A Universal Identity for Powers in Quadratic Algebras and a Matrix Derivation of a Fibonacci Identity

Marco Mantovanelli · 2026

Read A Universal Identity for Powers in Quadratic Algebras and a Matrix Derivation of a Fibonacci Identity by Marco Mantovanelli (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-19343.txt.

Introduction

Mantovanelli unifies several classic Fibonacci and Lucas identities under one algebraic trick. Take an element whose square is a linear mix of itself and the unit. Then each power of it shrinks to a first-degree form with set coefficients. Those coefficients are the generic Lucas sequence, also known as Dickson polynomials. The same rule, via Cayley-Hamilton, gives power formulas for two-by-two matrices. Applying test functionals to the Fibonacci matrix yields multi-index identities in one sweep. The note is expository and stresses the shared mechanism.

Catalog record

CREATORMarco Mantovanelli
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE5548
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.19343

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A Universal Identity for Powers in Quadratic Algebras and a Matrix Derivation of a Fibonacci Identity
Marco Mantovanelli

Abstract
We prove a universal identity for powers of elements in quadratic algebras, expressing x m x^{m} in terms of x x and the identity. As a consequence, we obtain a universal formula for powers of 2 × 2 2\times 2 matrices depending only on trace and determinant. Applying this to the Fibonacci matrix yields a binomial expansion formula for F n ​ m F_{nm} , recovering a recent identity of Vorobtsov. This shows that such identities are consequences of general algebraic principles rather than specific properties of Fibonacci numbers.



Keywords:
quadratic algebras; Cayley–Hamilton theorem; matrix powers; Fibonacci numbers; Chebyshev polynomials

MSC Classification:
15A24, 11B39, 05A19, 15A15

1
Introduction

Identities involving Fibonacci and Lucas numbers (see, e.g.,
[
2
,
6
]
) often exhibit binomial structures
and connections to Chebyshev polynomials (see, e.g.,
[
3
]
) and second-order recurrences. A recent result
of Vorobtsov
[
7
]
provides an explicit expansion of
F_{nm}
as a polynomial
in
L_{n}
with coefficients given by binomial sums.

From a structural point of view, such identities reflect a more general phenomenon:
whenever an element satisfies a quadratic relation, all powers of
x
lie in the
R
-submodule generated by
\{1,x\}
. In particular, Fibonacci and Lucas numbers arise from the same
second-order recurrence and admit representations in terms of Chebyshev polynomials,

F_{nm}=F_{n}\,U_{m-1}\!\left(\frac{L_{n}}{2}\right).

Another classical perspective is provided by matrix methods (see, e.g.,
[
1
,
5
]
). The Fibonacci sequence
is generated by powers of a
2\times 2
matrix, and more generally, powers of such matrices
are governed by the Cayley–Hamilton theorem. The coefficients in these expressions depend
only on the trace and determinant, which are invariants under similarity.

The Fibonacci identity considered here may be viewed as a specialization of
classical formulas for powers of
2\times 2
matrices expressed in terms of
trace and determinant; see, for example,
[
4
]
. Closely related
formulations also arise in the theory of second-order linear recurrences and matrix powers.

The present formulation highlights a universal quadratic algebra structure and provides
a direct conceptual derivation of Vorobtsov’s identity
[
7
]
as a corollary.

2
Universal quadratic identity

Throughout, we identify the ring
R
with the scalar subalgebra
R\cdot 1

of the given
R
-algebra.

Theorem 1

(Universal quadratic reduction)
.

Let
R
be a commutative ring with identity, and let
x
be an element of an

R
-algebra satisfying

x^{2}-tx+d=0

for some
t,d\in R
. Define polynomials
P_{m}(t,d)
by

P_{0}(t,d)=0,\qquad P_{1}(t,d)=1,

and

P_{m+1}(t,d)=tP_{m}(t,d)-dP_{m-1}(t,d)\qquad(m\geq 1).

