<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1 plus MathML 2.0//EN" "http://www.w3.org/Math/DTD/mathml2/xhtml-math11-f.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xmlns:xlink="http://www.w3.org/1999/xlink" xml:lang="en">
<head>
<title>Quotients of C*-algebras</title>
<link rel="stylesheet" type="text/css" href="http://www.ms.u-tokyo.ac.jp/~makotoy/makotosite.css" />
</head>
<body>
<h1 class="title">Quotients of C*-algebras</h1>
<p>Preliminaries. Let <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&epsi;</mi></math> be any strictly positive number. We can consider
<math xmlns='http://www.w3.org/1998/Math/MathML' display='block'><msub><mi>f</mi> <mi>&epsi;</mi></msub><mo stretchy="false">(</mo><mi>x</mi><mo stretchy="false">)</mo><mo>=</mo><mfrac><mrow><mi>x</mi></mrow><mrow><mi>x</mi><mo>+</mo><mi>&epsi;</mi></mrow></mfrac><mo>,</mo><msub><mi>g</mi> <mi>&epsi;</mi></msub><mo stretchy="false">(</mo><mi>x</mi><mo stretchy="false">)</mo><mo>=</mo><mfrac><mrow><mi>x</mi></mrow><mrow><mo stretchy="false">(</mo><mi>x</mi><mo>+</mo><mi>&epsi;</mi><msup><mo stretchy="false">)</mo> <mn>2 </mn></msup></mrow></mfrac></math>
on <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo stretchy="false">[</mo><mn>0 </mn><mo>,</mo><mn>&infin;</mn><mo stretchy="false">[</mo></math>. <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>f</mi> <mi>&epsi;</mi></msub></math> is an increasing function. <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>g</mi> <mi>&epsi;</mi></msub><mo stretchy="false">(</mo><mi>x</mi><mo stretchy="false">)</mo></math> reaches its maximum at <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>x</mi><mo>=</mo><mi>&epsi;</mi></math>.</p><p>Proposition. Any closed ideal <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>I</mi></math> in a C*-algebra <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi></math> is closed with respect to the adjoint operation.</p><p>Proposition. When <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>X</mi></math> is a compact space and <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>I</mi></math> is a closed ideal of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>C</mi><mo stretchy="false">(</mo><mi>X</mi><mo stretchy="false">)</mo></math>, there exists a closed subset <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>K</mi></math> of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>X</mi></math> such that <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>I</mi><mo>=</mo><mo stretchy="false">{</mo><mi>f</mi><mo>&Element;</mo><mi>C</mi><mo stretchy="false">(</mo><mi>X</mi><mo stretchy="false">)</mo><mo>&VerticalBar;</mo><mi>x</mi><mo>&Element;</mo><mi>K</mi><mo>&Rightarrow;</mo><mi>f</mi><mo stretchy="false">(</mo><mi>x</mi><mo stretchy="false">)</mo><mo>=</mo><mn>0 </mn><mo stretchy="false">}</mo></math>.</p><p>Proposition. Let <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi></math> be a C*-algebra, <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>I</mi></math> a closed (bilateral) ideal of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi></math> and <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>x</mi></math> an element of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi></math>. We have <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><msup><mo>&DoubleVerticalBar;</mo> <mn>2 </mn></msup><mo>=</mo><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo>&DoubleVerticalBar;</mo></math> in the quotient Banach algebra <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi><mo>/</mo><mi>I</mi></math>.
