Module 2.
Solution of Some First-Order
DE
Different types of differential equations have different methods of solving. There is no general method
of attack that will solve all types of differential equations.
2.1 Variable Separable
Many first-order de can be reduced to the form
g (y) y ′ = f (x) , (1a)
By algebraic manipulations. Since y ′ = dy/dx , we find it convenient to write
g (y) dy = f (x) dx, (1b)
But we keep in mind that this is merely another way of writing (1a). Such an equation is called a
separable equation, because in (1b) the variables x and y are separated so that x appears only on the
right and y appears only on the left.
To solve (1a), we integrate on both sides with respect to x, obtaining
∫ g (y) dy
dx dx = ∫ f (x) dx + c.
Now on the left we can switch to y as the variable of integration. By calculus, ( ) dx = dy , so we
dy
dx
get
∫ g (y) dy = ∫ f (x) dx + c, (2)
If we assume that f and g are continuous functions, the integrals in (3) will exist, and by evaluating
these integrals we obtain the general solution of (1a).
Example 1
Solve the de
9yy ′ + 4x = 0
Solution. By separating variables we have
9y dy =− 4x dx
By integrating on both sides we obtain the general solution
9 2 2 x2 y2 c̃
2 y =− 2x + c̃, thus 9 + 4 =c (c = 18 )
The figure below represents the solution curves.
Example 2
Solve the de
y′ = 1 + y2 .
Solution. By separating variables and integrating we obtain
dy
1+y 2 = dx, y = x + c, y = tan (x + c) .
It is of great importance to introduce the constant of integration immediately when the integration is
performed. y = tan x + c (with c≠0) would not be a solution!
Example 3 Exponential growth or decay
Exponential growth of decay is governed by the de
y′ = ky
With positive or negative (fixed) k , respectively. This equation is separable. We obtain
dy
y = k dx, ln |y | = k x + c̃
Taking exponentials (Note that eln a = a, b ln a = ln ab , and e(a+b) = ea eb ), we get
|y | = ekx+c̃ = ekx ec̃ , thus y = cekx
where c =+ ec̃ when y > 0 and c =− ec̃ when y < 0 , and we can also admit c = 0 (giving y ≡ 0 ). This is
the general solution. The figure below represents the solution curves. You will observe that when c is
negative, the curve is in quadrant III and IV only (y is less than 0).
Example 4 Initial value problem
Solve the initial value problem
dy 2y
dx = x, y(1) = 1
Solution: By separating variables we have
dy 2dx .
y = x
By integration, we obtain a family of solutions
ln |y | = 2 ln |x| + c
Exponentiation on both sides, (Note that eln a = a, b ln a = ln ab , and e(a+b) = ea eb )
2
eln y = e(ln x +c)
|y| = ec x2 .
If we now put c =+ ec when y > 0 and c =− ec when y < 0 , we can write
y = cx2 ,
This is the general solution. From this and the initial condition y (1) = 1
2
1 = c(1) , then c = 1
The answer is
y = x2
The figure below represents the solution curves. The first curve at the upper half is the particular
solution, y = x2 . You will observe that when c is negative, the curve is in quadrant III and IV only (y is
less than 0).
Example 5 Initial value problem. Bell-shaped curves
Solve the initial value problem
y ′ =− 2xy, y (0) = 1.
Solution. Separating variables
dy
y =− 2x dx
Integrating,
ln |y| =− x2 + c̃
Exponentiating,
2
|y | = e−x +c˜
Setting c =+ ec̃ when y > 0 , and c =− ec̃ when y < 0 , and admitting also c = 0 (which gives the
solution y ≡ 0 ), we get the general solution
2
y = ce−x .
This represent so-called “bell-shaped curves” which play a role in heat conduction and in probability and
statistics. The figure below shows some of them.
From our initial condition y (0) = 1
2
y = ce−x , 1 = ce0 , theref ore c = 1
The particular solution of the initial value problem is
2
y = e−x
2
The second curve at the upper half of the graph below is the particular solution, y = e−x . You will
observe that when c is negative, the curve is in quadrant III and IV only (y is less than 0).
