 , 𝑒3 } ∪ 𝑆]
 11

17. Let 𝑉 be a vector space with a basis {𝛼1 , 𝛼2 , ⋯ , 𝛼𝑛 }. Examine if

{𝛼1 + 𝛼2 , 𝛼2 + 𝛼3 , ⋯ , 𝛼𝑛−1 + 𝛼𝑛 , 𝛼𝑛 + 𝛼1 } is also a basis of V.
Answer: Since 𝑆 = {𝛼1 , 𝛼2 , ⋯ , 𝛼𝑛 } is a basis of 𝑉, so dimension of 𝑉 is 𝑛. Thus to show the set
𝑇 = {𝛼1 + 𝛼2 , 𝛼2 + 𝛼3 , ⋯ , 𝛼𝑛−1 + 𝛼𝑛 , 𝛼𝑛 + 𝛼1 } as a basis, it is suffices to show 𝑇 is L.I.
Now consider the equation 𝑐1 (𝛼1 + 𝛼2 ) + 𝑐2 (𝛼2 + 𝛼3 ) + ⋯ + 𝑐𝑛 ( 𝛼𝑛 + 𝛼1 ) = 𝜃
⇒ (𝑐1 + 𝑐𝑛 )𝛼1 + (𝑐1 + 𝑐2 )𝛼2 + (𝑐2 + 𝑐3 )𝛼3 + ⋯ + (𝑐𝑛−1 + 𝑐𝑛 )𝛼𝑛 = 𝜃
⇒ (𝑐1 + 𝑐𝑛 ) = (𝑐1 + 𝑐2 ) = (𝑐2 + 𝑐3 ) = ⋯ = (𝑐𝑛−1 + 𝑐𝑛 ) = 0 [∵ 𝑇 is L. I. ]
Which gives 𝑐1 = 𝑐2 = ⋯ = 𝑐𝑛 = 0. Therefore 𝑇 is L.I. and hence a basis of 𝑉.

18. Find the dimension of the subspace 𝑆 of ℝ3 defined by

(i) 𝑆 = {(𝑥, 𝑦, 𝑧) ∈ ℝ3 : 2𝑥 + 𝑦 − 𝑧 = 0}, (ii) 𝑆 = {(𝑥, 𝑦, 𝑧) ∈ ℝ3 : 𝑥 + 2𝑦 = 𝑧, 2𝑥 + 3𝑦 = 𝑧}.
[I think you know about the technique of “How to check a subset is subspace?”. If in question it
is given that 𝑆 are subspace, we have not to check yet. But in mind we can check easily by just
seeing that the addition conditions are linear homogeneous equation.]
Answer: (i) Let 𝜉 = (𝑎, 𝑏, 𝑐) ∈ 𝑆, then 2𝑎 + 𝑏 − 𝑐 = 0.
So 𝜉 = (𝑎, 𝑏, 𝑐) = (𝑎, 𝑏, 2𝑎 + 𝑏) = 𝑎(1,0,2) + 𝑏(0,1,1) ∈ 𝐿({(1,0,2), (0,1,1)})
Now obviously {(1,0,2), (0,1,1)} is L.I., so {(1,0,2), (0,1,1)} is a basis of 𝑆. Thus dim(𝑆) = 2.
(ii) Let 𝜉 = (𝑎, 𝑏, 𝑐) ∈ 𝑆, then 𝑎 + 2𝑏 = 𝑐 & 2𝑎 + 3𝑏 = 𝑐.
(eliminating 𝑐)Which also gives 𝑎 + 𝑏 = 0 or 𝑏 = −𝑎.
Thus 𝜉 = (𝑎, 𝑏, 𝑐) = (𝑎, 𝑏, 𝑎 + 2𝑏) = (𝑎, −𝑎, 𝑎 − 2𝑎) = 𝑎(1, −1, −1) ∈ 𝐿({(1, −1, −1)}).
Now obviously {(1, −1, −1)} is a basis of 𝑆 and hence dim(𝑆) = 1.

