Unit 9: First Order Dierential Equations with Applications

First Order Dierential Equations By a dierential equation, we mean an equation which involves the derivative of some unknown functon, say y , and the task at hand is to nd an explicit expression for y . We will study two types of rst order dierential equation, namely separable, and linear dierential equations ( shortened to D.E.s for the obvious reason)). Each will have its own prescribed method of solution.

Denition 9.1. By a separable dierential equation we will mean an equation which may be put into one of the following forms: dy f (x) dy g (y ) dy = or = or = f (x)g (y ) dx g (y ) dx f (x) dx To solve a dierential equation is to nd all possible solutions ( remember C in integration?). Strictly speaking we have already dealt with some D.E.s, as the following example will show. We rst outline some basic steps for solving such a problem.

Steps For Solving a separable Dierential Equation: Assuming that our D.E. is given in terms of being clear:
dy dx

with the generalization

(1) Recognize the problem as a separable D.E. (2) Seperate the equation into the form (terms not involving x)(dy ) = (terms not involving y)(dx) (3) Integrate both sides, adding one arbitrary constant, say C, to the x-side. This is done since adding an arbitrary constant to both sides after integrating is equivalent to adding just a single arbiratary constant to one side. (4) Solve for y if possible. If we can solve for y we get what is called the general solution, and if we can not solve for y we get what is called an implicit solution.
dy dx

Example 1. Solve:

= x2

Solution: First we recognize that we have a separable D.E. of either of the rst two forms listed above (with g(y)=1), and so proceed: dy dy y = x2 dx = =
x3 3

This is step (2) Begin step(3) step(3) and step (4)

x2 dx +C

So we have in this case found the general solution, x3 +C y= 3 2

Example 2. Solve:

dy dx

4x 3y 3

Solution: We rst recognize this as a separable D.E., and then move on: 3y 3 dy 3y 3 dy 3 y4 3 y4
4

= 4xdx = 4xdx
2

step (2) Begin step(3) step(3) and step (4)

= 4x +C 2 = 2x2 + C =4 2x + 4 C 3 2 3 =8 x2 + 4 C 3 3 =8 x2 + C 3 =
4 2

y4 y4 y4 y

8 2 x 3

+C
4

So once again we have a general solution, namely y =

8 2 x 3

+C

Solution: We rst recognize this as a separable D.E., and then move on: ey dy ey dy ey ln(ey ) y = (4x3 + 2x)dx = (4x3 + 2x)dx = x4 + x2 + C = ln(x4 + x2 + C ) = ln(x4 + x2 + C ) step (2) Begin step(3) step(3) and step (4)

So once again we have a general solution, namely y = ln(x4 + x2 + C )

Example 4. Solve:

dy dx

4x3 3x y +sin y

Solution:We rst recognize this as a separable D.E., and then move on: (y + sin y )dy (y + sin y )dy
y2 2

= (4x3 3x)dx

step (2)

= (4x3 3x)dx Begin step(3) = x4 3 x2 + C 2 = x4 3 x2 + C 2 step(3) and step (4)

cos y

1 y 2 cos y 2

Here, we see that we can not solve explicitly for y and so we have an implicit solution, namely 1 y 2 cos y = x4 3 x2 + C . 2 2

Side Conditions and D.E.s Recall that we used side conditions with integration to allow the constant of integration, C, to be found. Similarily we may use such conditions to solve for C here.

Example 5. Solve: Solve:

dy dx

y x2

given that y ( 1 ) = e. 2

) y( 1 2

Solution: We solve the D.E. as usual, and then use the side condition = e, to solve for C.:
1 dy y

= =

1 dx x2

step (2)

1 dy y

x2 dx Begin step(3)
x 1 1

ln(y ) =

+C

step(3) and step (4)

1 ln(y ) = x +C

eln(y) y We now use the fact that y ( 1 ) = e to get: 2 e e3 C Substituting C = 3 into (*) we get: y y y

= e x +C = e x eC
1

(*)

= e2 eC = eC =3 multiplying by e2

= e x e3 = e x +3 = e3 x
1 1 1

So here we have the specic solution, y = e3 x We now move on to the second type of D.E. mentioned above, linear differential equations.

Denition 9.2. By a linear dierential equation we will mean a D.E. which can be put into the following form: dy + p(x)y = q (x) dx where p(x) and q (x) are expressions that may or may not involve x, but very importantly they do not involve y .

The approach we learn here for solving linear D.E.s is much dierent from the approach we took to separable D.E.s. In order to give our method 5

we need the following denition.

