1.

2 Qualitative Methods: Phase Lines and Direction Fields 25

approximately two days, all solutions begin to exhibit similar behavior. From the general
solution (14), it is evident that for large t,
22.5
u(t) ≈ U(t) = 60 + (1.5 sin 2𝜋t − 2𝜋 cos 2𝜋t) , (15)
2.25 + 4𝜋 2
since ce−1.5t → 0 as t → ∞. The function U(t) in expression (15) is referred to as the steady-
state solution of Eq. (13). Using trigonometric identities, we can write U(t) in the form
(see Problem 31)
22.5
U(t) = 60 + √ sin(2𝜋t − 𝛿) ≈ 60 + 3.4831 sin(2𝜋t − 1.33645), (16)
2.25 + 4𝜋 2
√
where 𝛿 = cos−1 (1.5∕ 2.25 + 4𝜋 2 ). Comparing U(t) with T(t), we see that for large t the
air temperature within the building varies sinusoidally at the same frequency as the exter-
nal air temperature, but with a time lag of tlag = 1.33645∕(2𝜋) = 0.2127 days and an am-
plitude of only 3.4831◦ F about a mean temperature of 60◦ F. Does the qualitative behavior
of the steady state solution agree with what you expect based on physical reasoning and
experience?

u
External temperature
T(t) = 60 + 15 sin(2πt)
75

70

65

60

55

50

45

0 0.5 1 1.5 2 2.5 t

FIGURE 1.2.8 Direction field and integral curves for u′ + 1.5u = T(t). The variation in
external temperature is described by T(t) = 60 + 15 sin 2𝜋t.

PROBLEMS

Phase Line Diagrams. Problems 1 through 7 involve equa- 1. dy∕dt = y(y − 1)(y − 2), y0 ≥ 0
tions of the form dy∕dt = f (y). In each problem, sketch the 2. dy∕dt = ey − 1, −∞ < y0 < ∞
graph of f (y) versus y, determine the critical (equilibrium)
3. dy∕dt = e−y − 1, −∞ < y0 < ∞
points, and classify each one as asymptotically stable or un-
stable. Draw the phase line, and sketch several graphs of so- 4. dy∕dt = −2(arctan y)∕(1 + y2 ), −∞ < y0 < ∞
lutions in the ty-plane.
26 Chapter 1 Introduction

5. dy∕dt = y2 (y + 1)(y − 3), −∞ < y0 < ∞ (f) y′ = 1 + 2y

6. dy∕dt = ay + by , 2
a > 0, b > 0, y0 ≥ 0 (g) y′ = −2 − y
(h) y′ = y(3 − y)
7. dy∕dt = ay + by2 , a > 0, b > 0, −∞ < y0 < ∞
(i) y′ = 1 − 2y
Problems 8 through 13 involve equations of the form (j) y′ = 2 − y
dy∕dt = f (y). In each problem sketch the graph of f (y) ver-
sus y, determine the critical (equilibrium) points, and classify 24. The direction field of Figure 1.2.9.
each one as asymptotically stable, unstable, or semistable.
Draw the phase line, and sketch several graphs of solutions y
in the ty-plane.
4
8. dy∕dt = −k(y − 1)2 , k > 0, −∞ < y0 < ∞
3
9. dy∕dt = y2 (y2 − 1), −∞ < y0 < ∞
10. dy∕dt = y(1 − y2 ), −∞ < y0 < ∞ 2
√
11. dy∕dt = ay − b y, a > 0, b > 0, y0 ≥ 0
1
12. dy∕dt = y2 (4 − y2 ), −∞ < y0 < ∞
13. dy∕dt = y2 (1 − y)2 , −∞ < y0 < ∞
1 2 3 4 t
Direction Fields. In each of Problems 14 through 19 draw
a direction field for the given differential equation. Based on FIGURE 1.2.9 Direction field for Problem 24.
the direction field, determine the behavior of y as t → ∞. If
this behavior depends on the initial value of y at t = 0, de- 25. The direction field of Figure 1.2.10.
scribe the dependency.
14. y′ = 3 − 2y y
′
15. y = 2y − 3 4
16. y′ = 3 + 2y
3
17. y′ = −1 − 2y
18. y′ = 1 + 2y 2
19. y′ = y + 2
1
In each of Problems 20 through 23 draw a direction field for
the given differential equation. Based on the direction field,
determine the behavior of y as t → ∞. If this behavior de- 1 2 3 4 t
pends on the initial value of y at t = 0, describe this depen-
dency. Note that in these problems the equations are not of the FIGURE 1.2.10 Direction field for Problem 25.
form y′ = ay + b, and the behavior of their solutions is some-
what more complicated than the solutions shown in Figures 26. The direction field of Figure 1.2.11.
1.2.6 and 1.2.7.
20. y′ = y(4 − y) y
′
21. y = −y(5 − y) 1 2 3 4 t
′ 2
22. y = y
23. y′ = y(y − 2)2 –1
Consider the following list of differential equations, some of
which produced the direction fields shown in Figures 1.2.9 –2
through 1.2.14. In each of Problems 24 through 29 identify
the differential equation that corresponds to the given direc- –3
tion field.
(a) y′ = 2y − 1 –4
(b) y′ = 2 + y
(c) y′ = y − 2 FIGURE 1.2.11 Direction field for Problem 26.
(d) y′ = y(y + 3)
(e) y′ = y(y − 3)
 1.2 Qualitative Methods: Phase Lines and Direction Fields 27

