<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Changbin Bae on Medium]]></title>
        <description><![CDATA[Stories by Changbin Bae on Medium]]></description>
        <link>https://medium.com/@grhncgzsvb?source=rss-b40ef157b96d------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*TZvsaoWD3D2clb6Nns6ejw.jpeg</url>
            <title>Stories by Changbin Bae on Medium</title>
            <link>https://medium.com/@grhncgzsvb?source=rss-b40ef157b96d------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 22:31:28 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@grhncgzsvb/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Column 4 — From Lindblad Master Equation to Coherence Time]]></title>
            <link>https://medium.com/@grhncgzsvb/column-4-from-lindblad-master-equation-to-coherence-time-77d1b9ba08f1?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/77d1b9ba08f1</guid>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[quantum-physics]]></category>
            <category><![CDATA[quantum-computer]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Mon, 12 Oct 2026 00:45:34 GMT</pubDate>
            <atom:updated>2026-10-12T00:45:34.011Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Column 3 derived the Lindblad master equation from a microscopic qubit-bath model. In the equation the bath interacts only through jump operator and rates. This column evaluates the two channel that matter most for the qubits, energy exchange and pure dephasing, and shows that they reproduce the T1 and T2 of column 2 exactly.</em></p><p>Let’s look at the Lindblad master equation:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JgNN48DivCoZro4fbztkxg.png" /></figure><p>Here <em>L</em> is a jump operator, which is a channel through which the bath acts on the qubit. We can typically think of two kinds of channel, one is for the energy exchange, and one is for the pure phase relaxation. Let’s examine the master equation in these two frameworks.</p><p><strong>Amplitude Damping</strong></p><p>The first channel is amplitude damping. It happens when the qubit exchanges energy with the bath. It arises when the coupling has the form:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fj4i9aiPfuZ2VSTRx4gmurL&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fj4i9aiPfuZ2VSTRx4gmurL" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2401f03700ae4a68096c7c47e45c2619/href">https://medium.com/media/2401f03700ae4a68096c7c47e45c2619/href</a></iframe><p>The first term is when the qubit state drops from |1&gt; to |0&gt; while creating a bosonic excitation in the bath, and second term is the reverse process. Here the jump operator is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FhTETqjcW9ywycvZ9DtvRHh&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FhTETqjcW9ywycvZ9DtvRHh" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/80cf250c3c1e281cf99adfb4d67ae4d0/href">https://medium.com/media/80cf250c3c1e281cf99adfb4d67ae4d0/href</a></iframe><p>with fixed rate <em>γ</em>, and the master equation becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5KVN3n1KeFQwTrTABvUey9&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5KVN3n1KeFQwTrTABvUey9" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4df5664ed6e3e431a37442d5016109e4/href">https://medium.com/media/4df5664ed6e3e431a37442d5016109e4/href</a></iframe><p>We can derive the Bloch equations from this. Recall the convention of column 2: |0&gt; is the ground state at the north pole where <em>r_z = +1</em>. Using the parametrization:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6Rz8yafkKUff487VnzMd3a&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6Rz8yafkKUff487VnzMd3a" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0fc09b8057153f3b8376c89426e65176/href">https://medium.com/media/0fc09b8057153f3b8376c89426e65176/href</a></iframe><p>Plug in the <em>ρ_11 </em>element into the master equation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FacdvDYtKiHiPAVmhMLXiCw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FacdvDYtKiHiPAVmhMLXiCw" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/62a9aa77241c936c6134ee9afcfabf63/href">https://medium.com/media/62a9aa77241c936c6134ee9afcfabf63/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fs73wCUtML8a8H1huKXrnMT&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fs73wCUtML8a8H1huKXrnMT" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/23567a58b9885c17fd8ab5630f0de16f/href">https://medium.com/media/23567a58b9885c17fd8ab5630f0de16f/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fc2Zq1wRdokDREpWPMuyj4C&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fc2Zq1wRdokDREpWPMuyj4C" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a37f7881151307dd01083e510d802e72/href">https://medium.com/media/a37f7881151307dd01083e510d802e72/href</a></iframe><p>Convert into Bloch vector component:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqTMvXzQewNu9vWbjTFmoDi&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqTMvXzQewNu9vWbjTFmoDi" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a5a7b55a2b9b838765c6ac1882010263/href">https://medium.com/media/a5a7b55a2b9b838765c6ac1882010263/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6KVYKL6SMjfbdJm3ZDW7Nc&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6KVYKL6SMjfbdJm3ZDW7Nc" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c7cfb25d57c9d8d90b856c540d9d2762/href">https://medium.com/media/c7cfb25d57c9d8d90b856c540d9d2762/href</a></iframe><p>For the transverse component:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FApHvPTCdKH3q3K2HYJQNG&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FApHvPTCdKH3q3K2HYJQNG" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/f47078883e4b16902202e39f0d48a608/href">https://medium.com/media/f47078883e4b16902202e39f0d48a608/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9NN4RZ3wfHG22EuH5zFnSL&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9NN4RZ3wfHG22EuH5zFnSL" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a44217d09d04f7fc3d5db142fd2ff1ba/href">https://medium.com/media/a44217d09d04f7fc3d5db142fd2ff1ba/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Frt1wBQRoT8LsuiaYHpi831&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Frt1wBQRoT8LsuiaYHpi831" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/045771c5153e5b87ea039d3b8d0ba437/href">https://medium.com/media/045771c5153e5b87ea039d3b8d0ba437/href</a></iframe><p>Therefore:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2GVtSR6pvg12HDoh2nRxn3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2GVtSR6pvg12HDoh2nRxn3" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c4e8bd89a15fdc726b3e80d1c9077168/href">https://medium.com/media/c4e8bd89a15fdc726b3e80d1c9077168/href</a></iframe><p>This is exactly the same form of the Bloch equations from column 2, with:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fkjxopjhz2ZGV8ngq5zkkci&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fkjxopjhz2ZGV8ngq5zkkci" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/3471a33d46010116ee675e3f32d1eb27/href">https://medium.com/media/3471a33d46010116ee675e3f32d1eb27/href</a></iframe><p><strong>Pure Dephasing</strong></p><p>The second channel is pure dephasing. The bath randomizes the qubit’s phase without moving energy. It arises from the spin-boson coupling:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FipdUbyjNzzGHkWspYSgZ92&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FipdUbyjNzzGHkWspYSgZ92" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/87632d78e067ddfab68c40e7a53e68e4/href">https://medium.com/media/87632d78e067ddfab68c40e7a53e68e4/href</a></iframe><p>The jump operator is therefore:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fv4NpZzQyKnmVc2QoiTSoCH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fv4NpZzQyKnmVc2QoiTSoCH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/66c3951878612c7f929f9181e50bccbf/href">https://medium.com/media/66c3951878612c7f929f9181e50bccbf/href</a></iframe><p>With rate <em>γ_φ</em>, the master equation becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvTyA591gA7Lh8gtZSHtbrW&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvTyA591gA7Lh8gtZSHtbrW" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/282166eb93dc95816a167b0bcb791aae/href">https://medium.com/media/282166eb93dc95816a167b0bcb791aae/href</a></iframe><p>The equation of the longitudinal component case:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqASiDPShcfkVXBMB9E8x9w&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqASiDPShcfkVXBMB9E8x9w" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/85d4c416b2594b2013e7262ed9ce2a03/href">https://medium.com/media/85d4c416b2594b2013e7262ed9ce2a03/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F8aJNRJ94vyp53qCeA9jVNN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F8aJNRJ94vyp53qCeA9jVNN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/62544847b173e32837005f7306d40d79/href">https://medium.com/media/62544847b173e32837005f7306d40d79/href</a></iframe><p>We