THE UNIVERSITY OF HONG KONG

DEPARTMENT OF MATHEMATICS

MATH1013 University Mathematics II

Tutorial 4 Solutions

1. (a) Note that

5 x 5 x
lim (5x − 2x ) = lim 2x (( ) − 1) = ( lim 2x ) [ lim (( ) − 1)] = (∞)(∞ − 1) = ∞
x→∞ x→∞ 2 x→∞ x→∞ 2
by example 3.14.
√
x x2
(b) For x < 0, we have √ =− . Therefore, by proposition 3.9, we have
x2 + 4 +4x2
√ √
x ⎛ x2 ⎞ ⎛ x2 ⎞
lim √ = lim − = lim − = −1.
x→−∞ x2 + 4 x→−∞ ⎝ x2 + 4 ⎠ x→−∞ ⎝ x2 ⎠

(c) Since −1 ⩽ cos x ⩽ 1, we have 1 ⩽ 2 cos x + 3 ⩽ 5. It follows that

ex ⩽ (2 cos x + 3)ex ⩽ 5ex

for any x. As lim ex = 0 and lim 5ex = 0, we have lim (2 cos x + 3)ex = 0 by the
x→−∞ x→−∞ x→−∞
sandwich theorem.

2. (a) Note that √ √

1 x 3 1 3x 3 1 3x
lim (1 + ) = lim (1 + ) = lim (1 + ) .
x→∞ 3x x→∞ 3x x→∞ 3x
Using the substitution y = 3x, this is equal to
√
3 1 y √
lim (1 + ) = 3 e.
y→∞ y

(b) We have
x x− 12 1
2x + 1 x 2 x
1 1 1 2

lim ( ) = lim (1 + ) = lim (1 + ) = lim (1 + ) (1 + ) .

x→∞ 2x − 1 x→∞ 2x − 1 x→∞ x − 12 x→∞ x − 12 1
x− 2

x− 12
1 1 1 y
Using the substitution y = x − , we find that lim (1 + ) = lim (1 + ) = e.
2 x→∞ x − 12 y→∞ y
1
2
1 1
Also, it is clear that lim (1 + 1
) = (1) 2 = 1. Therefore, the given limit is equal to
x→∞ x− 2
(e)(1) = e.

1
 (c) Note that √
x 1
lim+ 1 + x = lim+ (1 + x) x .
x→0 x→0

1
Using the substitution y = , this is equal to
x
1 y
lim (1 + ) = e.
y→∞ y

3. Yes. Firstly, we have

2π
∣π + cos ∣ ⩽ π + 1 < 5.
x
2π
This implies ∣(π + cos ) sin (1 − cos x)∣ ⩽ 5 ∣sin (1 − cos x)∣, and hence
x
2π
−5 ∣sin (1 − cos x)∣ ⩽ (π + cos ) sin (1 − cos x) ⩽ 5 ∣sin (1 − cos x)∣ .
x
Note that lim sin (1 − cos x) = sin (1 − cos 0) = 0. Therefore, both the lower bound and the
x→0
upper bound approach 0 when x tends to 0. This implies lim f (x) = 0 by the sandwich
x→0
theorem. Therefore, we can define an extension f̃ of f such that f̃(0) = 0. Then

lim f̃(x) = lim f (x) = 0 = f̃(0),

x→0 x→0

which means f̃ is continuous at 0.

4. (a) f (x) = 0 for all x ∈ (0, 1)

(b) f (x) = x for all x ∈ (0, 1)

(c) f (x) = x for all x ∈ R

⎧
⎪ x if 0 ⩽ x < 1,
⎪
⎪
(d) f (x) = ⎨ 0 if x = 1,
⎪
⎪
⎩ x − 2 if 1 < x ⩽ 2
⎪

5. (a) Suppose on the contrary that there exist b < c in the same interval such that f (b) and
f (c) have different signs (note that f (b), f (c) ≠ 0). Note that f is defined on [b, c] since
b and c belong to the same interval. As f is continuous, there exists x ∈ (b, c) such that
f (x) = 0 by the intermediate value theorem, contradicting the assumption that b and c
belong to the same interval. Thus, all images in the same interval must have the same
sign.

Remark. As suggested by its name, the sign test can be used to test the sign of a
function. This will be a useful tool in section 4.

