The Power of 𝝅

Isaac Mortiboy
The notion of Pi, or 𝜋, can be a somewhat unsettling one; I remember the first time I
was introduced to the concept of Pi, it felt like some irregularity with a dissatisfying answer. I
recall thinking about how much simpler it would be if it were instead an integer, as I’m sure
many student’s first thoughts are. But now, having learnt much more about its nature, I’ve
gained a much greater appreciation for Pi and how it mysteriously unifies many branching
areas of mathematics under one irrational value. I hope to try and cover some of the history
behind Pi which is often left out of the classroom, as well as take a look into what makes Pi
so special to us, and the role it plays in places like mechanics and complex numbers. But
first, let us start with where Pi came from and how we define it.

Most students first come across Pi when studying circles, which is how it was first
discovered, almost 4,000 years ago. However, the most suitable value for Pi has been a
long-lasting debate. Some people propose that the value of Pi should in fact be 6.283… or
twice the size of what we call Pi. This constant however has been universally accepted as
Tau or 𝜏, which has been popularised by Micheal Hartl’s manifesto and is obsessed over due
to how it simplifies several mathematical formulae. This is because the value of Tau in
radians represents a full 360°, meaning any angle in radians is just that fraction of Tau. Alas,
the popularity of Pi being 3.14… has stood the test of time, with its definition simply being
the ratio between the diameter and the circumference of a circle, which can be written as:

π d = 2πr = C
From this simple definition, we can derive many more equations for Pi. If we also apply a
little bit of calculus, we have the formula:
r
∫ 2πr dr = π r 2 = Area 1
0

These two equations make up the basis of where Pi is

derived from, but can be hard to interpret from only algebra. I
personally find that the best way to visualise this equation is by
splitting a circle up into sectors, and lining them up side by side,
alternating them up and down2. By doing this, we can see that, as
the angle of the segments tends to 0, they more closely form a
rectangle of height r and length 𝜋r (or C/2), which when multiplied,
give the desired result, 𝜋r2. Now with that covered, we’re left with
the question of how to calculate Pi.

In c.250 BC, a Greek man named Archimedes of Syracuse, famed for his levers,
screws and his eureka moment, used polygons to approximate the area of a unit circle3 - a
circle with a radius of one and an area of Pi. His approach was to first draw the largest

1
[f(x) = Area] ⇒ [f'(x) = Length]
2
"CircleArea" by Jim.belk - Own work. Licensed under Public Domain via Wikimedia Commons.
3
PBS - Approximating Pi By Rick Groleau

1
possible hexagon within the circle and calculate its area using the
Pythagorean formula (which was only around 150 years old at the
time!) Because the hexagon fit entirely in the circle, Archimedes knew
that the area of the hexagon would be less than the area of the circle.
He applied the same logic to the smallest possible square outside the
circle, which he deemed to be just bigger than Pi. Hence Archimedes
defined an upper and lower bound for Pi such that 3 < 𝜋 < 4.

Whilst this may initially sound very inaccurate, the beauty of Archimedes’ method is
that the more sides the polygon has, the closer the approximation to Pi (since the area of the
shape become increasingly circular.) Many mathematicians used this in order to gradually
calculate more and more digits of Pi, using polygons with greater numbers of sides.
Eventually, this amounted to the work of Ludolph van Ceulen, who published his book ‘Van
den Circkel’ in 1596 which stated an approximation for Pi using a polygon with 4.6 Quintillion
sides4! This however only yielded a value accurate to 35 decimal places which despite being
ample for any real-world applications, seems rather small compared to our present-day
calculations, giving Pi to 50 Trillion decimal places5. Nevertheless, Archimedes and Van
Ceulen’s work remains a staple in mathematical history.

Although Archimedes’ method lay the foundations for approximating Pi, it was
exceptionally inefficient; having to calculate the area of two polygons with several quintillion
sides to only generate 35 digits of Pi didn't quite meet the mark for many mathematicians.
Globally, some of history’s best thinkers took it upon themselves to try and find the best
possible approximation of Pi, with contributions from figures such as Ramanujan, Euler,
Leibniz and Madhava, all producing work that soon surpassed the Archemedian method6.
However, my focus lies with Sir Isaac Newton who, at only the age of 23, made significant
advancements to western mathematics, such as his work on the binomial theorem which he
used to generate a whole new basis for calculating Pi. Whilst he was not the first, he proves
a more accessible example of how mathematical methods progressed and brings in ideas
that I will extend upon later.