19. Find the dimension of the subspace S of ℝ4 : (i) 𝑆 = {(𝑥, 𝑦, 𝑧, 𝑤) ∈ ℝ4 : 𝑥 + 𝑦 + 𝑧 + 𝑤 = 0},

(ii) 𝑆 = {(𝑥, 𝑦, 𝑧, 𝑤) ∈ ℝ4 : 𝑥 + 2𝑦 − 𝑧 = 2𝑥 + 𝑦 − 𝑤 = 0}.
Answer: (i) Let 𝜉 = (𝑎, 𝑏, 𝑐, 𝑑) ∈ 𝑆, then 𝑎 + 𝑏 + 𝑐 + 𝑑 = 0
Then 𝜉 = (𝑎, 𝑏, 𝑐, 𝑑) = (𝑎, 𝑏, 𝑐, −𝑎 − 𝑏 − 𝑐) = 𝑎(1,0,0, −1) + 𝑏(0,1,0, −1) + 𝑐(0,0,1, −1)
Thus 𝜉 ∈ 𝐿(𝐴), where 𝐴 = {(1,0,0, −1), (0,1,0, −1), (0,0,1, −1)}, which is L.I. So dim(𝑆) = 3.

(ii) Let 𝜉 = (𝑎, 𝑏, 𝑐, 𝑑) ∈ 𝑆, then 𝑎 + 2𝑏 − 𝑐 = 0 & 2𝑎 + 𝑏 − 𝑑 = 0.

So (𝑎, 𝑏, 𝑐, 𝑑) = (𝑎, 𝑏, 𝑎 + 2𝑏, 2𝑎 + 𝑏) = 𝑎(1,0,1,2) + 𝑏(0,1,2,1) ∈ 𝐿({(1,0,1,2), (0,1,2,1)}).
So obviously, {(1,0,1,2), (0,1,2,1)} is L.I. and hence a basis of 𝑆. Thus dim(𝑆) = 2.
 12

(*) Find a basis and the dimension of the following subspace of ℝ3

𝑆 = {(𝑥, 𝑦, 𝑧) ∈ ℝ3 : 𝑥 + 2𝑦 + 3𝑧 = 4𝑥 + 5𝑦 + 6𝑧 = 7𝑥 + 8𝑦 + 9𝑧 = 0}.
[This is a special type of problem because here is too much restriction equation and for this
(A) we cannot arrange 𝜉 easily to get the basis element and
(B) there is a problem of dependency of the equation.]
Answer: (i) Let 𝜉 = (𝑎, 𝑏, 𝑐) ∈ 𝑆, then 𝑎 + 2𝑏 + 3𝑐 = 4𝑎 + 5𝑏 + 6𝑐 = 7𝑎 + 8𝑏 + 9𝑐 = 0.
1 2 3
Then the coefficient matrix of defining conditions is 𝐴 = (4 5 6).
7 8 9
1
𝑅2′ =𝑅2 −4𝑅1 𝑅2′ =− 𝑅2
3
1 2 3 𝑅3′ =𝑅3 −7𝑅1
1 2 3 𝑅3′ =𝑅3 −6𝑅2′
1 2 3
Now (4 5 6) → (0 −3 −6 ) → (0 1 2)
7 8 9 0 −6 −12 0 0 0
Thus (𝑎, 𝑏, 𝑐) satisfies the equations 𝑎 + 2𝑏 + 3𝑐 = 𝑏 + 2𝑐 = 0.
So 𝜉 = (𝑎, 𝑏, 𝑐) = (−2𝑏 − 3𝑐, −2𝑐, 𝑐) = (4𝑐 − 3𝑐, −2𝑐, 𝑐) = 𝑐 (1, −2,1) ∈ 𝐿({(1, −2,1)}).
Obviously, {(1, −2,1)} is L.I. and hence {(1, −2,1)} is a basis of 𝑆 and dim(𝑆) = 1.

[Shortcut] Given 𝑆 = {(𝑥, 𝑦, 𝑧) ∈ ℝ3 : 𝑥 + 2𝑦 + 3𝑧 = 4𝑥 + 5𝑦 + 6𝑧 = 7𝑥 + 8𝑦 + 9𝑧 = 0}.