Denition 9.3. Given a linear D.E. dy + p(x)y = q (x) dx we dene the integrating factor (x) of the D.E. by: (x) = e Note: when calculating zero.
p(x)dx

p(x)dx here we set the integration constant C to

The use of the integration factor is illustrated below, but often these steps are skipped ( see next theorem). We begin with a linear D.E.: dy + p(x)y = q (x) dx e
p(x)dx dy dx

+e

p(x)dx

p(x)y

= q (x)e = q (x)e =

p(x)dx p(x)dx p(x)dx

- multiply both sides by (x) -recognize the equivalence of the left hand sides -integrating both sides
p(x)dx

d (ye p(x)dx ) dx

ye

p(x)dx

q (x)e

dx

= e

p(x)dx

q (x)e

dx -multiply both sides by (x)

1 q (x)(x)dx -use the fact that (x) = e y = ( x) So we have the general solution of the problem. We now be state the above result:

Theorem 9.4. Given a linear dierential equation

+ p(x)y = q (x)
p(x)dx

with integrating factor (x) = e

, the general solution is given by:

y= Example 6. Solve:
dy dx 2 +x y=

1 (x)

q (x)(x)dx

cos(5x) x2

Solution: By the previous theorem we need only nd (x), and then apply the formula. x) 2 and q (x) = cos(5 so we have p(x) = x x2 2 1 p(x)dx = x dx = 2 x dx = 2 ln(x), so the integrating factor is given by 2 (x) = e p(x)dx = e2 ln(x) = eln(x ) = x2 Substituting into the general solution formula we then have: 1 q (x)(x)dx y = ( x) = = =
1 x2 1 x2 1 1 ( x2 5 cos(5x) 2 x dx x2

cos(5x)dx sin(5x) + C ) REMEMBER YOUR BRACKETS!

1 C = 5x 2 sin(5x) + x2 ) So we have the general solution:y =

sin(5x) 5x2

C ) x2

Example 7. Solve:

dy dx

2y = xe3x

Solution: We need only nd (x), p(x) and q (x), and then apply the formula. p(x) = 2 and q (x) = xe3x so we have p(x)dx = So the integrating factor is given by (x) = e
p(x)dx

2dx = 2x

= e2x

Substituting into the general solution formula we then have:

= =

1 (x) 1 e2x

q (x)(x)dx e2x xe3x dx xex dx

= e2x

= e2x (xex ex + C ) (use integration by parts to get this) = xe3x e3x + Ce2x = (x 1)e3x + Ce2x So we have the general solution:y = (x 1)e3x + Ce2x We now introduce a subtopic of dierential equations, exponential growth and decay. Exponential functions are very useful when it comes to modeling many real world phenomena. Any such model will involve a dierential equation of the form dy = ky (which is separable), we will provide an easier method dt of solving such an equation. Theorem 9.5. The solution to the dierential equation y = y0 at t = 0 is given by y = y0 ekt Example 8. Suppose it is known that the cells of a given bacterial culture divide every 3.5 hours. If there are 1000 cells in a dish to begin with, how many will there be after 12 hours? Solution: Recognizing that this is an exponential growth problem, we let y = the number of cells present at time =t and so y = y0 ekt (since this is an exponential growth problem), and y0 = 500 (this is given) The only variable left to determine is k , and since the information provided 8
dy dt

= ky , subject to

that the bacterial culture divides every 3.5 hours has not been used yet, it makes sense that this is what we should use to nd k , indeed: The number of cells doubles every 3.5 hours, so after 3.5 hours there should be 1000 cells 1000 = 500ek(3.5) 2 = ek(3.5) ln(2) = (3.5)k ln(2) 0.198042 3.5

k=

Substituting this into the solution formula y = y0 ekt we have: y = 500e(0.198042)t Therefore after 12 hours there wil be 500e(0.198042)12 5383 cells in the dish. Denition 9.6. The Half-life of a radioactive element is the time required for half of the radioactive nuclei present in a sample to decay. Amazingly enough the half-life of an element is totally independent of the number of nuclei present initially. This can be easily seen: Let y0 be the number of radioactive nuclei present initially, then the number y of nuclei present at time t will be given by: y = y0 ekt Since we are looking for half-life we wish to know when 1 y0 ekt = y0 2 cancelling y0 s we have: ekt = 9 1 2

1 kt = ln( ) 2 kt = ln(2) ln(2) k

t=

Thus the half-life depends only on k and not on y0 at all, the formula above is worth noting for future use: Half lif e =
ln(2) k

Example 9. The radioactive lifetime of plutonium-210 is extremely short and so is measured in days rather than years. The number of atoms remaining after t days with an initial amount of y0 radioactive atoms is given by: y = y0 e(4.9510 Find the half-life of plutonium-210. Solution: Using the formula above we have that Half lif e = =
ln(2) k ln(2) 4.95103
3 )t

140 Therefore the half-life of plutonium-210 is approximately 140 days. Example 10. Scientists who do carbon-14 dating use a value of 5700 years for its half-life. Find the age of an object that has been excavated and found to have 90 Solution: Using the equation y = y0 ekt we see that we must nd two things: (a) the value of k 90 9 (b) the value of t for which y0 ekt = 100 y0 , i.e. when ekt = 10

10

(a) Find the value of k: Using the half-life equation we have k = =

1.2 104 (b) the value of t which makes ekt = e(1.210

9 10

= 0.9 (1.2 104 )t = ln(0.9) t =

(ln of both sides)

878 years Therefore the sample is approximately 878 years old.

11