can see there is no energy relaxation, just as we assumed. For the transverse case:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fi72eWVa6AuhPK1nTkrfdpv&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fi72eWVa6AuhPK1nTkrfdpv" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/390a980a1526751a635b0a1d9d05a915/href">https://medium.com/media/390a980a1526751a635b0a1d9d05a915/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqQsCeuRGBrhEeNR2osANH8&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqQsCeuRGBrhEeNR2osANH8" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4b3e50031d50f852496e186911b2f218/href">https://medium.com/media/4b3e50031d50f852496e186911b2f218/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9VSySRhDy52GuXwvRdbFHX&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9VSySRhDy52GuXwvRdbFHX" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e2e3f8b3be20bf534be31852ac796b4a/href">https://medium.com/media/e2e3f8b3be20bf534be31852ac796b4a/href</a></iframe><p>The Bloch sphere shrinks into the z-axis. The transverse rate defines the pure dephasing time:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FtuswAKSYo8dNFBP6qwW6Ep&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FtuswAKSYo8dNFBP6qwW6Ep" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b8cd8a9b9643b1fb439f7d4850d77ce0/href">https://medium.com/media/b8cd8a9b9643b1fb439f7d4850d77ce0/href</a></iframe><p>Now, suppose that both channels act on the system. Their dissipators simply add:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fxe6Lg3rb38hsxsXQE2Qvtw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fxe6Lg3rb38hsxsXQE2Qvtw" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a54d6e10dc055cbc8511066dcdecf2e4/href">https://medium.com/media/a54d6e10dc055cbc8511066dcdecf2e4/href</a></iframe><p>Therefore, we get the <em>T2</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FtybAandbh8HdpW7m4UXcZ4&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FtybAandbh8HdpW7m4UXcZ4" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/38193467ca26b7ee2a61b462d07fabdd/href">https://medium.com/media/38193467ca26b7ee2a61b462d07fabdd/href</a></iframe><p>This is the full decoherence of a qubit. It gives the inequality introduced in column 2:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Ff6BrKSoD34JEczDGbZrtNo&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Ff6BrKSoD34JEczDGbZrtNo" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ce5a2969ab7db34a775f6d5347dfb4e0/href">https://medium.com/media/ce5a2969ab7db34a775f6d5347dfb4e0/href</a></iframe><p>with equality only when the pure dephasing is switched off.</p><p><strong>When the Markov Approximations Break Down</strong></p><p>The Lindblad master equation is built on the assumption that the bath correlation time <em>τ_B </em>is far shorter than <em>T1 </em>and<em> T2</em>. However, for the dominant noise source in most superconducting qubits, this assumption fails. The culprit is 1/f noise, which comes from the ensemble of the two-level fluctuators from charged defects. The noise becomes larger at low frequencies, which means the low-frequency noise is correlated over very long times, the exact opposite of the Markov approximation. In column 4, we will further develop the theory of the noise and see how its spectral shape determines a qubit’s decoherence.</p><p><strong>References</strong></p><ul><li><em>Daniel A. Lidar, “Lecture Notes on the Theory of Open Quantum Systems”</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=77d1b9ba08f1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 3 — Derivation of the Lindblad Master Equation]]></title>
            <link>https://medium.com/@grhncgzsvb/column-3-derivation-of-the-lindblad-master-equation-561d12d5b2a3?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/561d12d5b2a3</guid>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[quantum-computer]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <category><![CDATA[quantum-physics]]></category>
            <category><![CDATA[physics]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Sat, 10 Oct 2026 18:40:49 GMT</pubDate>
            <atom:updated>2026-10-10T18:40:49.894Z</atom:updated>
            <content:encoded><![CDATA[<p><em>In column 2 we built the Bloch equations as phenomenology, with the decay rates installed by symmetry rather than derived. This column derives the Lindblad master equation, which is a fundamental model of the open quantum system such as the qubit that interacts with the bath. We apply three approximations to derive the master equation from first principles.</em></p><p>No quantum system is truly isolated. Every qubit, whether is a superconducting circuit, a trapped ion in an electromagnetic cage, or an electron spin in silicon, is embedded in a physical environment it cannot completely ignore. Whatever the physical manifestation is, the coupling between the qubit and the environment is real, and it degrades quantum information.</p><p>In column 2 we derived the Bloch equations from symmetry. The result is correct and useful, but it says nothing about how to determine the actual T1, T2 values. When we want to improve those parameters, we need to know which physical coupling controls it. Here, we will find out how to describe the physical coupling between the system and bath by deriving the Lindblad master equation from scratch.</p><p><strong>The System-Bath Hamiltonian</strong></p><p>Before we begin, we set ℏ = 1 for convenience.</p><p>We model the world (total <em>T</em>) as a qubit (system <em>S</em>) plus a bath (<em>B</em>), living in the tensor-product Hilbert space:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3NAgArdQj4XVy6RTdwDrRw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3NAgArdQj4XVy6RTdwDrRw" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/f300d1dec24ecff23462432457c96757/href">https://medium.com/media/f300d1dec24ecff23462432457c96757/href</a></iframe><p>The full density matrix evolves unitarily, but we only care about the qubit <em>ρ(t)</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fn7zYVF2SgwiBaVCTMEKxC1&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fn7zYVF2SgwiBaVCTMEKxC1" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c0ec2d7d63d522f8f654b35b2f7aa517/href">https://medium.com/media/c0ec2d7d63d522f8f654b35b2f7aa517/href</a></iframe><p>where Tr_<em>B</em> sums over the bath degrees of freedom. Our goal is to make the unitary evolution of <em>ρ_T </em>into a closed, self-contained equation of <em>ρ</em>.</p><p>The total Hamiltonian splits into three pieces:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6di8DCd4aM5NnmFqBUVNEC&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6di8DCd4aM5NnmFqBUVNEC" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d3619233b682ab16d85a015a648e7b28/href">https://medium.com/media/d3619233b682ab16d85a015a648e7b28/href</a></iframe><p><em>H_S</em> only acts on the qubit, <em>H_B</em> only on the bath, and <em>H_I</em> is the interaction. The parameter <em>α</em> measures the coupling strength.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iTrRmE-yfmWF9ldPZQ5Pfg.jpeg" /></figure><p>The simplest realization of the qubit and the bath is the spin-boson model. The qubit’s free Hamiltonian corresponds to a spin:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FuHhKLac5MeoTXiGVJHtJ7r&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FuHhKLac5MeoTXiGVJHtJ7r" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/1d04017e166cd27b0dca1b90f7a34d20/href">https://medium.com/media/1d04017e166cd27b0dca1b90f7a34d20/href</a></iframe><p>where <em>ω_q</em> is the qubit’s transition frequency. The bath is a collection of harmonic oscillators, which usually correspond to photons or phonons, which are bosons:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fun35cM1vsHKBjtezKc8F6c&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fun35cM1vsHKBjtezKc8F6c" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d81d0a5a43db945ecc5bc116e431a64c/href">https://medium.com/media/d81d0a5a43db945ecc5bc116e431a64c/href</a></iframe><p>where <em>b_k</em> is the bosonic creation/annihilation operators for mode <em>k</em>. The interaction can be expressed generally as a sum of products of a system operator with a bath operator:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2YwWKZaxXkWHiTvi6Psg8g&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2YwWKZaxXkWHiTvi6Psg8g" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/9a07c3e75374b8e12d4ffb484819312a/href">https://medium.com/media/9a07c3e75374b8e12d4ffb484819312a/href</a></iframe><p>In this case, the coupling is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2tbS4LxningAyNRaKpJeGv&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2tbS4LxningAyNRaKpJeGv" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/05a5027b8340be4d5dce03d2ee9074b0/href">https://medium.com/media/05a5027b8340be4d5dce03d2ee9074b0/href</a></iframe><p>which ties the qubit operator to the bath’s displacement. The mode-specific coupling constants <em>λ_k</em> determine the scale of the <em>α</em>.