maths.

org/step

STEP Support Programme

Pure STEP 3 Questions

2012 S3 Q6
1 Preparation
The STEP question involves complex numbers and the Argand diagram. However, after you
have put z = x + iy in the given equations and set the real and imaginary parts (separately)
equal to zero, there are no more complex numbers left, and the Argand diagram is just the
x-y plane.
In the first part, the locus given by the second equation (for which y = 0) is just the x-axis
or part of it. Rather surprisingly, we are told to obtain p in terms of x, not x in terms of
p. The question we are trying to answer is ‘What values of x are allowed’, in other words
which values of x correspond to some value of p (it has to be a real value of p since that is
what we were told and that is what we assumed when we took real and imaginary parts of
the given equations). The equation
1
−p = x +
x
shows that, for any value of x (except x = 0), there is a corresponding value of p. So the
required locus is the whole of the x-axis except for x = 0.
It may be best to plunge straight into the STEP question without any preparation. But some
of the sketches (especially for last part) are quite hard, so you might like to start with:

(i) Find the centre and radius of the circle x2 + y 2 + 2x = 0.

(ii) For which values of x is x4 − x non-negative (i.e. greater than or equal to zero)?
1
For which values of x is −x2 − non-negative?
x

√
(iii) Sketch the graphs y = x − 1, y = x − 1, and y 2 = x − 1, paying attention to the
gradient of the second and third graphs at x = 1.
Note that in the second graph y > 0, but in the third y can be negative (though we
must have x − 1 > 0).

Pure STEP 3 Questions 1

 maths.org/step

2 The STEP 3 question

(i) Let x + iy be a root of the quadratic equation z 2 + pz + 1 = 0, where p is a real number.

Show that x2 − y 2 + px + 1 = 0 and (2x + p)y = 0. Show further that

either p = −2x or p = −(x2 + 1)/x with x 6= 0 .

Hence show that the set of points in the Argand diagram that can (as p varies) represent
roots of the quadratic equation consists of the real axis with one point missing and a
circle. This set of points is called the root locus of the quadratic equation.

(ii) Obtain and sketch in the Argand diagram the root locus of the equation

pz 2 + z + 1 = 0 .

(iii) Obtain and sketch in the Argand diagram the root locus of the equation

pz 2 + p2 z + 2 = 0 .

Pure STEP 3 Questions 2

 maths.org/step

2001 S3 Q1
3 Preparation

(i) Use the chain rule to differentiate sin (ln x). Use this to help you differentiate e2 sin(ln x) .

d n
(ii) Use proof by induction to show that (x ) = nxn−1 .
dx
You can show the base case n = 1 by first principles, or even just by sketching the line.
It will help to note that x(k+1) = x × xk .

(iii) Maclaurin’s expansion is:

x 0 x2 x3
f(x) = f(0) + f (0) + f 00 (0) + f 000 (0) + · · · ,
1! 2! 3!
(assuming this series converges).
Use this to find the first three non-zero terms in the expansion of sin x. Hence estimate
sin 0.1 to seven decimal places, where 0.1 is in radians.
If 0.1 is in radians (not degrees), then we can use the result that the derivative of
sin x is cos x. You may well want to do the final calculation without a calculator, just
because you can.

4 The STEP 3 question

√ dy 1
Given that y = ln(x + x2 + 1), show that =√ .
dx 2
x +1
Prove by induction that, for n > 0 ,

x2 + 1 y (n+2) + (2n + 1) xy (n+1) + n2 y (n) = 0 ,

dn y
where y (n) = and y (0) = y .
dxn
Using this result in the case x = 0 , or otherwise, show that the Maclaurin series for y begins

x3 3x5
x− +
6 40
and find the next non-zero term.

Discussion Note that the induction requires you to prove the result for n > 0, so make your ‘base
case’ n = 0. Don’t overlook the request tucked away at the end of the question (to find the next
non-zero term)!

