Modeling with Seperable First Order Differential Equations

1 Separation of Variables
We first consider a simple first order differential equation:
dy
= f (x, y)
dx
This question can potentially be solved using the technique of Separation of Variables. In order to perform
separation of variables, f (x, y) must be separable. This means f (x, y) can be rewritten as two distinct
functions, g(x), h(y), where is g(x) is only a function of x and h(y) is only a function of y. Once this has
been done, the functions and their corresponding differential can be isolated, one to each side, to obtain:

h(y) dy = g(x) dx

Integrating both sides yields:

Z Z
h(y) dy = g(x) dx → H(y) + C1 = G(x) + C2

C1 and C2 are arbitrary constants, H(y) is the anti derivative of h(y), G(x) is the anti derivative of g(x).
To further simplify, C1 and C2 can be combined into a single arbitrary constant, C, as the sum or difference
of two arbitrary constants gives another arbitrary constant. The final expression is:

H(y) = G(x) + C

C is determined with using a given constraint, called a boundary condition.

It is typical that f (x, y) is a product of functions, f (x, y) = g(x) · h(y). In this case, separation of
variables gives:
dy
= g(x) dx
h(y)
If h(y) = y then the problem becomes: Z Z
dy
= g(x) dx
y
The integral on the left hand side occurs frequently and its integral is as follows:
Z
dy
= ln |y| + C
y
Separation of Variables is a simple and powerful technique that provides solutions to several different
physical models. The solutions provide insight into these physical phenomena.

1
2 The Various Models
Newton Cooling
Newton’s Law of Cooling states that the rate of cooling of an object at a temperature, H, with respect
to time, t, is proportional the difference between its temperature and the temperature of the surrounding
medium, Hs .
dH
= −k(H − Hs )
dt
Separate the variables so that H is on the same side as dH and dt is on the other side. Then integrate.
Z Z
dH
= −k dt
H − Hs

ln |H − Hs | = −kt + C
|H − Hs | = e−kt+C
note that: e−kt+C = e−kt eC
eC will always be greater than zero, so let’s call that an arbitrary, positive constant A.

|H − Hs | = Ae−kt

H − Hs = ±Ae−kt
H(t) = Hs ± Ae−kt
What will happen to the temperature of the object the longer it is kept in a medium with a temperature
difference? That is, as the time becomes arbitrarily long, what is the object’s expected temperature?

Consider: lim Hs ± Ae−kt

t→∞

Recall that limt→∞ e−t = 0, so:

lim Hs ± Ae−kt = Hs
t→∞

This says that after a long time has passed, the temperature of the object will be at the temperature of
the surrounding medium. In other words, the object tends toward achieving thermal equilibrium with its
surroundings. This result is consistent with everyday experience and thermodynamic principles.

2
 Radioactive Decay
The rate at which a radioactive isotope decays is quite nearly constant and proportional to the mass of
the isotope. The rate of change of the isotope’s mass with respect to time is proportional to its mass:
dm
= −km
dt
This proportionality is negative because the mass is decreasing. We want to find a function, m(t), so that
we can predict the mass at any given time or so we can find the time it will take to reach a given mass.

We need to find m(t), so separate the variables and integrate:

Z Z
dm
= −k dt
m

ln(m) = −kt + C
Mass is always a positive quantity, so we can remove the absolute value.

m = e−kt+C

note that: e−kt+C = e−kt eC

eC will always be greater than zero, so let’s call that a positive constant A.

m(t) = Ae−kt

Now at t = 0, m(t) = m0 :
m0 = Ae−k0 → m0 = A
m(t) = m0 e−kt
Suppose we let the sample decay to half of its original mass:
   
1 −kt 1 −kt 1 1
m0 = m0 e → =e → ln = −kt → − ln = kt → ln(2) = kt
2 2 2 2

ln(2)
t=
k
This expression is called the half-life, as it is the amount of time required for half of the radioactive nuclei
present in a sample to decay. It is an interesting fact that the half-life is a constant that does not depend
on the number of radioactive nuclei initially present in the sample, but only on the radioactive substance.

What happens to the sample if it was left to decay for long period of time?

Consider: lim m0 e−kt

t→∞
−t
Recall that limt→∞ e = 0, so:
lim m0 e−kt = 0
t→∞

This is as expected, since this is a decay model. After a long enough time has passed, the substance should
decay to a point to where it no longer exists.