</p><p>We then move to the interaction picture with respect to the free Hamiltonian:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fw3ktAR1mqGX9d6c1bpoE8i&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fw3ktAR1mqGX9d6c1bpoE8i" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d1a75325eac4bb8f8f8d1edb3c33094a/href">https://medium.com/media/d1a75325eac4bb8f8f8d1edb3c33094a/href</a></iframe><p>Any operator <em>O</em> becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fr1pN5riWhP73s93BgGEBr3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fr1pN5riWhP73s93BgGEBr3" width="227" height="136" frameborder="0" scrolling="no"><a href="https://medium.com/media/4bc7bccb2efba31feef50397a0372cc3/href">https://medium.com/media/4bc7bccb2efba31feef50397a0372cc3/href</a></iframe><p>and the state becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FxumHt4tVDrPgdKe4z8XAom&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FxumHt4tVDrPgdKe4z8XAom" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/489c9266d4b88e7c72d394875ff753a1/href">https://medium.com/media/489c9266d4b88e7c72d394875ff753a1/href</a></iframe><p>This strips away the fast free-evolution phases and leaves only the slow drift caused by the coupling. From now on, every tilde denotes an interaction-picture quantity, and the entire derivation lives in this interaction picture until we transform back at the very end.</p><p><strong>The Equation of Motion</strong></p><p>Let’s start with the von Neumann equation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FbRNHQWfPWwLzUFVrVLGpFE&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FbRNHQWfPWwLzUFVrVLGpFE" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e402ac1eb4c9932e9dc213e2a573ccb1/href">https://medium.com/media/e402ac1eb4c9932e9dc213e2a573ccb1/href</a></iframe><p>In the interaction picture, the equation collapses to the coupling term alone:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqsNKvw1FoJ6t1JMYkU63y7&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqsNKvw1FoJ6t1JMYkU63y7" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/07d12617d834e02927aac86b4f42bcc3/href">https://medium.com/media/07d12617d834e02927aac86b4f42bcc3/href</a></iframe><p>You can see the free Hamiltonian components <em>H_S</em> and <em>H_B</em> have dropped. We want the equation of the qubit’s density matrix, not the full joint state. To expose it, integrate it from <em>0</em> to <em>t</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FerKgZ3YEGc8smTujwme15d&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FerKgZ3YEGc8smTujwme15d" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/dc7514793b387ef9d4bb5479865871fd/href">https://medium.com/media/dc7514793b387ef9d4bb5479865871fd/href</a></iframe><p>and put it in the eq. (1):</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FknxdddyHckZniQit7GH9Vo&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FknxdddyHckZniQit7GH9Vo" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/47041868f162262846831350b7c37c89/href">https://medium.com/media/47041868f162262846831350b7c37c89/href</a></iframe><p>We call the second term a “memory integral”, since it represents the history of the state from <em>s = 0</em> to <em>t</em>. Now trace out the bath (Tr_<em>B</em>):</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FdaqR72T66hFJhHuZz7zaRg&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FdaqR72T66hFJhHuZz7zaRg" width="705" height="221" frameborder="0" scrolling="no"><a href="https://medium.com/media/d34ddc6821042691933da5963802f184/href">https://medium.com/media/d34ddc6821042691933da5963802f184/href</a></iframe><p>Let’s examine the <em>i</em>) term. At <em>t = 0</em>, the interaction-picture rotation is the identity:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqVvyiZWJY1e1zmqKNtSjQH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqVvyiZWJY1e1zmqKNtSjQH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/efa752750548b32ad173153d01f5790d/href">https://medium.com/media/efa752750548b32ad173153d01f5790d/href</a></iframe><p>We suppose the qubit and bath are initially uncorrelated (the “fresh qubit” assumption), with the bath in a stationary state <em>ρ_B</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FfCjtQ7pvpRxFU819AhWznS&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FfCjtQ7pvpRxFU819AhWznS" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/154246f1f52b873d510e82319b4d8847/href">https://medium.com/media/154246f1f52b873d510e82319b4d8847/href</a></iframe><p>Here, “stationary” means it does not change in time.</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F28weVVbdTc2mSZKVobCDaB&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F28weVVbdTc2mSZKVobCDaB" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ff15e53977060bf517d885fde624ca7e/href">https://medium.com/media/ff15e53977060bf517d885fde624ca7e/href</a></iframe><p>The coupling Hamiltonian in the interaction picture can be written as follows:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5e1XoU2JWM3qrrMkpADQJy&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5e1XoU2JWM3qrrMkpADQJy" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6652e2478e6e3eb70e061be133ea8bf6/href">https://medium.com/media/6652e2478e6e3eb70e061be133ea8bf6/href</a></iframe><p>Each system operator and bath operator rotates under their own Hamiltonian. Now expand the commutator in <em>i</em>):</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fxhnx5HCXLbn9dHf5NskLiU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fxhnx5HCXLbn9dHf5NskLiU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8e2dc8054f6ece46c8f7e3a6238a06dc/href">https://medium.com/media/8e2dc8054f6ece46c8f7e3a6238a06dc/href</a></iframe><p>Taking the partial trace over bath degrees of freedom and using the cyclicity of the trace:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FhKD4rxeMfq9VKLbHncFKiy&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FhKD4rxeMfq9VKLbHncFKiy" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/de5e89a32b500e687f53260fe760293b/href">https://medium.com/media/de5e89a32b500e687f53260fe760293b/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FitSM9wYLsaoAgtrcQDXZi2&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FitSM9wYLsaoAgtrcQDXZi2" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/fda66a0d4151d0f50048b9fb4494efb1/href">https://medium.com/media/fda66a0d4151d0f50048b9fb4494efb1/href</a></iframe><p>The partial trace term:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FggmvPK3Ug6XcBZzvhPct3g&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FggmvPK3Ug6XcBZzvhPct3g" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/08ca6b2898fac4225174266462f452d8/href">https://medium.com/media/08ca6b2898fac4225174266462f452d8/href</a></iframe><p>Since <em>ρ_B</em> is stationary, it commutes with <em>H_B</em>, thus we get the expectation value of the bath operator:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F25repYHMoudZdZUZ35Dret&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F25repYHMoudZdZUZ35Dret" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/464d54f06cc860e6c4f12e26f0379c74/href">https://medium.com/media/464d54f06cc860e6c4f12e26f0379c74/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FbF9YUo1awC2XPkd4JQxq9s&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FbF9YUo1awC2XPkd4JQxq9s" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e55124a0681c0841bbd99a5611e693e5/href">https://medium.com/media/e55124a0681c0841bbd99a5611e693e5/href</a></iframe><p>Thus, the <em>i</em>) term becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9k6ejruhHF8WmTK6mZyTpF&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9k6ejruhHF8WmTK6mZyTpF" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5ffbad0ea0777ccc2f23909356bd4e32/href">https://medium.com/media/5ffbad0ea0777ccc2f23909356bd4e32/href</a></iframe><p>Let’s suppose the bath operator has non-zero average in the bath’s stationary state. Then we can define the deviation of <em>B_k</em> from the mean:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4Fx1jHdLT4nsFHJct9zfYj&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4Fx1jHdLT4nsFHJct9zfYj" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/19fd07cc7efdfad583dcec5c4c928408/href">https://medium.com/media/19fd07cc7efdfad583dcec5c4c928408/href</a></iframe><p>Substituting this into coupling Hamiltonian:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkyQ6p843uB6Nj4oJQ6vPmJ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkyQ6p843uB6Nj4oJQ6vPmJ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6c884ae5c3e596ce5bd94ab191144251/href">https://medium.com/media/6c884ae5c3e596ce5bd94ab191144251/href</a></iframe><p>The second term acts only on the qubit, so we can move it into the system