1 2 3
1 2
Now |4 5 6| = 0 and | | = −1. So rank of 𝐴 is 2. So dimension of 𝑆 is 3 − 2 = 1.
4 5
7 8 9 But to find out basis we have to rearrange 𝜉

20. Find the dimension of the subspace 𝑆 ∩ 𝑇 of ℝ4 where

𝑆 = {(𝑥, 𝑦, 𝑧, 𝑤) ∈ ℝ4 : 𝑥 + 𝑦 + 𝑧 + 𝑤 = 0} and 𝑇 = {(𝑥, 𝑦, 𝑧, 𝑤) ∈ ℝ4 : 2𝑥 + 𝑦 − 𝑧 + 𝑤 = 0}.
Answer: From hypothesis, 𝑆 ∩ 𝑇 = {(𝑥, 𝑦, 𝑧, 𝑤) ∈ ℝ4 : 𝑥 + 𝑦 + 𝑧 + 𝑤 = 2𝑥 + 𝑦 − 𝑧 + 𝑤 = 0}.
Let 𝜉 = (𝑎, 𝑏, 𝑐, 𝑑) ∈ 𝑆 ∩ 𝑇, then 𝑎 + 𝑏 + 𝑐 + 𝑑 = 0, 2𝑎 + 𝑏 − 𝑐 + 𝑑 = 0
1 3
⇒ 𝑐 + 𝑑 = −(𝑎 + 𝑏), 𝑐 − 𝑑 = 2𝑎 + 𝑏 ⇒ 𝑐 = 𝑎, 𝑑 = − 𝑎 − 𝑏
2 2
𝑎 3 1 3
So 𝜉 = (𝑎, 𝑏, 𝑐, 𝑑) = (𝑎, 𝑏, , − 𝑎 − 𝑏) = 𝑎 (1,0, , − ) + 𝑏(0,1,0, −1).
2 2 2 2
1 3
Now {(1,0, , − ) , (0,1,0, −1)} is L.I. and hence a basis of 𝑆 ∩ 𝑇. Therefore dim(𝑆 ∩ 𝑇) = 2.
2 2

[Some observation to directly say the dimension: If the space is ℝ𝑛 and in the subspace 𝑆 there
are 𝑚 independent equations (obviously, linear and homogeneous), then dim(𝑆) = 𝑛 − 𝑚. This
technique is based on Rank-nullity theorem of linear transformation]
 13

21. Find a basis & determine the dimension of the subspace 𝑆 of the vector space ℝ2×2
𝑎 𝑏 𝑎 𝑏
(i) 𝑆 = {( ) ∈ ℝ2×2 : 𝑎 + 𝑏 = 0}, (ii) 𝑆 = {( ) ∈ ℝ2×2 : 𝑎 = 𝑑 = 0}
𝑐 𝑑 𝑐 𝑑
(iii) 𝑆 = the set of all 2 × 2 real diagonal matrices,
(iv) 𝑆 = the set of all 2 × 2 real symmetric matrices,
(v) 𝑆 = the set of all 2 × 2 real skew symmetric matrices.
𝑎 𝑏
Answer: (i) Let 𝜉 = ( ) ∈ 𝑆. Then 𝑎 + 𝑏 = 0 ⇒ 𝑏 = −𝑎.
𝑐 𝑑
𝑎 −𝑎 1 −1 0 0 0 0 1 −1 0 0 0 0
Thus 𝜉 = ( ) = 𝑎( )+𝑐( )+𝑑( ) ∈ 𝐿 ({( ),( ),( )}).
𝑐 𝑑 0 0 1 0 0 1 0 0 1 0 0 1
1 −1 0 0 0 0 0 0
Let now consider the equation 𝑐1 ( ) + 𝑐2 ( ) + 𝑐3 ( )=𝜃=( )
0 0 1 0 0 1 0 0
𝑐1 −𝑐1 0 0
⇒ (𝑐 𝑐 )=( ) ⇒ 𝑐1 = 0, 𝑐2 = 0, 𝑐3 = 0.
2 3 0 0
1 −1 0 0 0 0
Thus {( ),( ),( )} is L.I. and hence basis of 𝑆. Also dim(𝑆) = 3.
0 0 1 0 0 1
(ii) Left to the reader. [similar to (iii)]
𝑎 0 𝑎 0
(iii) By definition of diagonal matrix, here 𝑆 = {(
) : 𝑎, 𝑑 ∈ ℝ}. Now let 𝜉 = ( ) ∈ 𝑆.
0 𝑑 0 𝑑
1 0 0 0 1 0 0 0
Then 𝜉 = 𝑎 ( )+𝑑( ) ∈ 𝐿 ({( ),( )}).
0 0 0 1 0 0 0 1
1 0 0 0 1 0 0 0
Obviously, {( ),( )} is L.I. So {( ),( )} is a basis of 𝑆 and dim(𝑆) = 2.
0 0 0 1 0 0 0 1
𝑎 𝑏 𝑎 𝑏
(iv) By definition, here 𝑆 = {( ) : 𝑎, 𝑏, 𝑑 ∈ ℝ}. Now let 𝜉 = ( ) ∈ 𝑆.
𝑏 𝑑 𝑏 𝑑
1 0 0 1 0 0 1 0 0 1 0 0
Then 𝜉 = 𝑎 ( )+𝑏( )+𝑑( ) ∈ 𝐿 ({( ),( ),( )}).
0 0 1 0 0 1 0 0 1 0 0 1
1 0 0 1 0 0
Obviously, {( ),( )( )} is L.I. (check it!) and hence a basis of 𝑆. So dim(𝑆) = 3.
0 0 1 0 0 1
𝑎 𝑏
(v) By definition, here 𝑆 = {( ) : 𝑎, 𝑏, 𝑑 ∈ ℝ}.
−𝑏 𝑑
Remaining part left to the reader.

