The Product eπ Is Irrational

N. A. Carella

Abstract: This note shows that the product eπ of the natural base e and the circle
number π is an irrational number.
arXiv:1706.08394v7 [math.GM] 18 Jan 2023

Contents
1 Introduction 1

2 Foundation 2

3 Convergents Correlations 4

4 The Main Result 7

5 Algorithm And Numerical Data 8

1 Introduction
The number e = 2.718281828459 . . . was proved to be irrational by Euler, circa 1744. The
proof uses a differential equation to show that the continued fraction

e = [2; 2, 1, 2, 1, 1, 4, 1, 1, 6, 1, 1, 8, . . .] (1.1)

is infinite, see [6], [15, Theorem 3.10]. Later, a simpler proof based on the infinite series
X 1 1 1 1
e= = 1 + + + + ··· (1.2)
n! 2! 3! 4!
n≥0

was found by Fourier in 1815. Many versions of the Fourier classical proof are known for
er , where r ∈ Q is a rational number, and other numbers, see [13], [1, p. 35], [15]. The
number π = 3.141592653589 . . . was proved to be irrational by Lambert, circa 1760, see
[2, p. 129]. The proof uses the continued fraction of the tangent function tan(x), the fact
that the numbers tan(r) are irrationals for any nonzero rational number r ∈ Q, and the
value arctan(1) = π/4 to indirectly show that the continued fraction

π = [3; 7, 15, 1, 292, 1, 1, 1, 2, 1, 3, 1, 14, . . .] (1.3)

is infinite, see [2], [8], [11]. Later, simpler versions and new proofs were found by several
authors, [11], [1, p. 35], [15]. The above short compendium is a glimpse at the vast math-
ematical literature devoted to the analysis of the numbers e and π.

January 19, 2023

AMS MSC :Primary 11J72; Secondary 11A55.
Keywords: Irrational number; Natural base e; Circle number π.

1
The Product eπ Is Irrational 2

The arithmetic natures of the product e · π = 8.539734222673 . . ., and of the the sum e + π
are not known. In this note the known information on the continued fractions and the
convergents of the two irrational numbers e and π are used here to construct an infinite
subsequence of rational approximations for the product eπ.

Theorem 1.1. The product eπ is an irrational number.

The earlier sections cover the basic required background, and the proof of Theorem 1.1
appears in Section 4. An algorithm that illustrates the effectiveness of this result appears
in Section 5.

2 Foundation
Except for Theorem 2.4, all the materials covered in this section are standard results in
the literature, see [7], [9], [10], [14], [15], et alii.

A real number α ∈ R is called rational if α = a/b, where a, b ∈ Z are integers. Otherwise,

the number is irrational. The irrational numbers are further classified as algebraic if α
is the root of an irreducible polynomial f (x) ∈ Z[x] of degree deg(f ) > 1, otherwise it is
transcendental.

Theorem 2.1. If a real number α ∈ R is a rational number, then there exists a constant
c = c(α) such that
c p
≤ α− (2.1)
q q
holds for any rational fraction p/q 6= α. Specifically, c ≥ 1/b if α = a/b.

This is a statement about the lack of effective or good approximations for any arbitrary
rational number α ∈ Q by other rational numbers. On the other hand, irrational numbers
α ∈ R − Q have effective approximations by rational numbers. If the complementary
inequality |α − p/q| < c/q holds for infinitely many rational approximations p/q, then it
already shows that the real number α ∈ R is irrational, so it is sufficient to prove the
irrationality of real numbers.

Theorem 2.2 (Dirichlet). Suppose α ∈ R is an irrational number. Then there exists an

infinite sequence of rational numbers pn /qn satisfying

pn 1
0< α− < 2 (2.2)
qn qn

for all integers n ∈ N.

Theorem 2.3. Let α = [a0 , a1 , a2 , . . .] be the continued fraction of a real number, and let
{pn /qn : n ≥ 1} be the sequence of convergents. Then

pn 1
0< α− < (2.3)
qn an+1 qn2

for all integers n ∈ N.

The Product eπ Is Irrational 3

This is standard in the literature, the proof appears in [7, Theorem 171], [15, Corollary
3.7], and similar references.