Pure STEP 3 Questions 3

 maths.org/step

2009 S3 Q8
5 Preparation

(i) Although you probably won’t study limits rigorously until you get to university, there
are various properties of limits that you can understand and use without formal defi-
nitions.
For example, if lim f(x) = L and lim g(x) = K (where L and K are finite) then:
x→a x→a

i. lim cf(x) = c lim f(x) = cL

x→a x→a


ii. lim f(x) + g(x) = lim f(x) + lim g(x) = L + K
x→a x→a x→a


iii. lim f(x)g(x) = lim f(x) × lim g(x) = L × K
x→a x→a x→a

f(x) lim f(x) L

iv. lim = x→a = (K 6= 0).
x→a g(x) lim g(x) K
x→a



v. If h(x) is continuous, then lim h g(x) = h lim g(x) = h(K) .
x→a x→a

These all make good sense, and you would probably have assumed them to be true.
Evaluate the following limits, stating which of the above properties you have used:
sin θ
(a) lim
θ→0 cos θ
(b) lim 5(e−x + 1)
x→∞

(c) lim cos 1t

t→∞
1
(d) lim e x
x→∞

t
(ii) By first writing e−t t as and then using the series expansion for et show that:
et
lim e−t t = 0 .
t→∞

There are various ways you can do this. You could, for example, start by dividing top
and bottom by t. Note that you cannot use rule iii. above as the limit of t is not finite.

Show that lim e−t tn = 0 where n is a positive integer.

t→∞

(iii) If x = e−t , how does x behave as t → ∞? By setting substitution x = e−t in the

expression x ln x, show that lim (x ln x) = 0.
x→0

Pure STEP 3 Questions 4

 maths.org/step

Z 1
(iv) Consider the integral ln x dx. The function ln x is not defined when x = 0, but the
0
integral can be defined as a limit, as follows.
Z 1
(a) Evaluate ln x dx (see Assignment 24 if you are stuck).
r

(b) Now take the limit of this result as r → 0.

6 The STEP 3 question

Let m be a positive integer and let n be a non-negative integer. Use the result lim e−mt tn = 0
t→∞
to show that
lim xm (ln x)n = 0 .
x→0

(i) By writing xx as ex ln x show that

lim xx = 1 .
x→0

Z 1
(ii) Let In = xm (ln x)n dx . Show that
0

n+1
In+1 = − In
m+1
and hence evaluate In .

(iii) Show that

Z 1
1 2 1 3 1 4
xx dx = 1 −




2 + 3 − 4 + ··· .
0

Discussion

Who would have thought the final integral would come out as such an interesting-looking series?

A small point: towards the end of part (iii) you have to integrate an infinite series, which you do
by integrating each term and adding them all up. You have used the result that the integral of
a sum is equal to the sum of the integrals, which certainly holds for the sum of a finite number
of terms. However, it may not hold for an infinite number of terms; it depends on how well the
series converges (as you will discover in your first university year). We are OK here, because the
exponential series converges splendidly well, and you weren’t intended to worry about it.

Pure STEP 3 Questions 5

 maths.org/step

2006 S3 Q7
7 Preparation
Some remarks on hyperbolic functions

• Hyperbolic functions are defined in terms of exponentials (look it up if you don’t know
these definitions), in the same way as trigonometric functions could be defined in terms
of complex exponentials (eiθ ).
• All the formulae for trigonometric functions work similarly for hyperbolic functions
provided you remember to change the sign in front of any product of two sinh functions.
For example cosh2 θ − sinh2 θ = 1. This is not magic: it follows immediately from the
identities sinh x = −i sin(ix) and cosh x = cos(ix) which themselves follow directly
from the definitions (try it if you haven’t already done so). This “rule” is known as
“Osbourn’s rule”, and first appeared in 1902 as “Mnemonic for hyperbolic functions”
in the Mathematical Gazette.
• Although trigonometric and hyperbolic functions are algebraically related as above, they
are very distinct geometrically. For example, cosh x > 1 if x is real; and the equation
sinh x = 0 has only one real solution.

The hyperbolic functions sinh x (pronounced “sinch or shine”) and cosh x are defined by:

ex − e−x ex + e−x
sinh x = and cosh x = .
2 2
d d
(i) Show that (cosh x) = sinh x and find (sinh x).
dx dx

(ii) Sketch cosh x and sinh x.

Note that when x → +∞, we can ignore the e−x term.

(iii) Simplify cosh x + sinh x and cosh x − sinh x.

(iv) Simplify (cosh x)2 − (sinh x)2 .

dy
(v) Solve the equation = sinh x, where y = 2 when x = 0.
dx

dy sinh y
(vi) Solve the equation = tanh y, where tanh y = and is pronounced “tanch” or
dx cosh y
“th-an”. Write your answer in the form sinh y = · · · .

Pure STEP 3 Questions 6

 maths.org/step

8 The STEP 3 question

(i) Solve the equation u2 + 2u sinh x − 1 = 0 giving u in terms of x.