27. The direction field of Figure 1.2.12. Applications.

32. If in the exponential model for population growth,
y dy∕dt = ry, the constant growth rate r is replaced by a growth
1 2 3 4 t rate r(1 − y∕K) that decreases linearly as the size of the pop-
ulation increases, we obtain the logistic model for population
growth,
–1
dy ( y)
= ry 1 − , (i)
–2 dt K
in which K is referred to as the carrying capacity of the pop-
–3 ulation. Sketch the graph of f (y), find the critical points, and
determine whether each is asymptotically stable or unstable.
–4 33. An equation that is frequently used to model the pop-
ulation growth of cancer cells in a tumor is the Gompertz
equation
FIGURE 1.2.12 Direction field for Problem 27.
dy
= ry ln(K∕y),
dt
28. The direction field of Figure 1.2.13.
where r and K are positive constants.
y (a) Sketch the graph of f (y) versus y, find the critical points,
and determine whether each is asymptotically stable or un-
5
stable.
4 (b) For each y in 0 < y ≤ K, show that dy∕dt, as given by the
3 Gompertz equation, is never less than dy∕dt, as given by the
logistic equation, Eq. (i) in Problem 32.
2
34. In addition to the Gompertz equation (see Problem 33),
1 another equation used to model the growth of cancerous tu-
mors is the Bertalanffy equation
1 2 3 4 t
dV
–1 = aV 2∕3 − bV,
dt
where a and b are positive constants. This model assumes
FIGURE 1.2.13 Direction field for Problem 28. that the tumor grows at a rate proportional to surface area,
while the loss of tumor mass due to cell death is proportional
29. The direction field of Figure 1.2.14. to the volume of the tumor. Sketch the graph of f (V) ver-
sus V, find the critical points, and determine whether each is
y asymptotically stable or unstable.
35. A chemical of fixed concentration ci flows into a con-
5
tinuously stirred tank reactor at a constant volume flow rate
4 ri and flows out at the same rate. While in the reactor, the
3 chemical undergoes a simple reaction in which it disappears
at a rate proportional to the concentration:
2
dc r c
1 = i ci − ri − kc, (i)
dt V V
where V is the volume of the reactor and k is the rate of re-
1 2 3 4 t
action.
–1
(a) Use the dimensionless variables
c t
FIGURE 1.2.14 Direction field for Problem 29. C= , 𝜏=
ci V∕ri
to express Eq. (i) in dimensionless form
30. Verify that the function in Eq. (11) is a solution of Eq. dC
(10). = 1 − C − 𝛼C, (ii)
d𝜏
31. Show
√ that A sin 𝜔t + B cos 𝜔t = R sin(𝜔t − 𝛿), where where
R = A2 + B2 and 𝛿 is the angle defined by R cos 𝛿 = A and kV
𝛼= .
R sin 𝛿 = −B. ri
28 Chapter 1 Introduction

(b) Determine the equilibrium solution of Eq. (ii), draw a The parameters n and g, growth rates of labor L and technol-
phase line diagram, and determine whether the equilibrium ogy A, respectively, appear in the equations
solution is asymptotically stable or unstable. Then sketch a dL dA
phase portrait with a representative set of solution curves. = nL, = gA.
dt dt
36. A pond forms as water collects in a conical depression of The product A(t)L(t) is referred to as effective labor and the
radius a and depth h. Suppose that water flows in at a constant output Y of the economy is given by Y = ALf (k). Assume
rate k and is lost through evaporation at a rate proportional that the production function f (k) satisfies the following con-
to the surface area. ditions:
(a) Show that the volume V(t) of water in the pond at time t (i) f (0) = 0, f (k) > 0 for k > 0,
satisfies the differential equation (ii) f ′ (k) > 0, f ′′ (k) < 0 for k > 0,
dV∕dt = k − 𝛼𝜋(3a∕𝜋h)2∕3 V 2∕3 , (iii) limk→0 f ′ (k) = ∞, limk→∞ f ′ (k) = 0.
For example, the function f (k) = ck𝛼 where 0 < 𝛼 < 1 satis-
where 𝛼 is the coefficient of evaporation.
fies conditions (i), (ii), and (iii).
(b) Find the equilibrium depth of water in the pond. Is the
equilibrium asymptotically stable? (a) Draw a phase line diagram of Eq. (i) by sketching the
(c) Find a condition that must be satisfied if the pond is not graphs of actual investment 𝜎f (k) and break-even invest-
to overflow. ment (n + g)k on the same set of coordinate axes. Show that
Eq. (i) has an asymptotically stable equilibrium solution k∗ .
37. The Solow model of economic growth (ignoring the ef-
(b) When k = k∗ , we say the Solow economy is on its
fects of capital stock depreciation) is
balanced growth path. Show that when k = k∗ , the output of
k′ = 𝜎f (k) − (n + g)k, (i) the economy grows at the combined growth rates of labor and
where k is capital stock per unit of effective labor, f (k) is technology,
GDP per unit of effective labor, and 𝜎, 0 < 𝜎 < 1, is the frac- dY
tion of gross domestic product (GDP) devoted to investment. = (n + g)Y.
dt

1.3 Definitions, Classification, and

Terminology
In Sections 1.1 and 1.2 we gave examples to introduce you to a number of important topics in
the context of first order differential equations: mathematical modeling, solutions, integral
curves, initial value problems, phase line diagrams, direction fields, equilibrium points,
and concepts of stability. Prior to embarking on an in-depth study of first order equations,
we briefly step back and give you a broader view of differential equations by presenting a
few important definitions, introducing some commonly used terminology, and discussing
different ways that differential equations are classified. This background information will
enhance your understanding of the subject in the following ways:
▶ It will provide you with an organizational framework for the subject;
▶ it will acquaint you with some of the language used to discuss the subject in a sensible
manner;
▶ it will give you perspective on the subject as a whole.

We begin with a definition of a differential equation.

DEFINITION Differential Equation. An equation that contains derivatives of one or more unknown
1.3.1 functions with respect to one or more independent variables is said to be a differential
equation.