A basic extension of the previous inequalities in Theorem 2.2 and Theorem 2.3 provided
here uses a pair of distinct irrational numbers and the corresponding parameters.
Theorem 2.4. Let α = [a0 , a1 , a2 , . . .] and β = [b0 , b1 , b2 , . . .] be distinct continued frac-
tions for two distinct irrational numbers α and β ∈ R such that αβ 6= ±1, respectively.
Then
pn um 2β 2α
0 < αβ − < 2
+ 2
, (2.4)
q n vm an+1 qn bm+1 vm
where {pn /qn : n ≥ 1} and {um /vm : m ≥ 1} are the sequences of convergents respectively.
Proof. By Theorem 2.3, there exists a sequence of convergents {pn /qn : n ∈ N} such that
pn 1
α− < , (2.5)
qn an+1 qn2
and the corresponding long form is
pn 1 pn 1
− <α< + . (2.6)
qn an+1 qn2 qn an+1 qn2
Similarly, there exists a sequence of convergents {um /vm : m ∈ N} such that
um 1
β− < 2
, (2.7)
vm bm+1 vm
and the corresponding long form is
um 1 um 1
− 2
<β< + 2
. (2.8)
vm bm+1 vm vm bm+1 vm
The product of the last two long forms returns

     
pn 1 um 1 pn 1 um 1
− − < αβ < + + . (2.9)
qn an+1 qn2 vm 2
bm+1 vm qn an+1 qn2 vm 2
bm+1 vm
Expanding these expressions produces
p n um 1 um 1 pn 1
− − +
qn vm an+1 qn2 vm 2 q
bm+1 vm n a 2 2
n+1 m+1 qn vm
b
< αβ (2.10)
p n um 1 um 1 pn 1
< + 2
+ 2
+ .
qn vm an+1 qn vm bm+1 vm qn an+1 bm+1 qn2 vm
2

The second order term 1/(an+1 bm+1 qn2 vm2 ) on the left side and right side is absorbed into

the larger first order terms on the right side. Thus, rearranging the inequality yield
1 um 1 pn p n um 2 um 2 pn
− − < αβ − < + . (2.11)
an+1 qn2 vm 2
bm+1 vm qn qn vm 2
an+1 qn vm 2
bm+1 vm qn
To complete the proof, rewrite it as a standard inequality
pn um 2 um 2 pn
0 < αβ − < 2
+ 2
, (2.12)
q n vm an+1 qn vm bm+1 vm qn
The Product eπ Is Irrational 4

and use the trivial upper bound

pn um
≤ 2α and ≤ 2β, (2.13)
qn vm

for all large integers n, m ≥ 1, confer (2.6) and (2.8).

For distinct irrationals α, β ∈ (0, 1), the simpler version

pn um 2 2
0 < αβ − < 2
+ 2
(2.14)
q n vm an+1 qn bm+1 vm

can be used to stream line the proof of a result such as Theorem 1.1.

Theorem 2.5 (Euler). The continued fraction e = [2, 12, 1, 1, 4, 1, 1, 6, 1, 1, 8, . . .] of the

natural base has unbounded quotients and the subsequence of convergents pn /qn satisfies
the inequality
pn 1
e− < . (2.15)
qn an+1 qn2
The quotients have the precise form

a0 = 2, a3k = a3k−2 = 1, a3k−1 = 2k, (2.16)

for k ≥ 1. The derivation appears in [12], [9, Theorem 2], [15, Theorem 3.10], [5], and
other.

3 Convergents Correlations
The correlation of a pair of convergents {pn /qn : n ≥ 1} and {um /vm : n ≥ 1} provides
information on the distribution of nearly equal values of the continuants qn and vm .

The regular pattern and unbounded properties of the partial quotients an = a3k−1 = 2k
of the continued fraction of e, see Theorem 2.5, are used here to generate a pair of infinite
subsequences of rational approximations {p3k−2 /q3k−2 : k ≥ 1} and {umk /vmk : k ≥ 1},
for which the product
p3k−2 umk
−→ eπ as k, mk −→ ∞. (3.1)
q3k−2 vmk

Furthermore, the values q3k−2 ≍ vmk are sufficiently correlated. The notation f (x) ≍ g(x)
is defined by g(x) ≪ f (x) ≪ g(x).

The recursive relations

p−1 = 1, p 0 = a0 , pn = an pn−1 + pn−2 ,

q−1 = 0, q0 = 1, qn = an qn−1 + qn−2 , (3.2)

for all n ≥ 1, see [7], [10], [14], are used to estimate the rate of growth of the subsequences
of continuants {qn : n ≥ 1} and {vm : m ≥ 1}.
The Product eπ Is Irrational 5

Lemma 3.1. Let e = [a0 , a1 , a2 , . . .] and π = [b0 , b1 , b2 , . . .] be the continued fractions of

this pair of irrational numbers. Let δ > 0 and ε > 0 be a pair of arbitrary small numbers.
Then, the followings hold.