Hamiltonian:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FaHr92TT95cg1r68avcdQ9i&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FaHr92TT95cg1r68avcdQ9i" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c567b4ad776bbbf7da8664a643960964/href">https://medium.com/media/c567b4ad776bbbf7da8664a643960964/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FmvGbpDj3sZym73fUEDNPob&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FmvGbpDj3sZym73fUEDNPob" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0ddaa601c31302e27dccfa58f280ef0f/href">https://medium.com/media/0ddaa601c31302e27dccfa58f280ef0f/href</a></iframe><p>The total Hamiltonian is untouched. We have only changed the form of the components. The new bath operators now have zero mean:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Few6kyea3Rm9j2DZ8Uy1x6c&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Few6kyea3Rm9j2DZ8Uy1x6c" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/419d32a7ea4f25bf86454a8edb9ad66f/href">https://medium.com/media/419d32a7ea4f25bf86454a8edb9ad66f/href</a></iframe><p>Therefore, we can always enforce the expectation value of the bath operator <em>B</em> to be zero, making the term <em>i</em>) also equals to zero. The eq. (2) then becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqrAgsAncFueube2QPPAUxZ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqrAgsAncFueube2QPPAUxZ" width="678" height="192" frameborder="0" scrolling="no"><a href="https://medium.com/media/887ac650b737309506729d34bec47cee/href">https://medium.com/media/887ac650b737309506729d34bec47cee/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FjjKAkxWiezgMJjfa7p8Pjg&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FjjKAkxWiezgMJjfa7p8Pjg" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/cde7528b88e87d4dd8a308fe0976c7f3/href">https://medium.com/media/cde7528b88e87d4dd8a308fe0976c7f3/href</a></iframe><p><strong>The Born Approximation</strong></p><p>Imagine putting a drop of dye into the ocean. The dye disperses, but the ocean is so enormous that it barely makes any change. Similarly, the bath is so large compared to the qubit, and the coupling is so weak, that the back action of the qubit on the bath is negligible. The bath stays in its stationary state, and the qubit never build up meaningful correlations. This is the Born approximation. Thus, the joint state factorizes at all times:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqtqL5Lks9QinpzU26LBYLz&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqtqL5Lks9QinpzU26LBYLz" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/29b087ac71c0dfdbcb1fc0bbd51a5f05/href">https://medium.com/media/29b087ac71c0dfdbcb1fc0bbd51a5f05/href</a></iframe><p>The correlation term is <em>O(α)</em>, and it is inside the integral which is already <em>O(α²)</em> in eq. (3), so dropping it costs only <em>O(α³)</em> which makes sense for small <em>α</em>. Thus the eq. (3) becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fn5FMWAvoX3YfatTwEWEZqt&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fn5FMWAvoX3YfatTwEWEZqt" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c7330e4ef06f5c09abb20622dbf3c978/href">https://medium.com/media/c7330e4ef06f5c09abb20622dbf3c978/href</a></iframe><p><strong>The Markov Approximation</strong></p><p>Now we obtained the equation in qubit state <em>ρ</em>, but it still contains the memory integral. The qubit state depends on the entire history between <em>0</em> and <em>t </em>through <em>ρ(s)</em>. First, we carry out the bath trace of eq. (4):</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FuFYzREfsNYVnT9EXbz4ibH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FuFYzREfsNYVnT9EXbz4ibH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/35193380443da33095eded6b4957e759/href">https://medium.com/media/35193380443da33095eded6b4957e759/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FeFyUPJohz1iV48S5DNeiR3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FeFyUPJohz1iV48S5DNeiR3" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/332dcad4277636fb198a4f7bdc4484fc/href">https://medium.com/media/332dcad4277636fb198a4f7bdc4484fc/href</a></iframe><p>Substituting the explicit form of interaction:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3YSSJUQPZkcHYftLiieWoJ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3YSSJUQPZkcHYftLiieWoJ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/338d519d5ef7448b3786c8fc3c16d3f2/href">https://medium.com/media/338d519d5ef7448b3786c8fc3c16d3f2/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FtsVHCoTs1o8rEj2AhVtYiy&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FtsVHCoTs1o8rEj2AhVtYiy" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/3caa38ca4211d11c5f42295c4e1c9c1a/href">https://medium.com/media/3caa38ca4211d11c5f42295c4e1c9c1a/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FsC2ykcnuxNkTwFjkYaYdqS&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FsC2ykcnuxNkTwFjkYaYdqS" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/507f9be45fcafd5860f9d9a2c9e44c4a/href">https://medium.com/media/507f9be45fcafd5860f9d9a2c9e44c4a/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FeHyqo2Fdb1UCZQ5wesHwnU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FeHyqo2Fdb1UCZQ5wesHwnU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/1677894745aacde72d1c3c9d19748875/href">https://medium.com/media/1677894745aacde72d1c3c9d19748875/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fc9XaGSLrXk1euFyqFpKqF7&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fc9XaGSLrXk1euFyqFpKqF7" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4af3ef59ed4b35bb716016a4d6b461cb/href">https://medium.com/media/4af3ef59ed4b35bb716016a4d6b461cb/href</a></iframe><p>Tracing out the bath for each term:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FdADuJx9EzrsS2AuG8yxQpM&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FdADuJx9EzrsS2AuG8yxQpM" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b3523eaa5dc8816a427fc5f7c610fbfb/href">https://medium.com/media/b3523eaa5dc8816a427fc5f7c610fbfb/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3dVcpv2zFZoZRn3yiaFaMy&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3dVcpv2zFZoZRn3yiaFaMy" width="270" height="177" frameborder="0" scrolling="no"><a href="https://medium.com/media/9c17a44feda7f592cec4dc41ba80ca50/href">https://medium.com/media/9c17a44feda7f592cec4dc41ba80ca50/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FdXrpN6zjftovbcDpgY3fUQ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FdXrpN6zjftovbcDpgY3fUQ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0870c988817a83197d7cd2dbb10a8146/href">https://medium.com/media/0870c988817a83197d7cd2dbb10a8146/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fkeh4LZSCfs7i1DQcbGsUXs&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fkeh4LZSCfs7i1DQcbGsUXs" width="270" height="177" frameborder="0" scrolling="no"><a href="https://medium.com/media/411220976a8068fa092924f7009fbac2/href">https://medium.com/media/411220976a8068fa092924f7009fbac2/href</a></iframe><p>Only two types appear. Look at the term <em>i</em>). Using the definition of the interaction-picture and the fact that <em>ρ_B </em>commutes with <em>H_B</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqzvVdR2BoJoWyqSZCAE88V&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqzvVdR2BoJoWyqSZCAE88V" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/13db1b5eece39315a68594123d90d773/href">https://medium.com/media/13db1b5eece39315a68594123d90d773/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F68h8WjHE62EWmrFoY6Rv2M&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F68h8WjHE62EWmrFoY6Rv2M" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/86d11f643ead31a9c1cd891a777035ff/href">https://medium.com/media/86d11f643ead31a9c1cd891a777035ff/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FoYztwR7NQtFF8mqPMGtULL&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FoYztwR7NQtFF8mqPMGtULL" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e437e6a7c2f89f27587068d387462a02/href">https://medium.com/media/e437e6a7c2f89f27587068d387462a02/href</a></iframe><p>We call <em>C</em> a bath correlation function, which measures how strongly a bath fluctuation at time 0 is still felt at the time <em>t — s</em> later. The term <em>ii</em>) becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkRW4uXh7vGTWhMwRiWkAQ1&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkRW4uXh7vGTWhMwRiWkAQ1" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7dc2f56fee1d16e6b7c379cc54fe2bcb/href">https://medium.com/media/7dc2f56fee1d16e6b7c379cc54fe2bcb/href</a></iframe><p>Therefore the