Then for every integer
m\geq 1
,

x^{m}=P_{m}(t,d)\,x-d\,P_{m-1}(t,d).

Moreover,

P_{m}(t,d)=\sum_{i=0}^{\lfloor(m-1)/2\rfloor}\binom{m-1-i}{i}t^{m-1-2i}(-d)^{i}.

Proof.

From the relation
x^{2}=tx-d
, it follows inductively that every power
x^{m}

can be written in the form

x^{m}=a_{m}x+b_{m}

for suitable coefficients
a_{m},b_{m}\in R
. Multiplying by
x
gives

x^{m+1}=a_{m}x^{2}+b_{m}x=a_{m}(tx-d)+b_{m}x=(ta_{m}+b_{m})x-da_{m}.

Thus the coefficients satisfy

a_{m+1}=ta_{m}+b_{m},\qquad b_{m+1}=-da_{m}.

Eliminating
b_{m}
yields

a_{m+1}=ta_{m}-da_{m-1},\qquad a_{0}=0,\ a_{1}=1.

Hence
a_{m}=P_{m}(t,d)
, and
b_{m}=-dP_{m-1}(t,d)
, giving the claimed formula.

The explicit expression for
P_{m}
follows by induction on
m
, verifying that
the binomial sum satisfies the same recurrence and initial conditions.
∎

Proposition 1
.

For each integer
m\geq 1
, there exists a polynomial
P_{m}(T,D)\in\mathbb{Z}[T,D]

such that for every commutative ring
R
and every element
x
of an
R
-algebra
satisfying
x^{2}-Tx+D=0
, one has

x^{m}=P_{m}(T,D)x-DP_{m-1}(T,D),

where
P_{m}
is given by the same recurrence and initial conditions as above.

3
Matrix formulation

Corollary 1
.

Let
M\in M_{2}(R)
, where
R
is a commutative ring, and let

t=\operatorname{tr}(M),\qquad d=\det(M).

Then for all integers
m\geq 1
,

M^{m}=P_{m}(t,d)\,M-d\,P_{m-1}(t,d)\,I.

Proof.

By the Cayley–Hamilton theorem,
M
satisfies

M^{2}-tM+dI=0.

The result follows by applying the theorem to
x=M
.
∎

4
Remark on Chebyshev polynomials

The polynomials
P_{m}(t,d)
are closely related to Chebyshev (or Dickson)
polynomials. If
\sqrt{d}
exists in a suitable extension of
R
, then

P_{m}(t,d)=d^{(m-1)/2}\,U_{m-1}\!\left(\frac{t}{2\sqrt{d}}\right).

Thus the above identity recovers and generalizes the classical Chebyshev representation
of matrix powers.

5
Application to Fibonacci numbers

Consider the Fibonacci matrix

A=\begin{pmatrix}1&1\\
1&0\end{pmatrix}.

Then

A^{n}=\begin{pmatrix}F_{n+1}&F_{n}\\
F_{n}&F_{n-1}\end{pmatrix},

so that

\operatorname{tr}(A^{n})=L_{n},\qquad\det(A^{n})=(-1)^{n}.

Let
M=A^{n}
. Then

A^{nm}=M^{m}.

Taking the
(1,2)
entry gives

F_{nm}=(M^{m})_{12}.

Since
(I)_{12}=0
, we obtain:

Corollary 2
.

For all integers
m,n\geq 1
,

F_{nm}=F_{n}\sum_{i=0}^{\lfloor(m-1)/2\rfloor}\binom{m-1-i}{i}L_{n}^{m-1-2i}(-1)^{i(n+1)}.

This recovers the identity of Vorobtsov
[
7
]
.

6
Conclusion

We have shown that identities expressing
F_{nm}
in terms of
L_{n}
and binomial
coefficients arise from a universal reduction formula in quadratic algebras.
The Fibonacci case is thus a specialization of a general algebraic principle

Read the complete text