2
 (x − 5)(x + 2) 4
(b) We have f (x) = . The function is undefined when x = − , 5, and f (x) = 0
(x − 5)(3x + 4) 3
4 4
when x = −2. Thus, we consider the intervals (−∞, −2), (−2, − ), (− , 5) and (5, ∞).
3 3
By part (a), as f is continuous, it has a constant sign on each interval. Thus, we can pick
any point in each interval to test for the sign.
1
ˆ Since f (−3) = > 0, f has a positive sign on (−∞, −2).
5
5 1 4
ˆ Since f (− ) = − < 0, f has a negative sign on (−2, − ).
3 3 3
1 4
ˆ Since f (0) = > 0, f has a positive sign on (− , 5).
2 3
4
ˆ Since f (6) = > 0, f has a positive sign on (5, ∞).
11

Additional problems:

6. (a) Note that x 2 ln x = (eln x ) 2 ln x = e(ln x)⋅ 2 ln x = e 2 . Therefore, it is clear that lim x 2 ln x = e 2 .
1 1 1 1 1 1

x→∞

1 1
(b) Note that lim+ ln x = −∞ and lim = 0. By proposition 3.11, we get lim+ = 0.
x→0 y→−∞ y2 x→0 (ln x)2

√ √
(c) Note that x2 − sin2 x < x2 = −x for all x < −1 (which ensures x2 − sin2 x > 0). As x3 < 0
for x < −1, we have √
x2 − sin2 x −x 1
0⩾ > 3 = − 2.
x3 x x
√
1 x2 − sin2 x
As lim 0 = 0 and lim (− 2 ) = 0, we have lim = 0 by the sandwich
x→−∞ x→−∞ x x→−∞ x3
theorem.

(d) We have

1 x 1 x+2 1 −2
lim (1 + ) = lim (1 + ) (1 + ) = e ⋅ (1)−2 = e.
x→∞ x+2 x→∞ x+2 x+2

7. (a) Both x and ∣x∣ − 1 are continuous functions. So their quotient is also continuous by
proposition 3.12(d). Thus, there is no point at which f is discontinuous.

Remark. ±1 do not belong to the domain of f , so we do not talk about the continuity
at these two points.

(b) It suffices to check the endpoints x = ±1 as the function must be continuous at other
points (since the formulae only involve elementary functions).
ˆ For x = −1, we check that
x2 − 1
lim − g(x) = lim − = lim (x − 1) = −2
x→−1 x→−1 x + 1 x→−1−
and
lim g(x) = lim + (x3 − 1) = (−1)3 − 1 = −2 = lim − g(x).
x→−1+ x→−1 x→−1

3
 This shows lim g(x) = −2 = g(−1), and hence g is continuous at −1.
x→−1

ˆ For x = 1, we check that

lim g(x) = lim− (x3 − 1) = 0

x→1− x→1

and
lim g(x) = lim+ (2x2 − 1) = 1 ≠ lim− g(x).
x→1+ x→1 x→1

This shows lim g(x) does not exist, and hence g is discontinuous at 1.
x→1

(c) It suffices to check the endpoints x = 0, 1.

ˆ For x = 0, we check that
sin x sin x √
lim− h(x) = lim− √ = lim− ( ⋅ (− −x)) = 1 ⋅ 0 = 0
x→0 x→0 −x x→0 x
and
lim h(x) = lim+ ln (1 − x) = ln 1 = 0 = lim− h(x).
x→0+ x→0 x→0

This shows lim h(x) = 0 = h(0), and hence h is continuous at 0.

x→0

ˆ For x = 1, using the substitution y = 1 − x, we find that

lim h(x) = lim− ln (1 − x) = lim+ ln y = −∞.

x→1− x→1 y→0

This shows lim h(x) does not exist, and hence h is discontinuous at 1.
x→1

8. Since f is a polynomial in each of the open intervals (−∞, −2), (−2, 0), (0, 1) and (3, ∞), it
must be continuous at all points in these intervals. It remains to check the endpoints of the
intervals.

ˆ f is continuous at −2 if and only if lim − (2kx + ℓ) = lim + (−ℓx2 + 2k + 2) (since it is

x→−2 x→−2
clear that f (−2) is equal to the right-hand limit). This means −4k + ℓ = −4ℓ + 2k + 2, i.e.