eq. (5) becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fv7T9gPvRFQLsiRWdPYEoJ5&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fv7T9gPvRFQLsiRWdPYEoJ5" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2b27efb677a8cbd6ae50b4d74a3a786f/href">https://medium.com/media/2b27efb677a8cbd6ae50b4d74a3a786f/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fu8HxgdJ1QKNPM7wpLmnUwv&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fu8HxgdJ1QKNPM7wpLmnUwv" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ba91ea8abe006a05fb7d01f05993681e/href">https://medium.com/media/ba91ea8abe006a05fb7d01f05993681e/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FbaKA4xesfJQay18SzTtbh2&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FbaKA4xesfJQay18SzTtbh2" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0f89ab3ada9b7eecff9278921ee8b753/href">https://medium.com/media/0f89ab3ada9b7eecff9278921ee8b753/href</a></iframe><p>The equation of motion therefore becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FgVYWtjrtjk87j2qyiySqt7&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FgVYWtjrtjk87j2qyiySqt7" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2e4b945e5f4e785b879a44f918aba46d/href">https://medium.com/media/2e4b945e5f4e785b879a44f918aba46d/href</a></iframe><p>Writing <em>t — s = τ</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4mj5GQwKwJKUk91uYmPBFA&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4mj5GQwKwJKUk91uYmPBFA" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d5de9567e4e02b75190700a8b5359dcd/href">https://medium.com/media/d5de9567e4e02b75190700a8b5359dcd/href</a></iframe><p>Now it’s time to introduce the Markov approximation. The memory integral is weighted by the correlation function. For bath in a stationary state, the correlation function decays to zero over a characteristic time <em>τ_B, </em>often modeled as an exponential decay:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FhunU2Six83jWXB1XRrdEP&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FhunU2Six83jWXB1XRrdEP" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/597f67dcac95f25c2158e3553eeb3d91/href">https://medium.com/media/597f67dcac95f25c2158e3553eeb3d91/href</a></iframe><p>Then we can extend the upper limit of the integral from <em>t</em> to ∞ without any cost. Also, If the qubit evolves slowly on the scale of <em>τ_B</em>, that is, if <em>τ_B</em> &lt;&lt; <em>T1, T2</em>, then the qubit state barely changes over the time window, and we can approximate it to be constant:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FcGDfL3JuuiV2D63Q4jvZPC&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FcGDfL3JuuiV2D63Q4jvZPC" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/9ca7a85e0c431a6dddfa3bc829fc15df/href">https://medium.com/media/9ca7a85e0c431a6dddfa3bc829fc15df/href</a></iframe><p>Putting it all together, the result is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9iEYVwprnQUaZ4wuxHZG1S&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9iEYVwprnQUaZ4wuxHZG1S" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a8551a16dbb880885c402dd310f97634/href">https://medium.com/media/a8551a16dbb880885c402dd310f97634/href</a></iframe><p>This is called the Redfield equation.</p><p><strong>The Secular Approximation and the Lindblad Equation</strong></p><p>The Redfield equation (6) has a hidden defect. Iterated over long time, it can lead the density matrix to have negative eigenvalues (negative probabilities), which are physically unreasonable. To handle this problem, we start with the spectral decomposition of <em>H_S</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqH2RpX3FvMvmVX3xftzTC9&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqH2RpX3FvMvmVX3xftzTC9" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5eb12c6f21558b8306b46b26a65fd644/href">https://medium.com/media/5eb12c6f21558b8306b46b26a65fd644/href</a></iframe><p>where <em>Π(ε)</em> is the projection operator onto the eigenspace of the qubit with energy ε. Putting the identity on both side of the system operator:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Foh7SyTFZhoUA2UiWRxVjTU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Foh7SyTFZhoUA2UiWRxVjTU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e9410c116dbaa62bc3621505988a3908/href">https://medium.com/media/e9410c116dbaa62bc3621505988a3908/href</a></iframe><p>The interaction-picture system operator then becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fie789HVKV5J6KNeN3WaAee&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fie789HVKV5J6KNeN3WaAee" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/70b604480058b201c6e2b6dbde5b9eef/href">https://medium.com/media/70b604480058b201c6e2b6dbde5b9eef/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FiEhGCopoVJTi4pq8PPYGJC&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FiEhGCopoVJTi4pq8PPYGJC" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/877132ccbf25124aa4b68457000693fb/href">https://medium.com/media/877132ccbf25124aa4b68457000693fb/href</a></iframe><p>The phase depends only on the energy difference <em>ω = ε’ — ε</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FbsLPfo7tjWh61176J5Qmef&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FbsLPfo7tjWh61176J5Qmef" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a58276455e7aa98379f4f349f32e6139/href">https://medium.com/media/a58276455e7aa98379f4f349f32e6139/href</a></iframe><p>where:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F21nRHGcedzi1ZJWEeVCayn&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F21nRHGcedzi1ZJWEeVCayn" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/77e2f896238418f0381187526691212a/href">https://medium.com/media/77e2f896238418f0381187526691212a/href</a></iframe><p>If <em>ω &gt; 0</em>, the qubit emits the energy into the bath, and vice versa. The Hermitian conjugate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fd474pVL96ncym2dsojDvtH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fd474pVL96ncym2dsojDvtH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c739930d61e42ec3a10ee4a1decc4ebc/href">https://medium.com/media/c739930d61e42ec3a10ee4a1decc4ebc/href</a></iframe><p>Now substitute this into eq. (6). Using the conjugated form for the one at time <em>t</em>, and the original form for the one at time <em>t — τ</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5GBEtpgDnrvxKSfaT5QmsJ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5GBEtpgDnrvxKSfaT5QmsJ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/fb6e95bd4c7cf34eaa39980227709580/href">https://medium.com/media/fb6e95bd4c7cf34eaa39980227709580/href</a></iframe><p>The commutator in eq. (6) then becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Ft96YzZcS976tejUkaqWKxr&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Ft96YzZcS976tejUkaqWKxr" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c7e8c7899213bb9d83c8c90e6929dfb6/href">https://medium.com/media/c7e8c7899213bb9d83c8c90e6929dfb6/href</a></iframe><p>Now let’s define the one-sided (<em>τ </em>&lt; 0) Fourier transform of the bath correlation function:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4SztXuD7YKxBfA7qhYy3Co&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4SztXuD7YKxBfA7qhYy3Co" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/084ba1387d94e05fa4311983a851d90b/href">https://medium.com/media/084ba1387d94e05fa4311983a851d90b/href</a></iframe><p>Then eq. (6) becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7tNy4MnDQV232iDSizjXCb&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7tNy4MnDQV232iDSizjXCb" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/96d25b986e36d6d12bca70cd1b53f95c/href">https://medium.com/media/96d25b986e36d6d12bca70cd1b53f95c/href</a></iframe><p>We can split the terms of eq. (7) into two parts: <em>ω</em> = <em>ω’ </em>(“secular”), and <em>ω ≠ ω’ </em>(“non-secular”). The non-zero frequencies are order of qubit frequency<em> ω_q</em>, since it is constructed upon the difference of the qubit energy level. Compared to the time scale that we consider in the qubit evolution, which is <em>T1</em> and <em>T2</em>, usually:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4oVfVjV4aq1uhZPEhNy84h&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4oVfVjV4aq1uhZPEhNy84h" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/33276c171c20734206857440a9673d8d/href">https://medium.com/media/33276c171c20734206857440a9673d8d/href</a></iframe><p>Hence, the non-secular term oscillates rapidly within the timescale of our interest, which makes the contribution equal to zero, so we can drop all the non-secular terms. This is the secular approximation, and therefore the eq. (7) becomes;</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2cDD5sLEMRmYKXXqcxdiha&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2cDD5sLEMRmYKXXqcxdiha" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b2e1190bef2544420687104562141b72/href">https://medium.com/media/b2e1190bef2544420687104562141b72/href</a></iframe><p>Now we need to analyze the commutator of the eq. (8). Expand the commutator and its conjugate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvX7yqS5HNJstiXCUxhWoMY&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvX7yqS5HNJstiXCUxhWoMY" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/31326225925330649cb5294daa3d9c85/href">https://medium.com/media/31326225925330649cb5294daa3d9c85/href</a></iframe><p>Now