Newton’s Binomial theorem is a generalization for the expansion of the sum of two
unique terms, raised to some power7, such that:

n(n−1)a2 bn−2
(a + b)n = (0n)an b0 + (1n)an−1 b1 + ... = an + nabn−1 + 2!
+ ...
Initially, it's hard to make any meaningful connection between the binomial theorem,
and approximating Pi, but Newton was able to make the logical steps necessary to create a
whole new approach. The key into understanding his method lies in the equation for a unit
circle, which he rearranged as follows:

1/2
[x2 + y 2 = 1] ⇒ [y = (1 − x2 ) ]

4
Mathematical Treasure: Van Ceulen on the Circle
5
Calculating Pi: My attempt at breaking the Pi World Record
6
https://en.wikipedia.org/wiki/Approximations_of_π
7
Math Is Fun - Binomial Theorem

2
 By presenting it this way, we have an equation for the positive half of the circle with
an area of 𝜋/2, which can be applied to the binomial theorem to generate an infinite series:

1/2
(1 − x2 ) = 1 − 21 x2 − 81 x4 − 1 6
16
x ...
In my opinion, Newton's next step was equally astute, as he
used this knowledge in conjunction with his newly derived fundamental
theorem of calculus by integrating this series with respect to 1 and 0.
The result of this integral will be a quadrant (shaded grey in the
diagram) with an area of 𝜋/4. In this way, Newton generated the
following equation:

1 1
π 2 1/2
4 = ∫ (1 − x ) dx = ∫ [1 − 21 x2 − 81 x4 − 1 6
16 x ...]dx
0 0

And by using the power rule and some cancelling, we can simplify this expression to just:

1 x3 1 x5 1 x7 1 2 1 1 5
π = 4[x − 2
· 3 − 8
· 5 − 16
· 7 − ...]0 = 4 − 3 − 10 − 28 − 288 − ...
While it might not seem too special at face value, I find this equation to be
awe-inspiring! The fact that we are able to express an irrational value as the sum of a series
of rational numbers seems unbelievable. What Newton has achieved is proving that Pi can
be expressed as a converging infinite series, where each term gets increasingly smaller,
meaning it becomes very accurate very quickly. In fact, Newton went on to improve on this
formula, by deriving other infinite series with faster convergence to Pi8.

Nowadays, we have a plethora of ways of generating Pi proven by a range of great

mathematical minds. Not all evaluations are made equal, and while some series approach Pi
rapidly, they tend to do so at the cost of being more complex. Below are a few particularly
significant ones - especially the last two, whose variations are most notably used to generate
the figure I previously mentioned of 50 trillion digits:

∞ k
π (−1) 1 1 1 1
Leibniz Formula : 9
4 = ∑ 2k+1 =1− 3 + 5 − 7 + 9 ...
k=0

∞
1 2√2 (4k)!(1103+26390k)
Ramanujan Formula : 10
π = ∑
9801 (k!)4 3964k
k=0

∞ k
1 (−1) (6k)!(13591409+545140134k)
Chudnovsky Algorithm : 11
= 12 ∑
π (3k)!(k!)3 6403203k+3/2
k=0

8
The Discovery That Transformed Pi
9
https://en.wikipedia.org/wiki/Leibniz_formula_for_π
10
Ramanujan–Sato series
11
Chudnovsky algorithm

3
Today, the algorithms we use to determine Pi are the ones with the most optimal time
complexity, though calculating digits of Pi has now become a test of the power of
supercomputers, and does little in the way of helping us uncover some of the true mysteries
of this number.