22. 𝑉 is the vector space of all 2 × 2 real matrices. Show that the set 𝑆 is a basis of 𝑉.
1 1 1 1 1 0 0 1 1 1 0 0 1 1 0 1
(i) = { ( ),( ),( ),( )} , (ii) 𝑆 = { ( ),( ),( ),( )}.
1 0 0 1 1 1 1 1 0 0 1 1 1 0 1 1
 14

1 1 1 1 1 0 0 1
Answer: (i) Let 𝑆 = { ( ),( ),( ),( )} and consider the equation
1 0 0 1 1 1 1 1
1 1 1 1 1 0 0 1 𝑎 𝑏
𝑐1 ( ) + 𝑐2 ( ) + 𝑐3 ( ) + 𝑐4 ( )=𝜉=( ). ------------ (1)
1 0 0 1 1 1 1 1 𝑐 𝑑
𝑐1 + 𝑐2 + 𝑐3 = 𝑎 𝑐1 𝑎 1 1 1 0
𝑐1 + 𝑐2 + 𝑐4 = 𝑏 𝑐2 𝑏 1 1 0 1
or, 𝑐 + 𝑐 + 𝑐 = 𝑐 or, 𝐴 (𝑐 ) = ( ) , where 𝐴 = ( ).
1 3 4 3 𝑐 1 0 1 1
𝑐2 + 𝑐3 + 𝑐4 = 𝑑 𝑐4 𝑑 0 1 1 1
0 0
When 𝜉 = ( ), then unique solution (zero solution) of (1) gives the linear independence of 𝑆.
0 0
Again for arbitrary 𝜉 ∈ ℝ2×2 , unique solution of (1) gives 𝐿(𝑆) = ℝ2×2 .
From properties of system of linear equations, for both these conditions we have to det(𝐴) ≠ 0.
Now det(𝐴) = −3 ≠ 0. So 𝑆 is L.I. and 𝐿(𝑆) = ℝ2×2 , i.e., 𝑆 is a basis of ℝ2×2 .
(ii) Left to the reader.

23. 𝑉 is the vector space of all 3 × 1 real matrices. Show that the set 𝑆 is a basis of 𝑉.
1 1 1 2 1 1
(i) = { (1) , (1) , (0)} , (ii) 𝑆 = { (1) , (2) , (1)}.
1 0 0 1 1 2
1 1 1 0
Answer: (i) Let us consider the equation 𝑐1 (1) + 𝑐2 (1) + 𝑐3 (0) = (0).
1 0 0 0
⇒ 𝑐1 + 𝑐2 + 𝑐3 = 0, 𝑐1 + 𝑐2 = 0, 𝑐1 = 0 ⇒ 𝑐1 = 𝑐2 = 𝑐3 = 0.
𝑎 1 1 1 𝑎
So 𝑆 is L.I. Again let 𝜉 = (𝑏 ) ∈ 𝑉 and suppose 𝑑1 (1) + 𝑑2 (1) + 𝑑3 (0) = (𝑏 ).
𝑐 1 0 0 𝑐
1 1 1 𝑑1 𝑎
⇒ (1 1 0) (𝑑2 ) = (𝑏 ) … … … … (1)
1 0 0 𝑑3 𝑐
1 1 1 We can also find out the solution
Now |1 1 0| = −1 ≠ 0, so (1) has unique solution. So 𝜉 ∈ 𝐿(𝑆). 𝑑1 = 𝑐, 𝑑2 = 𝑏 − 𝑐, 𝑑3 = 𝑎 − 𝑏
1 0 0
Again as 𝜉 is arbitrary in 𝑉, so 𝐿(𝑆) = 𝑉. Therefore 𝑆 is a basis of 𝑉.
[(second part using homomorphism): Again we know the vector space 𝑉 of all 3 × 1 real matrices is
isomorphic to ℝ3 . So dimension of 𝑉 is 3. Now 𝑆 is L.I. and 𝑜(𝑆) = 3, so 𝑆 is a basis of 𝑉.
Note: We can also do the both L.I. and linear span of 𝑆 as similar in Ex. 22 (i)]
(ii) Left to the reader.
 15