(i) If bm = o(m), then the exists a pair of subsequences of convergents p3k−2 /q3k−2 and
umk /vmk such that

(2k)1−ε q3k−2 ≪ vmk ≪ 2(2k)1−ε q3k−2 . (3.3)

(ii) If bm = O(m), then the exists a pair of subsequences of convergents p3k−2 /q3k−2 and
umk /vmk such that

(2k)1−ε q3k−2 ≪ vmk ≪ 2(2k)1−ε q3k−2 . (3.4)

(iii) If bm = O(m1+δ ), then the exists a pair of subsequences of convergents p3k−2 /q3k−2
and umk /vmk such that

(mk )1−ε vmk ≪ qnk ≪ 2(mk )1−ε vmk . (3.5)

Proof. Case (i): The partial quotients bm = o(m) are bounded or unbounded. Make the
change of index n ≡ 1 mod 3 −→ k = (n + 2)/3 to focus on the subsequence of convergents
p3k−2 /q3k−2 of the number e as k → ∞, see Lemma ?? for more details. Observe that

.. .. ..
. . .
q3k−2 = q3k−3 + q3k−4 = q3k−2
q3k−1 = 2kq3k−2 + q3k−3 ≍ 22 kq3k−2
q3k = q3k−1 + q3k−2 ≍ 23 kq3k−2
q3(k+1)−2 = q3(k+1)−3 + q3(k+1)−4 ≍ 24 kq3k−2
(3.6)
q3(k+1)−1 = 2(k + 1)q3(k+1)−2 + q3(k+1)−3 ≍ 25 k(k + 1)q3k−2
q3(k+1) = q3(k+1)−1 + q3(k+1) ≍ 26 k(k + 1)q3k−2
q3(k+2)−2 = q3(k+2)−3 + q3(k+2)−4 ≍ 27 k(k + 1)q3k−2
q3(k+2)−1 = 2(k + 2)q3(k+1)−2 + q3(k+2)−3 ≍ 28 k(k + 1)(k + 2)q3k−2
.. .. ..
. . .
This verifies that these numbers has exponential rate of growth in k of the form

q3(k+t)−1 ≍ (4k)t+1 q3k−2 , (3.7)

for some t ≥ 0, as k → ∞. Next, consider the sequence of convergents um /vm of the

number π. By hypothesis, the partial quotients bm = o(m) are bounded or unbounded.
Furthermore, to simplify the notation, assume that bmk ≍ m1−δ k for infinitely many integers
mk = mk0 , mk1 , mk2 , mk3 , . . . ≥ 1. Then, this implies the existence of a subsequence of
convergents umk /vmk such that
.. .. ..
. . .
vmk = bmk vmk −1 + vmk −2 = vmk
vmk1 = bmk1 vmk1 −1 + vmk1 −2 ≍ (2k1 −k m1−δ
k )vmk
(3.8)
vmk2 = bmk2 vmk2 −1 + vmk2 −2 ≍ (2k 2 −k m1−δ 2
k ) vmk
1−δ
vmk3 = bmk3 vmk3 −1 + vmk3 −2 ≍ (2k3 −k mk )3 vmk
.. .. ..
. . .
The Product eπ Is Irrational 6

Moreover, the existence of a single value vmk > 1, see Tables 1 and 2, such that

(2k)1−ε q3k−2 ≪ vmk ≪ 2(2k)1−ε q3k−2 , (3.9)

implies the existence of an infinite subsequence of lower bounds

vmks = (2ks −k m1−δ s

k ) vmk (3.10)
≫ (2ks −k m1−δ s
k ) (2k)
1−ε
q3k−2
≫ (2k)1−ε q3(k+t)−2 ,

where
q3(k+t)−2 = (2ks −k m1−δ s
k ) q3k−2 , (3.11)
and use (3.7) to identify the relation

(2ks −k mk1−δ )s = o((4k)t+1 ) (3.12)

for some s ≥ 1 depending on t ≥ 1. The corresponding subsequence of upper bounds

satisfies

vmks = (2ks −k mk1−δ )s vmk (3.13)

≪ 2(2ks −k m1−δ s
k ) (2k)
1−ε
q3k−2
≪ 2(2k)1−ε q3(k+t)−2 .

Combining the last two inequalities yields the required relation

(2k)1−ε q3(k+t)−2 ≪ vmks ≪ 2(2k)1−ε q3(k+t)−2 (3.14)

for some s, t ≥ 1 as k → ∞.

Case (ii): The the partial quotients bm are bounded or unbounded, and bm = O(m). The
proof for this case is similar to Case (i).