Proof. Step 1. First assume that <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>x</mi></math> is normal and hence the C*-algebra <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>C</mi> <mi>x</mi></msub></math> generated by <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>x</mi></math> is commutative. The composition of the inclusion <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>C</mi> <mi>x</mi></msub><mo>&rightarrow;</mo><mi>A</mi></math> with the projection <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi><mo>&rightarrow;</mo><mi>A</mi><mo>/</mo><mi>I</mi></math> induces an injection <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&iota;</mi></math> of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>C</mi> <mi>x</mi></msub><mo>/</mo><mo stretchy="false">(</mo><msub><mi>C</mi> <mi>x</mi></msub><mo>&cap;</mo><mi>I</mi><mo stretchy="false">)</mo></math> into <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi><mo>/</mo><mi>I</mi></math>. Let <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>D</mi></math> denote the closure of the image of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&iota;</mi></math>. <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>C</mi> <mi>x</mi></msub></math> is a C*-algebra by the proposition above. To show that <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>D</mi></math> is a C*-algebra (hence the statement of this proposition for <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>x</mi></math>), it is enough to show that <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&iota;</mi></math> is an isometry.</p><p>We have the (contractive) Gelfand transform <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&Gamma;</mi><mo>:</mo><mi>D</mi><mo>&rightarrow;</mo><mi>C</mi><mo stretchy="false">(</mo><msub><mi>M</mi> <mi>D</mi></msub><mo stretchy="false">)</mo></math> where <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>M</mi> <mi>D</mi></msub></math> is the maximal ideal space of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>D</mi></math>. The composition <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&Gamma;</mi><mi>&iota;</mi><mo>:</mo><msub><mi>C</mi> <mi>x</mi></msub><mo>&rightarrow;</mo><mi>C</mi><mo stretchy="false">(</mo><msub><mi>M</mi> <mi>D</mi></msub><mo stretchy="false">)</mo></math> induces a continuous mapping <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo stretchy="false">(</mo><mi>&Gamma;</mi><mi>&iota;</mi><msup><mo stretchy="false">)</mo> <mo>#</mo></msup><mo>:</mo><msub><mi>M</mi> <mi>D</mi></msub><mo>&rightarrow;</mo><msub><mi>M</mi> <mi>x</mi></msub></math> where <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>M</mi> <mi>x</mi></msub></math> is the maximal ideal space of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>C</mi> <mi>x</mi></msub></math>. If this mapping is surjective, <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&Gamma;</mi><mi>&iota;</mi></math> becomes an isometry and so is <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&iota;</mi></math>.</p><p>As <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>D</mi></math> is commutative, ``<math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>f</mi><mo>&Element;</mo><mi>D</mi></math> is invertible'' is equivalent to ``<math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&Gamma;</mi><mo stretchy="false">(</mo><mi>f</mi><mo stretchy="false">)</mo></math> is invertible.'' If <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo stretchy="false">(</mo><mi>&Gamma;</mi><mi>&iota;</mi><msup><mo stretchy="false">)</mo> <mo>#</mo></msup></math> was not surjective, there would exist a non-invertible element <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>y</mi></math> in <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msub><mi>C</mi> <mi>x</mi></msub></math> such that <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&Gamma;</mi><mi>&iota;</mi><mi>y</mi><mo>=</mo><mn>1 </mn></math>. Then we should have <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&iota;</mi><mi>y</mi></math> invertible in <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>D</mi></math>, which is a contradiction. Hence <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo stretchy="false">(</mo><mi>&Gamma;</mi><mi>&iota;</mi><msup><mo stretchy="false">)</mo> <mo>#</mo></msup></math> is surjective.</p><p>Step 2. Let <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>x</mi></math> be any element of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi></math>. Put <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&delta;</mi><mo>=</mo><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo>&DoubleVerticalBar;</mo></math>. We are going to establsh <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><msup><mo>&DoubleVerticalBar;</mo> <mn>2 </mn></msup><mo>&leq;</mo><mi>&delta;</mi></math>.</p><p>Step 2-1. Suppose <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo>&not;</mo><mo>&Element;</mo><mi>I</mi></math> so that <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&delta;</mi><mi>&gt;</mi><mn>0 </mn></math> and we have 
<math xmlns='http://www.w3.org/1998/Math/MathML' display='block'><mi>x</mi><mo>=</mo><mi>x</mi><mfrac><mrow><mi>&delta;</mi></mrow><mrow><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo>+</mo><mi>&delta;</mi></mrow></mfrac><mo>+</mo><mi>x</mi><mfrac><mrow><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi></mrow><mrow><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo>+</mo><mi>&delta;</mi></mrow></mfrac><mo>.</mo></math>