let’s split Γ into Hermitian and anti-Hermitian parts by defining two new matrices:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvRCTyYrGjGRtmg3BzGRoKf&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvRCTyYrGjGRtmg3BzGRoKf" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d81b628183d64c60cac0ce79f18d977c/href">https://medium.com/media/d81b628183d64c60cac0ce79f18d977c/href</a></iframe><p>Both of them are Hermitian:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FfXi7mg8UeT8zh7oRtDxrvw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FfXi7mg8UeT8zh7oRtDxrvw" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0892ea6234363f24afb49f40cdd218c1/href">https://medium.com/media/0892ea6234363f24afb49f40cdd218c1/href</a></iframe><p>Substituting these matrices into the commutator gives:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FdcUPc5CV94PfsSUpCJvJJi&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FdcUPc5CV94PfsSUpCJvJJi" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8283fddbbc6aa9f95680a07648d4ce43/href">https://medium.com/media/8283fddbbc6aa9f95680a07648d4ce43/href</a></iframe><p>And the eq. (8) becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5xYMvkzsPSMLtYFFWkKtie&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5xYMvkzsPSMLtYFFWkKtie" width="964" height="234" frameborder="0" scrolling="no"><a href="https://medium.com/media/bf71e7d361bd772cae6360e7b06a0599/href">https://medium.com/media/bf71e7d361bd772cae6360e7b06a0599/href</a></iframe><p><em>H_LS</em> is the term that describes the Lamb shift, which is small renormalization of the qubit’s transition frequency. Since:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FrSUcK61xVrqgtAEM952XEm&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FrSUcK61xVrqgtAEM952XEm" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/39ac2bef50cd4f493939d32311f6ffec/href">https://medium.com/media/39ac2bef50cd4f493939d32311f6ffec/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6aoP2v8HNfbQs2K7dARmsD&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6aoP2v8HNfbQs2K7dARmsD" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0ce5ceffcfb65c73de3b7eecff59b3d5/href">https://medium.com/media/0ce5ceffcfb65c73de3b7eecff59b3d5/href</a></iframe><p>The Lamb shift Hamiltonian commutes with <em>H_S</em> and in practice simply absorbed into it.</p><p>Now we are almost done. Two facts finish the job. First, for each <em>ω</em>, the <em>γ_kl (ω)</em> is positive semi-definite (Appendix). Second, a positive semi-definite Hermitian matrix can be diagonalized:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fk81yUQtPYV1hdRqZVeVR1Q&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fk81yUQtPYV1hdRqZVeVR1Q" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/98923cf2028d6ee06093000b28c83ce2/href">https://medium.com/media/98923cf2028d6ee06093000b28c83ce2/href</a></iframe><p>where <em>U</em> is a unitary matrix. Then part of the second term of eq. (9) becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqJ8F74fHqNtDhSYJBaTyXh&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqJ8F74fHqNtDhSYJBaTyXh" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ef4faadad4ccaf78bf95716b43334f7e/href">https://medium.com/media/ef4faadad4ccaf78bf95716b43334f7e/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqWcb6VhWQpAcwdDeDvmvBm&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqWcb6VhWQpAcwdDeDvmvBm" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/9d25a4b416bfbfff760db5c228933428/href">https://medium.com/media/9d25a4b416bfbfff760db5c228933428/href</a></iframe><p>where:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FtUtx2CYJUyuUk3qq8n4G8D&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FtUtx2CYJUyuUk3qq8n4G8D" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e41143b3aca366a11cdb870887a24943/href">https://medium.com/media/e41143b3aca366a11cdb870887a24943/href</a></iframe><p>Using the same approach the second term of eq. (9) becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3gMH14Rxt5tvANgRVvgWC4&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3gMH14Rxt5tvANgRVvgWC4" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/fc5ce9ea51bf9e9308b69777dc7ceeef/href">https://medium.com/media/fc5ce9ea51bf9e9308b69777dc7ceeef/href</a></iframe><p>The double sum over <em>k, l</em> is now a single sum over <em>j</em>. Each eigenvector of <em>γ</em> is one independent channel <em>L_j</em>, and its eigenvalue is that channel’s rate. Finally, we merge (<em>ω, j</em>) into one channel index k, and absorb the coupling parameter into the rate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FekwbJAJJdKZmxCh8dfBBJE&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FekwbJAJJdKZmxCh8dfBBJE" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/57abe881baa26ef066b8123c3fd2b0f8/href">https://medium.com/media/57abe881baa26ef066b8123c3fd2b0f8/href</a></iframe><p>we get:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FiqpAWkovWKmG8ffuhNBciC&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FiqpAWkovWKmG8ffuhNBciC" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ab98f1140073c5d5db20a570a1fc20d0/href">https://medium.com/media/ab98f1140073c5d5db20a570a1fc20d0/href</a></iframe><p><strong>Back to the Schrödinger Picture</strong></p><p>Finally, we leave the interaction picture and return to the original Schrödinger picture. Inverting the rotation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Foz8Q4JFRDqd8SjUNZTfkhQ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Foz8Q4JFRDqd8SjUNZTfkhQ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/90a4424f0066a61892a2feccd0027d97/href">https://medium.com/media/90a4424f0066a61892a2feccd0027d97/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FwBKk8g7hWHPkry4CgVYE3h&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FwBKk8g7hWHPkry4CgVYE3h" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/69a1e81c05b6567ac2bfbd88bf5379ac/href">https://medium.com/media/69a1e81c05b6567ac2bfbd88bf5379ac/href</a></iframe><p>Inverting the relation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FdrKkPrEY4dMs9Q7BJYsT6y&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FdrKkPrEY4dMs9Q7BJYsT6y" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/298b1d5334285490f39b243b7ab2444d/href">https://medium.com/media/298b1d5334285490f39b243b7ab2444d/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fni3Nu6TfyecEnuFvaycGBw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fni3Nu6TfyecEnuFvaycGBw" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7dec7874f692386b477940fb8d06e131/href">https://medium.com/media/7dec7874f692386b477940fb8d06e131/href</a></iframe><p>Since <em>L</em> is a unitary transformation of <em>S(ω)</em>, it obeys the same rule:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7GrqLvKjFr6SCEmraDRAjb&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7GrqLvKjFr6SCEmraDRAjb" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/131cecead1a0cb85f2e87db99247bc31/href">https://medium.com/media/131cecead1a0cb85f2e87db99247bc31/href</a></iframe><p>Now we apply the transformation term-by-term in eq. (10).</p><p><em>i) Lamb-shift term</em></p><p>We have proven that <em>H_LS</em> commutes with <em>H_S</em>. So:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FpX3iZVHYz7DsvBPkuLGiuS&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FpX3iZVHYz7DsvBPkuLGiuS" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6ae868e8dfefc6a0bb512236bbb807bd/href">https://medium.com/media/6ae868e8dfefc6a0bb512236bbb807bd/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FmzcG86oVFr3Bk9gqz2FeqU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FmzcG86oVFr3Bk9gqz2FeqU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/075e391e381e06aafdc213231a0071b2/href">https://medium.com/media/075e391e381e06aafdc213231a0071b2/href</a></iframe><p><em>ii) First part of the second term in eq. (10)</em></p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FesoNu3K6ZvMZQwj6V9hLxs&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FesoNu3K6ZvMZQwj6V9hLxs" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/83b576e0f3f162f4706b1f5e9ab0edf6/href">https://medium.com/media/83b576e0f3f162f4706b1f5e9ab0edf6/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fu3dTPdStc7bJJoWpRmcusb&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fu3dTPdStc7bJJoWpRmcusb" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a28cd446f37f72546a0159b75b0dbf49/href">https://medium.com/media/a28cd446f37f72546a0159b75b0dbf49/href</a></iframe><p>Note that we use <em>ω_k, </em>since <em>k</em> had absorbed the degrees of freedom of <em>ω</em>.