Case (ii): The partial quotients bm are unbounded, and bm = O(m1+δ ). In this case, (3.14)
can fail, but since the inequality in Theorem 2.4 is symmetric in qn and vm , the proof is
almost the same as Case (i), but the subsequences of convergents are switched to obtain
the required relation

(2kt −k m1−δ
k )
t+1
vmkt ≪ qnk+s ≪ 2(2kt −k m1−δ
k )
t+1
vmkt (3.15)

for some s, t ≥ 1 as k → ∞. 

The distribution of all the continuants {qn : n ≥ 1} associated with a subset of continued
fractions of bounded partial quotients is the subject of Zeremba conjecture, see [4] for
advanced details. For any continued fraction, the numbers {qn : n ≥ 1} have exponential
growth  √ n
qn = an qn−1 + qn−2 ≥ (1 + 5)/2 , (3.16)

which√ is very sparse subsequence of integers. The least asymptotic growth occurs for the
(1+ 5)/2 = [1, 1, 1, . . .]. But the combined subset of continuants for a subset of continued
fractions of bounded partial fractions has positive density in the sunset of integers N =
{1, 2, 3, . . .}.
The Product eπ Is Irrational 7

4 The Main Result

Proof. (Theorem 1.1) Let e = [a0 , a1 , a2 , . . .] be the continued fraction of the irrational
number e. By Theorem 2.3, there exists a sequence of convergents {pn /qn : n ∈ N} such
that
pn 1
e− < . (4.1)
qn an+1 qn2
Similarly, let π = [b0 , b1 , b2 , . . .] be the continued fraction of the irrational number π, and
let {um /vm : m ∈ N} be the sequence of convergents such that
um 1
π− < 2
. (4.2)
vm bm+1 vm
Now, suppose that the product eπ = r/s ∈ Q is a rational number. Then
1 1 p n um 2π 2e
≤ eπ − < + . (4.3)
s q n vm q n vm an+1 qn2 2
bm+1 vm
The left side follows from Theorem 2.1, and the right side follows from Theorem 2.4.

Next, use the subsequence of unbounded partial quotients an = a3k−1 of the continued
fraction of e to generate an infinite subsequence of rational approximations
p3k−2 umk
−→ eπ as k, mk −→ ∞. (4.4)
q3k−2 vmk
The subsequence of rational approximations is generated by the following algorithm.

Fix an arbitrary small number ε > 0.

1. Input: The integer n ≡ 1 mod 3.
2. Let k = (n + 2)/3.
3. Let an+1 = a3k−1 = 2k, and fix the convergent pn /qn = p3k−2 /q3k−2 of e, see
Theorem 2.5.
4. Choose a convergent umk /vmk of π in the range
(2k)1−ε q3k−2 ≤ vmk ≤ 2(2k)1−ε q3k−2 . (4.5)

5. Output: The quotient a3k−1 = 2k and the pair of convergents

p3k−2 umk
and . (4.6)
q3k−2 vmk

Various versions of this algorithm are possible, for example, by modifying the interval in
(4.5).

Replacing the subsequence of rational approximations constructed in (4.5) into (4.3) yields
1 1 p3k−2 umk
1−ε 2 ≤ eπ − (4.7)
s 2(2k) q3k−2 q3k−2 vmk
2π 2e
≤ 2 + 2(1−ε) 2
2kq3k−2 (2k) bmk +1 q3k−1
π 2e
≤ 2 + 2
.
kq3k−2 (2k)2(1−ε) bmk +1 q3k−2
The Product eπ Is Irrational 8

Multiplying (4.7) by (2k)1−ε q3k−2

2 and using bmk +1 ≥ 1 returns

1 21−ε π 2e
≤ ε
+ 1−ε
(4.8)
2s k (2k) bmk +1
1−ε
2 π 2e
≤ + .
k ε (2k)1−ε

Since eπ = r/s is rational constant, and s ≥ 1, it is clear that the inequality (4.8) is a
contradiction for infinitely many large rational approximations
p3k−2 umk
(4.9)
q3k−2 vmk

as k, mk → ∞. Ergo, the product eπ is not a rational number. 

The structure of the proof, in equations (4.3), and (4.7), is similar to some standard proofs
of irrational numbers. Among these well known proofs are the Fourier proof of the irra-
tionality of e, see [1, p. 35], the proofs for ζ(2), and ζ(3) in [3], et alii.

An algorithm and sample of numerical data is compiled in Section 5 to demonstrate the

practicality of this technique.

5 Algorithm And Numerical Data

For each fixed pair of index (n, m), the basic product inequality

pn um 2π 2e
0 < eπ − < 2
+ 2
(5.1)
qn vm an qn bm vm

is used to test each rational approximation. It is quite easy to find the pairs (n, m) to
construct a subsequence of rational approximations
p n um
→ eπ (5.2)
qn vm
as n, m → ∞. The subsequence of rational approximations is generated by the following
algorithm.