3
 Unlimited Population Growth
The population, P, grows at a rate that is proportional to its size: dP
dt
= kP
To obtain a function from the given differential equation, separate the variables and integrate both sides:
Z Z
dP
= k dt
P
ln |P | = kt + C
|P | = ekt+C
note that: ekt+C = ekt eC
eC will always be greater than zero, so let’s call that an arbitrary positive constant, A.

|P | = Aekt

P (t) = ±Aekt
Suppose the growth continue for a long time. What would happen to the population after a long time has
passed? To find out what happens to the population after a very long time passes, take the limt→∞ P (t).

Consider: lim Aekt

t→∞

Recall that limt→∞ et = ∞, so:

lim Aekt = ∞
t→∞

This is clearly not a realistic result as a population cannot grow infinitely here on Earth, which has a
finite number of resources. This model fails to account for the environment’s ability to sustain population
growth.

4
 The Logistic Population Growth Model
One population growth is P (t) = Aekt , but as shown earlier it does not consider real world conditions.

A different model that takes into account real world constraints is called the Logistic Model.
The Logistic Equation states:
dP
= P (b − P )
dt
P is the population and b is a constant called the carrying capacity; we shall see the significance of this
number later. Both quantities are typically measured in the thousands or millions.

To find P (t), apply separation of variables and integrate:

Z Z
dP
= dt
P (b − P )

Use partial fraction decomposition in order to integrate the left hand side:
Z  
1 1 1
+ dP = t + C1
b P b−P

ln |P | − ln |b − P | = bt + bC1
Since C1 is just an arbitrary constant and b is just a constant as well, their product is just another constant,
which we will call C.  
 P 
ln   = bt + C
b−P
 
 P  bt+C
b − P  = e
 

note that: ebt+C = ebt eC

eC will always be greater than zero, so let’s call that arbitrary, positive constant A.
P
= Aebt → P = Aebt (b − P ) → P = bAebt − P Aebt
b−P

P + P Aebt = bAebt → P (1 + Aebt ) = bAebt

bAebt
P (t) =
1 + Aebt
bAebt
Consider: lim
t→∞ 1 + Aebt

bAe bt bAe bt
Notice that dominant term in the denominator is bAebt . As t becomes arbitrarily large, 1+Ae bt ≈ Aebt . The

top and bottom terms cancel and we are just left with b. So after a long time has passed, the population
will be at b. It turns out that no matter where you start, time approaches infinity, the value of P (t) will
always approach b. This is called the carrying capacity of an environment, as that value is the maximum
population that environment can sustain.

5
Numerical Examples
Cooling Eggs
A hard-boiled egg at 98 ◦ C is put into a sink filled with water at 18 ◦ C. After five minutes, the egg’s tem-
perature is 38 ◦ C. Assuming that the water has not warmed appreciably, how much longer will it take the
egg to reach 20 ◦ C?

The initial conditions are: Hs =18 ◦ C, H(0)=98 ◦ C, and H(5)=38 ◦ C.

Use the previously derived result: H(t) = Hs ± Ae−kt

Use the first initial condition to see if A should be positive or negative.

H(0) = 18 ± Ae−k0 → 98 = 18 ± A → 80 = ±A

Recall that A must be a positive constant, so that means A = 80

Now we must find what k is, so use the second condition to find it:

ln(0.25)
H(5) = 18 + 80e−5k → 38 = 18 + 80e−5k → k = − ≈ 0.28
5
So this means H(t) = 18 + 80e−0.28t

To find how long it will take to reach 20 ◦ C, solve:

ln(0.025)
20 = 18 + 80e−0.28t → t = − ≈ 13.17
0.28
So it will take about 13 minutes for the egg to cool to 20 ◦ C

6
 Plutonium-239
The half-life of plutonium-239 is 24,360 years. If 10 g of plutonium is released into the atmosphere by a
nuclear accident, how many years will it take for 80% of the isotope to decay?

Use the previously derived general formula: m(t) = m0 e−kt

One can see that m0 = 10.

To find k, use the half-life equation:

ln(2) ln(2) ln(2)

t= →k= →k=
k t 24360
ln(2)
m(t) = 10e− 24360 t
When 80% of the sample has decayed, there will only be 2 g left:
ln(2) ln(2) ln(2) ln(2)
2 = 10e− 24360 t → 0.2 = e− 24360 t → ln(0.2) = − t → ln(5) = t
24360 24360
24360 ln(5)
t= ≈ 56, 562
ln(2)
It will take about 56,562 years for the sample to decay to 80% of its original mass.
Bottom line: handle radioactive waste with extreme caution!