</p><p><em>iii) Anti-commutator term</em></p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FfxMcCQraXQJwc9Aui9N16q&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FfxMcCQraXQJwc9Aui9N16q" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/be2ec8fb43aa5ff66eadd07d434a1a67/href">https://medium.com/media/be2ec8fb43aa5ff66eadd07d434a1a67/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FcMsFYt8xSAKriF5g6Wwuoo&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FcMsFYt8xSAKriF5g6Wwuoo" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/cb54a71c1c32291af8b9349d0d3c567b/href">https://medium.com/media/cb54a71c1c32291af8b9349d0d3c567b/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkALXr4qokG8XrjjYUW2XsN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkALXr4qokG8XrjjYUW2XsN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/f9b923bd3a816e38b3e5531ab15dd121/href">https://medium.com/media/f9b923bd3a816e38b3e5531ab15dd121/href</a></iframe><p>Therefore, we finally arrive at the celebrated, Lindblad-Gorini-Kossakowski-Sudrashan equation, better known as the Lindblad master equation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fxn5pvXUs25CsqpggbeTY3R&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fxn5pvXUs25CsqpggbeTY3R" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d539c169bb43aabb6a3374b9e9dac890/href">https://medium.com/media/d539c169bb43aabb6a3374b9e9dac890/href</a></iframe><p>There are few things worth mentioning. The commutator represents a coherent evolution under the qubit Hamiltonian and its Lamb-shifted frequency. Each <em>L</em> is a jump operator, which is a channel through which the bath acts on the qubit. And because every <em>γ_k </em>is greater or equal to 0 (positive semi-definite), the evolution keeps the density matrix positive at all times. Now it became physically reasonable.</p><p>It was a truly long way to get here. What remains open is what the jump operators and rates are for a real qubit. The next column answers that for the two channels that dominate qubits, that is, the energy exchange and pure dephasing.</p><p><strong>Appendix</strong></p><p>Hermitian matrix is positive semidefinite when:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2sESf72z37SW9Te7F4Mwcc&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2sESf72z37SW9Te7F4Mwcc" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/445b5dab50bd224f7ef37131a2556cf4/href">https://medium.com/media/445b5dab50bd224f7ef37131a2556cf4/href</a></iframe><p>for every complex vector <em>v</em>. Take an arbitrary vector <em>v</em> and build the combined bath operator:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3pTGDHMJiYPLwgyU5cATiG&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3pTGDHMJiYPLwgyU5cATiG" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/683ea42bb34ea93ee84aa05745b8058d/href">https://medium.com/media/683ea42bb34ea93ee84aa05745b8058d/href</a></iframe><p>We will start in the time-domain, evaluate the result and Fourier transform it to the frequency domain:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9tU8Lxw4K3hJkLDVCwpFyj&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9tU8Lxw4K3hJkLDVCwpFyj" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5378b892a706609a7a36be262722fc13/href">https://medium.com/media/5378b892a706609a7a36be262722fc13/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FnJQmu7fgcuDe3j9YiXcyuu&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FnJQmu7fgcuDe3j9YiXcyuu" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/939a7cb0f053ceba1cb700ed2d107c6f/href">https://medium.com/media/939a7cb0f053ceba1cb700ed2d107c6f/href</a></iframe><p>Evaluate <em>f(τ)</em> in the bath’s eigenspace. Using:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9SUCmWLSeNy7jpJ2wWoctX&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9SUCmWLSeNy7jpJ2wWoctX" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d8f18926784f035061692d83bbdc9485/href">https://medium.com/media/d8f18926784f035061692d83bbdc9485/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fb6Sn4TBkCfqexemnbpA8CW&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fb6Sn4TBkCfqexemnbpA8CW" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/663bc0fac3817525d9e6603c106b648f/href">https://medium.com/media/663bc0fac3817525d9e6603c106b648f/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5QAQDSTbSR3rYJ7pjkKaGS&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5QAQDSTbSR3rYJ7pjkKaGS" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/9b7f8f8d075a6f276b61cc615fac0957/href">https://medium.com/media/9b7f8f8d075a6f276b61cc615fac0957/href</a></iframe><p>Using the property:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fty6t3y3wHDWXxUDLN7x8xd&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fty6t3y3wHDWXxUDLN7x8xd" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a158c1f3fb5c52ea8b8a3d23362aff02/href">https://medium.com/media/a158c1f3fb5c52ea8b8a3d23362aff02/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FnUb4YvRF2Sw5bSHBe2mtmo&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FnUb4YvRF2Sw5bSHBe2mtmo" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/171e6d693d1559686f31b57ab7d287d9/href">https://medium.com/media/171e6d693d1559686f31b57ab7d287d9/href</a></iframe><p>Every factor is non-negative, thus the <em>γ </em>is positive semi-definite.</p><p><strong>References</strong></p><ul><li><em>Daniel A. Lidar, “Lecture Notes on the Theory of Open Quantum Systems”</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=561d12d5b2a3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 2 — T1, T2, and the Bloch Equations]]></title>
            <link>https://medium.com/@grhncgzsvb/column-2-t1-t2-and-the-bloch-equations-c6ef1181a39e?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/c6ef1181a39e</guid>
            <category><![CDATA[quantum-computer]]></category>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <category><![CDATA[quantum-mechanics]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Wed, 05 Aug 2026 11:54:08 GMT</pubDate>
            <atom:updated>2026-08-05T23:42:24.798Z</atom:updated>
            <content:encoded><![CDATA[<p><em>One of the most widely used metrics to measure the performance of qubit systems is T1 and T2. This column introduces both through the geometry of the Bloch sphere. The result is the Bloch equations, which give the fundamental inequality between the two quantities.</em></p><p><strong>Quantum Information Decays</strong></p><p>When you leave a qubit alone — no drive, no gates — the quantum information the qubit was holding inevitably and irreversibly leaks away into the environment. The question is not whether it happens, but how fast — that is, how long the qubit can hold the quantum information. Two timescales govern this process: <em>T1</em>, the time for the qubit to lose energy, and <em>T2</em>, the time for the qubit to lose its phase coherence. Every performance metric in quantum hardware — such as gate fidelity, error correction thresholds — ultimately traces back to these two numbers.</p><p><strong>What the Bloch Sphere Tells Us About Decay</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OBlkH4virpWJjTwtIHDJaw.jpeg" /></figure><p>In column 1 we studied the Bloch vector representation. The Bloch vector of a pure state lies on the surface of the sphere, while that of a mixed state lies inside it. The central question of this column is where the qubit state ends up when we leave it alone, and how fast it decays.</p><p>T1 and T2 relaxation can be viewed as a shrinking of the Bloch sphere itself. There are exactly two ways the Bloch sphere can shrink. First, the transverse components <em>r_x</em> and <em>r_y</em> collapse while <em>r_z</em> is left untouched. We can imagine all points on the sphere being pulled inward toward the z-axis. This is pure phase decay: the qubit loses the information of its phase angle in the equatorial plane without any energy exchange.</p><p>Second, the entire sphere is pulled toward the north pole, which corresponds to a ground state. All three components of the Bloch vector are compressed, but asymmetrically — the longitudinal components <em>r_z</em> and the transverse components <em>r_x, r_y</em> are compressed at different rates. This is the energy decay: the qubit emits energy into the environment and relaxes toward the ground state.</p><p>These two geometric pictures are physically independent. One involves phase randomization without energy exchange, and the other involves energy exchange that affects both longitudinal and transverse components.</p><p><strong>The Symmetry Argument and the Bloch Equations</strong></p><p>The Bloch equations are the equation of motion of the Bloch vector that captures the phase/energy relaxation dynamics. They follow from a rotational symmetry along the z-axis of the qubit: no direction around the equator of the Bloch sphere is preferred.