Algorithm 1. Fix an arbitrary small number ε > 0.

1. Input an integer n ≡ 1 mod 3.

2. Let k = (n + 2)/3.

3. Let an+1 = a3k−1 = 2k, and fix the convergent pn /qn = p3k−2 /q3k−2 of the natural
base e.

4. Choose a convergent umk /vmk of π such that

(2k)1−ε q3k−2 ≤ vmk ≤ 2(2k)1−ε q3k−2 . (5.3)

5. If Step 4 fail, then increment n ≡ 1 mod 3, and repeat Step 2, the existence is proved
in Lemma 3.1, see also Remark 5.1.
The Product eπ Is Irrational 9

(n, m) = (19,10)
(an , bm ) = (12, 1)
pn = 13580623
qn = 4996032
um = 5419351
vm = 1725033
p n um
eπ − qn vm = 0.00000000000012256862192
1 um 1 pn 1
2 v
an qn m
+ 2 q
bm vm n
+ an+1 bm+1 qn2 v2 = 0.000000000000136378880
m

Table 1: A Rational Approximation Of The Product eπ.

(n, m) = (31,21)
(an , bm ) = (20, 2)
pn = 22526049624551
qn = 8286870547680
um = 3587785776203
vm = 1142027682075
p n um
eπ − qn vm = 8.32849575322710174432272 × 10−25
1 um 1 pn 1
2 vm
an qn
+ 2
bm vm qn + an+1 bm+1 qn2 v2 = 1.04439176914510045201022 × 10−24
m

Table 2: A Rational Approximation Of The Product eπ.

6. Output a3k−1 = 2k, p3k−2 /q3k−2 and umk /vmk .

The parameters for two small but very accurate approximations are listed in the Tables 1
and 2. These examples demonstrate the practicality of the algorithm.

Remark 5.1. Step 5 makes the algorithm independent of the rate of growth of the partial
quotients of the number π. Various versions of this algorithm are possible, for example,
by modifying the interval in (5.3).
The Product eπ Is Irrational 10

References
[1] Aigner, Martin; Ziegler, Gunter M. Proofs from The Book. Fifth edition. Springer-
Verlag, Berlin, 2014. 1, 8

[2] Berggren, Lennart; Borwein, Jonathan; Borwein, Peter. Pi: a source book. Third
edition. Springer-Verlag, New York, 2004. 1

[3] Beukers, F. A note on the irrationality of ζ(2) and ζ(3). Bull. London Math. Soc.
11 (1979), no. 3, 268-272. 8

[4] Jean Bourgain, Alex Kontorovich. On Zaremba’s Conjecture. arXiv:1107.3776. 6

[5] Cohn, Henry. A short proof of the simple continued fraction expansion of e. Amer.
Math. Monthly 113 (2006), no. 1, 57-62. 4

[6] Leonard Euler, De Fractionibus Continuis, Desertation, 1744. 1

[7] Hardy, G. H.; Wright, E. M. An introduction to the theory of numbers. Sixth edition.
Oxford University Press, Oxford, 2008. 2, 3, 4

[8] M. Laczkovich. On Lambert’s Proof of the Irrationality of π. American Mathemat-

ical Monthly, Vol. 104, No. 5 (May, 1997), pp. 439-443.

[9] Lang, Serge. Introduction to Diophantine approximations. Second edition. Springer-

Verlag, New York, 1995. 1

[10] Niven, Ivan; Zuckerman, Herbert S.; Montgomery, Hug h L. An introduction to the
theory of numbers. Fifth edition. John Wiley & Sons, Inc., New York, 1991. 2, 4

[11] Niven, Ivan. A simple proof that π is irrational. Bulletin of the American Mathe-
matical Society, 53 (6), p. 509, 1947. 2, 4

[12] Olds, C. D. The Simple Continued Fraction Expansion of e . Amer. Math. Monthly
77 (1970), no. 9, 968-974. 1

[13] L. L. Pennisi. Elementary Proof that e is Irrational. The American Mathematical

Monthly, Vol. 60, No. 7 (Aug. - Sep., 1953), p. 474. 4

[14] Rose, H. E. A course in number theory. Second edition. Oxford Science Publications.
The Clarendon Press, Oxford University Press, New York, 1994. 1

[15] Steuding, Jorn. Diophantine analysis. Discrete Mathematics and its Applications.
Chapman- Hall/CRC, Boca Raton, FL, 2005. 2, 4
1, 2, 3, 4

IrrationalProducts-01-18-23-23-arxiv.tex.