Find the solution of the differential equation
 2
dy dy
+2 sinh x − 1 = 0
dx dx

dy
that satisfies y = 0 and > 0 at x = 0.
dx

(ii) Find the solution, not identically zero, of the differential equation
 2
dy dy
sinh y +2 − sinh y = 0
dx dx

that satisfies y = 0 at x = 0, expressing your solution in the form cosh y = f(x). Show
that the asymptotes to the solution curve are y = ±(−x + ln 4).

Discussion

Asymptotes are lines that the curve approaches as either x or y or both tend to infinity. As some
terms
√ √ big, other terms (such as constant terms) can be ignored. For example, as x → +∞,
get very
x2 + 1 ≈ x2 = x (since x → +∞ we have x > 0).

Pure STEP 3 Questions 7

 maths.org/step

1988 S2 Q5
9 Preparation

(i) (a) Prove by induction that (cos θ + i sin θ)n = cos nθ + i sin nθ.
(b) Find all solutions of the equation z 5 = 1, by writing z as r(cos θ + i sin θ), where
r > 0. You should justify each step carefully.
(c) Express cos 5θ in terms of cos θ and sin θ.

(ii) If the roots of the equation xn + a1 xn−1 + · · · + an−1 x + an = 0, are α1 , α2 , . . . , αn ,

what is the expression for the product of the roots in terms of the coefficients ai ?
It just involves one coefficient, and depends on n. If you haven’t seen this before,
factorise the equation in the form (x − α1 ) · · · (x − αn ) = 0

10 The STEP 3 question

By considering the imaginary part of the equation z 7 = 1, or otherwise, find all the roots of
the equation
t6 − 21t4 + 35t2 − 7 = 0.
You should justify each step carefully.
Hence, or otherwise, prove that
2π 4π 6π √
tan tan tan = 7.
7 7 7
Find the corresponding result for

2π 4π (n − 1)π
tan tan · · · tan
n n n
in the two cases n = 9 and n = 11.

This is a 30 year old STEP II question, included here as it involves roots of unity which are now
on the STEP 3 specification (rather than STEP 2).

Pure STEP 3 Questions 8

 maths.org/step

2013 S3 Q3
11 Preparation
In geometry the centroid of a set of points is, roughly speaking, the average position position
of the points. If particles of equal mass were placed at each point, the centroid would coincide
with their centre of mass (or centre of gravity). In vector language, if the points have position
vectors p1 , p2 , . . . , pn , then the centroid is at (p1 + p2 + · · · + pn )/n.
The three vertices P1 , P2 and P3 of an equilateral triangle lie on a circle of radius 1, and
their position vectors with respect to an origin at the centre of the circle are p1 , p2 and p3 .

(i) Explain without calculation, using a symmetry argument, why p1 + p2 + p3 = 0.

This takes a bit of thought; or, rather, a bit of thought to come up with a careful
explanation. Even if you are not sure about this part, don’t let that stop you from
continuing to the rest of the question.

(ii) Evaluate p1 .p1 , (p1 + p2 + p3 ).p1 , p2 .p1 + p3 .p1 and p2 .p1 .

(iii) If the vector x has length k, show that (x − p1 ).(x − p1 ) = k 2 − 2x.p1 + 1 and find
3
X
(x − pi ).(x − pi ).
i=1

(iv) Given that P1 has coordinates (0, 1), use the values of p2 .p1 and p2 .p2 to find the
coordinates of P2 and P3 .

Pure STEP 3 Questions 9

 maths.org/step

12 The STEP 3 question

The four vertices Pi (i = 1, 2, 3, 4) of a regular tetrahedron lie on the surface of a sphere with
centre at O and of radius 1. The position vector of Pi with respect to O is pi (i = 1, 2, 3, 4).
Use the fact that p1 + p2 + p3 + p4 = 0 to show that pi . pj = − 13 for i 6= j.
Let X be any point on the surface of the sphere, and let XPi denote the length of the line
joining X and Pi (i = 1, 2, 3, 4).

(i) By writing (XPi )2 as (pi − x) . (pi − x), where x is the position vector of X with
respect to O, show that
X4
(XPi )2 = 8 .
i=1

(ii) Given that P1 has coordinates (0, 0, 1)√and that the coordinates of P2 are of the form
(a, 0, b), where a > 0, show that a = 2 2/3 and b = −1/3, and find the coordinates of
P3 and P4 .

(iii) Show that

4
X 4
X
4
(XPi ) = 4 (1 − x . pi )2 .
i=1 i=1
4
(XPi )4 is independent
P
By letting the coordinates of X be (x, y, z), show further that
i=1
of the position of X.