6k + 2 = 5ℓ. (1)

ˆ f is continuous at 0 if and only if lim− (−ℓx2 + 2k + 2) = lim+ (x + k + ℓ + 1) (again since

x→0 x→0
f (0) is equal to the right-hand limit). This means 2k + 2 = k + ℓ + 1, i.e.

k + 1 = ℓ. (2)

ˆ f is continuous at 3 because lim f (x) = lim+ f (x) = f (3).

x→3 x→3

Thus, we only need (1) and (2). Combining these equations, we obtain 6k + 2 = 5(k + 1).
This gives k = 3, and hence ℓ = k + 1 = 4. We check that both (1) and (2) are satisfied when
(k, ℓ) = (3, 4). Therefore, f is a continuous function if and only if (k, ℓ) = (3, 4).

Remark. We do not need to check the endpoint 1 because it is not in the domain of f . The
checking of the continuity at 3 may also be regarded as ‘obvious’ (similar to those interior
points).

4
 1
9. Since −1 ⩽ sin ⩽ 1, we have
g(x)
1
2 − ∣sin g(x)∣ ⩽ 2 + sin g(x) sin ⩽ 2 + ∣sin g(x)∣ .
g(x)

Note that lim ∣sin g(x)∣ = 0 since g(x) → 0. It follows that both 2 − ∣sin g(x)∣ and 2 + ∣sin g(x)∣
x→0
tend to 2 when x tends to 0. By the sandwich theorem, we have lim f (x) = 2. Therefore, f
x→0
has an extension which is continuous at 0 as we can define f (0) = 2.

10. (a) True. By the intermediate value theorem, there exists a root in (1, 2) and there exists
another root in (2, 3).

(b) False. The function f (x) = (x − 1)2 satisfies f (0) = f (2) = 1 > 0, but f (1) = 0.

(c) False. Define

−1 if x ⩽ 0,
f (x) = {
1 if x ⩾ 2.
Note that f is continuous. Also, f (0) = −1 < 0 and f (2) = 1 > 0, but f (x) = 0 has no
solution.

Remark. The intermediate value theorem cannot be used since f is not defined on [0, 2]
in this example.

11. For every k ∈ R, we can find a sufficiently small a such that f (a) < k. Such an a exists
since lim f (x) = −∞. Similarly, we can find a sufficiently large b such that f (b) > k since
x→−∞
lim f (x) = ∞. As f is continuous on [a, b], there exists c ∈ (a, b) such that f (c) = k by the
x→∞
intermediate value theorem. This proves f is surjective as every k belongs to the range of f .

12. (a) Let f (t) = t3 + t2 − 2t − 1. Then f is a continuous function defined on [−1, 1]. As
f (−1) = 1 > 0 and f (1) = −1 < 0, there must be x ∈ (−1, 1) such that f (x) = 0 by the
intermediate value theorem. This gives x3 + x2 = 2x + 1.

(b) Let f (t) = cos t − 1000t2 . Then f is a continuous function defined on [0, π]. Since
f (0) = 1 > 0 and f (π) = −1 − 1000π 2 < 0, there must be x ∈ (0, π) such that f (x) = 0 by
cos x
the intermediate value theorem. As x ≠ 0, this gives = 1000.
x2
cos x
Remark. It is incorrect to apply the intermediate value theorem to g(x) = on an
x2
interval [a, b] with a < 0 < b. This is because g(0) is undefined. It is fine if the interval
[a, b] does not contain 0 instead as g is also continuous.

13. Let f (x) = 2x3 − 3x2 + 4x + 1. Note that f (−1) = −8 < 0 and f (0) = 1 > 0. Therefore, we can
apply the bisection method to the interval [−1, 0] to search for a root. The following table

5
gives the computation (correct to 6 decimal places).

interval midpoint c f (c)

(−1, 0) −0.5 −2
(−0.5, 0) −0.25 −0.21875
(−0.25, 0) −0.125 0.449219
(−0.25, −0.125) −0.1875 0.131348
(−0.25, −0.1875) −0.21875 −0.039490
(−0.21875, −0.1875) −0.203125 0.046959

This shows there is a root in the interval (−0.21875, −0.203125). We can use its midpoint
−0.2109375 to serve as an approximation to the root. The error is less than

(−0.203125) − (−0.21875)
= 0.0078125 < 0.01.
2