For the first term in the right hand side, its norm squared is equal to the norm of
<math xmlns='http://www.w3.org/1998/Math/MathML' display='block'><mo stretchy="false">(</mo><mi>x</mi><mfrac><mrow><mi>&delta;</mi></mrow><mrow><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo>+</mo><mi>&delta;</mi></mrow></mfrac><msup><mo stretchy="false">)</mo> <mo>*</mo></msup><mi>x</mi><mfrac><mrow><mi>&delta;</mi></mrow><mrow><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo>+</mo><mi>&delta;</mi></mrow></mfrac><mo>=</mo><msup><mi>&delta;</mi> <mn>2 </mn></msup><msub><mi>g</mi> <mi>&delta;</mi></msub><mo stretchy="false">(</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">)</mo><mo>,</mo></math>
which is equal to <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&delta;</mi><mo>/</mo><mn>4 </mn></math>. Hence the norm of the first term of the RHS in
<math xmlns='http://www.w3.org/1998/Math/MathML' display='block'><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mo>=</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mfrac><mrow><mi>&delta;</mi></mrow><mrow><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo>+</mo><mi>&delta;</mi></mrow></mfrac><mo>+</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mfrac><mrow><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo></mrow><mrow><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo>+</mo><mi>&delta;</mi></mrow></mfrac><mo>.</mo></math>
is majored by <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msqrt><mrow><mi>&delta;</mi></mrow></msqrt><mo>/</mo><mn>2 </mn></math>. For the second term, by the step 1 the Banach subalgebra of <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi><mo>/</mo><mi>I</mi></math> generated by <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo></math> is a C*-algebra and we can do functional calculus there. We have <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo>&DoubleVerticalBar;</mo><msub><mi>f</mi> <mi>&delta;</mi></msub><mo stretchy="false">(</mo><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo stretchy="false">)</mo><mo>&DoubleVerticalBar;</mo><mo>=</mo><mn>1 </mn><mo>/</mo><mn>2 </mn></math>. Hence we obtain
<math xmlns='http://www.w3.org/1998/Math/MathML' display='block'><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mo>&DoubleVerticalBar;</mo><mo>&leq;</mo><mfrac><mrow><msqrt><mrow><mi>&delta;</mi></mrow></msqrt></mrow><mrow><mn>2 </mn></mrow></mfrac><mo>+</mo><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mo>&DoubleVerticalBar;</mo><mo>.</mo><mfrac><mrow><mn>1 </mn></mrow><mrow><mn>2 </mn></mrow></mfrac><mo>,</mo></math>
that leads to <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mo>&DoubleVerticalBar;</mo><mo>&leq;</mo><mi>&delta;</mi></math>.</p><p>Step 2-2. When <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo>&Element;</mo><mi>I</mi></math>, for any strictly positive real number <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&epsi;</mi></math> we have
<math xmlns='http://www.w3.org/1998/Math/MathML' display='block'><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mo>=</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mfrac><mrow><mi>&epsi;</mi></mrow><mrow><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo>+</mo><mi>&epsi;</mi></mrow></mfrac><mo>+</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mfrac><mrow><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo></mrow><mrow><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo>+</mo><mi>&epsi;</mi></mrow></mfrac><mo>=</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mfrac><mrow><mi>&epsi;</mi></mrow><mrow><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo>+</mo><mi>&epsi;</mi></mrow></mfrac></math>
in <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>A</mi><mo>/</mo><mi>I</mi></math>. When <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&epsi;</mi><mi>&lt;</mi><mo>&DoubleVerticalBar;</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo>&DoubleVerticalBar;</mo></math>, we have
<math xmlns='http://www.w3.org/1998/Math/MathML' display='block'><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mfrac><mrow><mi>&epsi;</mi></mrow><mrow><mo stretchy="false">[</mo><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo stretchy="false">]</mo><mo>+</mo><mi>&epsi;</mi></mrow></mfrac><mo>&DoubleVerticalBar;</mo><mo>&leq;</mo><mo>&DoubleVerticalBar;</mo><mi>x</mi><mfrac><mrow><mi>&epsi;</mi></mrow><mrow><msup><mi>x</mi> <mo>*</mo></msup><mi>x</mi><mo>+</mo><mi>&epsi;</mi></mrow></mfrac><mo>&DoubleVerticalBar;</mo><mo>=</mo><mfrac><mrow><msqrt><mrow><mi>&epsi;</mi></mrow></msqrt></mrow><mrow><mn>2 </mn></mrow></mfrac><mo>.</mo></math>
By taking the limit <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mi>&epsi;</mi><mo>&rightarrow;</mo><mn>0 </mn></math>, we get <math xmlns='http://www.w3.org/1998/Math/MathML' display='inline'><mo>&DoubleVerticalBar;</mo><mo stretchy="false">[</mo><mi>x</mi><mo stretchy="false">]</mo><mo>&DoubleVerticalBar;</mo><mo>=</mo><mn>0 </mn></math>.</p>
<div class="footer"><p>This document was written by <a href="mailto:makotoy@ms.u-tokyo.ac.jp">Yamashita Makoto</a>. Last update on 2005-10-09.</p></div>
</body>
</html>