Discussion

The very last part is a bit of slog; sorry about that!

4
X
It is probably easiest to expand and simplify (1 − x . pi )2 before putting in the coordinates of
i=1
x and pi .
 √ √ 2
You should find that the worst thing you have to calculate is − 32 x ± √2 y
3
− 13 z .

Pure STEP 3 Questions 10

 maths.org/step

2008 S3 Q4
13 Preparation

(i) Sketch the graph y = tanh x.

sinh x
You can differentiate to find the gradient. You can also express tanh x in terms
x
cosh x
of e which will help when explaining the behaviour as x → ±∞.

r
cosh x − 1
(ii) Show that, for x > 0, = tanh(x/2). What is the corresponding result
cosh x + 1
if x 6 0?

d
(iii) By considering cosh y = x, find arcoshx in terms of x.
dx

Z
x
(iv) By using the substitution sinh y = x, find √ dx and check your answer by
x2+1
direct integration.

(v) Give a sketch to show that if f(t) > g(t) for t > 0 and f(0) = g(0) then
Z x Z x
f(t)dt > g(t)dt
0 0

and use this result to show that 21 x2 > 1 − cos x for x > 0.

14 The STEP 3 question

(i) Show, with the aid of a sketch, that y > tanh(y/2) for y > 0 and deduce that
x−1
arcoshx > √ for x > 1. (∗)
x2 − 1

x−1
(ii) By integrating (∗), show that arcoshx > 2 √ for x > 1.
x2 − 1

√
x2 − 1
(iii) Show that arcoshx > 3 for x > 1.
x+2

[Note: arcoshx is another notation for cosh−1 x. ]

Pure STEP 3 Questions 11

 maths.org/step

Discussion

This is a very nice question (where nice is not equivalent to easy), you obtain the first inequality to
obtain the second inequality, and the second to obtain the third, and obviously you could keep going,
though the denominators get nasty. You are effectively pulling yourself up by your bootstraps.

And just when you feel you can’t face another integration, you find that you have already done
everything required for part (iii).

But it is not a perfect question, because in fact the inequality of part (ii) could have been obtained
from a sketch similar to that of part (i), showing y/2 and tanh(y/2).

Pure STEP 3 Questions 12

 maths.org/step

2013 S3 Q7
15 Preparation

(i) If f(0) = 0 and f 0 (x) > 0 for x > 0 what can you say about f(x)? (Be careful to make
an accurate statement!)

(ii) Differentiate with respect to x:

(a) esin x ;
(b) cosh(x2 )
(c) sinh(f(x)), where f is a given function with derivative f 0 ;
(d) xf 0 (x);
(e) (f 0 (x))2 .

(iii) Integrate Ax sin x + B cos x with respect to x;

(iv) Integrate f(x)ef(x) f 0 (x) with respect to x.

(v) By writing the hyperbolic functions in terms of exponentials, show that cosh x > 1.

Pure STEP 3 Questions 13

 maths.org/step

16 The STEP 3 question

d2 y dy
(i) Let y(x) be a solution of the differential equation 2
+ y 3 = 0 with y = 1 and =0
dx dx
at x = 0, and let
 2
dy
E(x) = + 12 y 4 .
dx
Show by differentiation that E is constant and deduce that |y(x)| 6 1 for all x.

d2 v dv
(ii) Let v(x) be a solution of the differential equation 2
+ x + sinh v = 0 with v = ln 3
dx dx
dv
and = 0 at x = 0, and let
dx
 2
dv
E(x) = + 2 cosh v .
dx

dE 5
Show that 6 0 for x > 0 and deduce that cosh v(x) 6 3 for x > 0.
dx

(iii) Let w(x) be a solution of the differential equation

d2 w dw
2
+ (5 cosh x − 4 sinh x − 3) + (w cosh w + 2 sinh w) = 0
dx dx
dw 1 5
with = √ and w = 0 at x = 0. Show that cosh w(x) 6 4 for x > 0.
dx 2

Discussion

Don’t attempt to solve any of the differential equations in this question! They are all non-linear
(i.e. they depend on y 2 or worse), and in general such equations can only be solved numerically by
computer.

Even the equation in part (i), which doesn’t look too bad, can only be solved in terms of what are
called elliptic functions.

This is not an easy question, not because any one step is hard, but because there are lots of steps
each of which needs an idea which may well be unfamiliar.

Pure STEP 3 Questions 14