Despite Pi’s strong connection to infinite series, its tendency to appear in unexpected
places can be seen in the Basel problem. The premise was first posed by Pietro Mengoli in
1650, who asked the question, ‘what is the value of the sum of the reciprocals of the squares
of natural numbers?’ This can be written algebraically as:

∞
1 1 1 1 1
∑ n2
= + + + ... = ?
12 22 32 42
n=1

This question remained long unanswered for almost a century despite its relatively
simple concept. Ultimately, the Basel problem’s solution was first presented in 1734 by
Leonard Euler, who was notoriously fascinated by puzzles and made large contributions to
the ways in which we use Pi today, such as affirming its ties to the field of complex numbers
(which I will come on to later). Euler's original method was based on rigorous algebra12,
making the solution rather obtuse, however, more modern approaches make it easier to see
how Pi falls into this problem’s solution. In particular, my favourite comes from Grant
Sanderson - better known as 3Blue1Brown, who contextualizes this infinite series as a set of
lighthouses, and uses their light intensity alongside the inverse Pythagorean theorem to
compute the solution. He explains his inventive approach excellently in his video13, which I’d
highly recommend along with the rest of his channel.

Either way you approach the problem, you end up with a value of 𝜋2/6, a seemingly
unlikely result, and certainly one which wouldn’t be expected when first considering the
problem. It is fascinating how such a regularly defined series generates the value of Pi in this
way, and once again furthers our appreciation for how impressive mathematics really can be.
However Euler’s obsession with Pi did not stop here; most people know Leonard Euler’s
name through his work on exponentials and the aptly named Euler’s number, whose defining
property is that:

d x
dx e = ex
Euler’s number shares similarities with Pi, in both its irrational value and ability to be
represented as an infinite series14, and the two are seamlessly combined in what is
renowned as the most beautiful and simple piece of mathematics - Euler’s equation:

eiπ + 1 = 0
The explanation to this equation lies within the complex plane, which is highlighted
through the presence of i, which is defined as the square root of -1. i can have a variety of
magnitudes, as with any real number, and is represented on an argand diagram, with a real
axis and an imaginary axis. This can be a challenging concept to grasp if you’ve never

12
An infinite series of surprises
13
3B1B’s - Why is pi here? And why is it squared? A geometric answer to the Basel problem
14
e = 1/1! + 1/2! + 1/3! + 1/4! ...

4
encountered it before, and for that reason I recommend another of 3B1B’s videos15 which
gives a more in-depth and visual representation of what this equation is truly saying.

By raising something to the power of i, for example, 1,

what we are doing is performing a series of multiplications by i,
initially setting its trajectory towards i, then continuing to rotate
our value on a vector 90° anticlockwise. This transformation
takes us around the complex plane in a circular motion about
the origin and immediately opens our eyes to why Pi might
appear in this scenario. To finish off this explanation, we bring
back in Euler’s number and its defining properties of
differentiation. Here, the derivative is used to describe the rate
at which we rotate around the unit circle, and since we are
using Euler’s number here, much like how the gradient of f(x) = ex at any point is ex, we are
travelling at a constant velocity. The last piece of the puzzle here is that if we travel a
distance of Pi radians around our circle, we finish at a value of -1, and if we add 1 to this we
spectacularly land at 0 - our desired result.

In my opinion, Euler’s equation is the perfect demonstration of how captivating the

beauty of maths is. In one equation we are able to combine 3 of math’s most perplexing and
irregular constants, as well as the two fundamental values of 1 and 0, into a single line of
algebra. Whilst I’ve tried to give credit to how amazing Pi really is, 2000 words simply don't
do it justice, so I’ve cited some more great explanations of places where Pi really shows its
colours. These range from its use in integrating the normal distribution curve16, to the
number of collisions between two blocks17, to how it can be found from simply dropping
needles18 and many more19! Although its roots lie within circles, Pi’s presence and sense of
mystery permeates through the rest of mathematics and is the reason why, on 3/14 every
year, I and many others come to celebrate it, on the occasion of Pi day.

16
Vcubingx’s - Why does pi show up here? | The Gaussian Integral, explained
17
3B1B’s - Why do colliding blocks compute pi?
18
Numberphile’s - Pi and Buffon's Matches - Numberphile
19
- Happy Pi Day!
- Understanding Euler's Formula | BetterExplained
- The Wallis product for pi, proved geometrically