24. Find the co-ordinate vector of 𝛼 in ℝ3 relative to the ordered basis (𝛼1 , 𝛼2 , 𝛼3 ) .
(i) 𝛼1 = (0,3,1), 𝛼2 = (1,1,0), 𝛼2 = (1,0,1) and 𝛼 = (0,1,1). It is given that (𝛼1 , 𝛼2 , 𝛼3 ) is basis,
so we just find the coordinates.
(ii) 𝛼1 = (2,3,3), 𝛼2 = (2,1,1), 𝛼2 = (1,2,1) and 𝛼 = (1,1,2).
Answer: (i) Consider the equation 𝑐1 𝛼1 + 𝑐2 𝛼2 + 𝑐3 𝛼3 = 𝛼. Then
𝑐2 + 𝑐3 = 0 0 1 1 𝑐1 0
3𝑐
𝑐1 (0,3,1) + 𝑐2 (1,1,0) + 𝑐3 (1,0,1) = (0,1,1) ⇒ 1 + 𝑐2 = 1 ⇒ (3 𝑐
1 0) ( 2 ) = (1).
𝑐1 + 𝑐3 = 1 1 0 1 𝑐3 1
1 1 1 1 1 1
Solving we get 𝑐1 = , 𝑐2 = − , 𝑐3 = . So coordinate of 𝛼 is ( , − , ).
2 2 2 2 2 2

(ii) Left to the reader.

25. Show that the set ℂ (the set of all complex numbers) forms a vector space over the field ℝ.
What is the dimension of ℂ over ℝ?
Answer: The first part is obvious.
Now ℂ = {𝑎 + 𝑖𝑏: 𝑎, 𝑏 ∈ ℝ}
So an arbitrary complex number can be expressed as 𝑎 + 𝑖𝑏 = 𝑎. 1 + 𝑏. 𝑖. Therfore ℂ = 𝐿({1, 𝑖}).
Again {1, 𝑖} is L.I. So dimension of ℂ over ℝ is 2 and its standard basis = {1, 𝑖}.

26. Show that the set ℝ (the set of all real numbers) forms a vector space over the field ℚ. What
is the dimension of ℝ over ℚ?
Answer: First part is obvious.
Now we claim that dimension of ℝ over ℚ is infinite (uncountable).
If possible, let dimension of ℝ over ℚ is finite, say, 𝑛 and 𝑆 = {𝑥1 , 𝑥2 , ⋯ , 𝑥𝑛 } ⊂ ℝ is a basis.
Then any 𝑥 ∈ ℝ can be expressed as 𝑥 = ∑𝑛𝑖=1 𝑐𝑖 𝑥𝑖 , for some 𝑐𝑖 ∈ ℚ. So
𝑛

ℝ = {∑ 𝑐𝑖 𝑥𝑖 : 𝑥𝑖 ∈ 𝑆, 𝑐𝑖 ∈ ℚ} = 𝑥1 ℚ + 𝑥2 ℚ + ⋯ + 𝑥𝑛 ℚ.
𝑖=1

Now each 𝑥𝑖 ℚ’s is countable, so the R.H.S. is countable set, but we know ℝ is uncountable.
Thus we arise at a contradiction. Hence we can conclude that dimension of ℝ over ℚ is infinite.