</p><p>Suppose the bath is unchanging in time and has no memory of the qubit in the past, so that the same rule governs the motion at every moment. The rate of change of each component of the Bloch vector can be any linear combination of the three components:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fv9rUibgCAV2YqaVgZBgNP2&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fv9rUibgCAV2YqaVgZBgNP2" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/821502222c369926af91cb0ca48c0dc9/href">https://medium.com/media/821502222c369926af91cb0ca48c0dc9/href</a></iframe><p>Here <em>A</em> is a real-valued, constant <em>3 x 3</em> array and <em>b</em> is a constant real vector. Both are independent of time and the current state. There are 12 free parameters in total. Eliminating 9 of them and consider only 3 is the whole story of the argument.</p><p>The z-component of the Bloch sphere tells us which state the qubit favours: it is the qubit’s energy coordinate. All points on the equatorial plane are an equal-weight superposition of <em>|0&gt;</em> and <em>|1&gt;</em>. The angle around the equator is the relative phase between them. This means rotating any Bloch vector with respect to the z-axis by arbitrary angle <em>ϕ</em> cannot change the equation of motion. This gives the requirement:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FshxK7o7EK8EwDN5u2QE8et&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FshxK7o7EK8EwDN5u2QE8et" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a03830684b92ff60a34bc7182e035a27/href">https://medium.com/media/a03830684b92ff60a34bc7182e035a27/href</a></iframe><p>where <em>R(ϕ)</em> is rotation matrix.</p><p>Let’s look at <em>b</em> first. A vector that is unchanged by rotation about <em>z</em> through every angle could only have z-component. So <em>b</em> becomes:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3CLPHcPwKF7vK1Bti1yzyN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3CLPHcPwKF7vK1Bti1yzyN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/151613ac07d7045fd93c17faa26aa69a/href">https://medium.com/media/151613ac07d7045fd93c17faa26aa69a/href</a></iframe><p>Now <em>A</em>. Because the symmetry treats the transverse plane and the z-axis differently, it would be better to separate them:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5Ytrp7ZiR5EJ2HgGmCdiey&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5Ytrp7ZiR5EJ2HgGmCdiey" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/901cf18ddaa4fa46bab63070802a12aa/href">https://medium.com/media/901cf18ddaa4fa46bab63070802a12aa/href</a></iframe><p>The <em>2 x 2</em> block <em>A_t</em> (<em>t</em> stands for transverse) governs how <em>r_x</em> and <em>r_y</em> interact. The two-component vector <em>u</em> and <em>v</em> are the cross-couplings: <em>u</em> says how <em>r_z</em> affects the transverse decay and <em>v</em> says how the transverse components affect the time derivative of <em>r_z</em>.</p><p><em>u</em> and <em>v</em> are vectors lying in the equatorial plane, and they must be unchanged through every rotation of the plane. Let:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FeQrVMr9K7BGeYZVCmpGEDs&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FeQrVMr9K7BGeYZVCmpGEDs" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/f1cfb2117e684a030a35e2bf454ddb6d/href">https://medium.com/media/f1cfb2117e684a030a35e2bf454ddb6d/href</a></iframe><p>Then:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fq3YLTPazRtSViee4Rkop5b&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fq3YLTPazRtSViee4Rkop5b" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/089c8d7fe448abbf220c678356738e47/href">https://medium.com/media/089c8d7fe448abbf220c678356738e47/href</a></iframe><p>Since <em>u</em> and <em>v</em> must be invariant under rotation, they cannot have any of the x, y component, thus both vanish. Physically, this means the rate at which the qubit loses energy cannot depend on the phase, and the rate at which it loses phase cannot depend on the energy.</p><p>Next, <em>A_t</em> satisfies:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FpHAQY49ABc5QrYanmDnT32&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FpHAQY49ABc5QrYanmDnT32" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/951603bcbd169f38cabe451d82b8f686/href">https://medium.com/media/951603bcbd169f38cabe451d82b8f686/href</a></iframe><p>So<em> A_t </em>commute with every rotation of the equatorial plane. Only two independent matrices satisfy this condition: the identity and the rotation by 90 degrees:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FfkhD3aBfcQbHmNgd8XUExC&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FfkhD3aBfcQbHmNgd8XUExC" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e10e59b42a6e91a8f908fd97f6feb77e/href">https://medium.com/media/e10e59b42a6e91a8f908fd97f6feb77e/href</a></iframe><p>The identity part shrinks <em>r_x</em> and <em>r_y</em> together at a rate <em>α</em>, and the J part rotates them together at a rate <em>Ω</em>. So the transverse motion can only be a linear combination of uniform contraction and precession. It is noteworthy that <em>r_x </em>and <em>r_y</em> shrink at the same rate. The directional bias within the equatorial plane cannot exist.</p><p>Putting all the results in the matrix and write <em>A</em> and <em>b </em>in explicit form:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FnJbSLEKqDqouGzWb5R4TL5&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FnJbSLEKqDqouGzWb5R4TL5" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/45b839e5251d97bd7886046e1b0b9d46/href">https://medium.com/media/45b839e5251d97bd7886046e1b0b9d46/href</a></iframe><p>Here the precession rate <em>Ω</em> describes the rotation of the Bloch sphere, not a contraction of it. If we set our reference frame to be the qubit’s frame (<em>Ω=0</em>), the three parameters remain. Setting them as:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FwkHGqPXqoe4Gc2tRTS9fw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FwkHGqPXqoe4Gc2tRTS9fw" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/170ff5d206c483f04c1d18bd49a8e450/href">https://medium.com/media/170ff5d206c483f04c1d18bd49a8e450/href</a></iframe><p>we get three differential equations:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FeW4xAFLR6g7UbHdsNveb2&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FeW4xAFLR6g7UbHdsNveb2" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/bcfbbb14f94b8914e1d359403e989714/href">https://medium.com/media/bcfbbb14f94b8914e1d359403e989714/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FpY9vJxgYUfA2CUhKJJz6uQ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FpY9vJxgYUfA2CUhKJJz6uQ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e8257bb0bee9b90aa0963ee71ef6c269/href">https://medium.com/media/e8257bb0bee9b90aa0963ee71ef6c269/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FggqBccDkLC8u35GvMHSPh3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FggqBccDkLC8u35GvMHSPh3" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/3053d70062b686bc9a9812cbd7dd177e/href">https://medium.com/media/3053d70062b686bc9a9812cbd7dd177e/href</a></iframe><p>When a qubit coupled to a thermal bath reaches the thermal equilibrium, the Bloch vector sits at a fixed point: <em>r_x = r_y = 0, r_z = r^eq, </em>where<em> r^eq</em> is determined by the bath temperature. Since a qubit coupled to a cold bath always favors the ground state in equilibrium, we can set <em>r^eq = |0&gt; = +1</em>.</p><p>The differential equations are solved as follows:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2shHgGQq2MZeTT5wYAmy5o&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2shHgGQq2MZeTT5wYAmy5o" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e2ee6e3d974af964da9b84bd32140996/href">https://medium.com/media/e2ee6e3d974af964da9b84bd32140996/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FhvhMeZ5bXNPcV9jb9o9NuP&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FhvhMeZ5bXNPcV9jb9o9NuP" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8f9d886e02ea79498683fc672e2c3ee0/href">https://medium.com/media/8f9d886e02ea79498683fc672e2c3ee0/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FiMmqNRT4DRHx4rLEhXcs1B&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FiMmqNRT4DRHx4rLEhXcs1B" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5b472acf5b6b093f773b8acb6fb30146/href">https://medium.com/media/5b472acf5b6b093f773b8acb6fb30146/href</a></iframe><p>These are the Bloch equations, in which T1 and T2 appear as two independent parameters. They describe entirely different physical processes, and it is worth being explicit about this.</p><p>T1 is an energy timescale. An energy quantum <em>ℏω</em> is emitted from the qubit into the environment. The qubit’s excited state population decays exponentially toward its thermal equilibrium at rate <em>1/T1</em>. In equilibrium the excited state population is essentially zero: the qubit ends up in the ground state.</p><p>T2 is a phase timescale. The qubit’s superposition maintains its amplitude but loses track of the relative phase between the two components <em>|0&gt;</em> and <em>|1&gt;</em>. <em>r_z</em> is unaffected but <em>r_x</em> and <em>r_y</em> decay to zero. The dephasing process is often driven by slowly fluctuating environmental fields (e.g., stray magnetic fields, charge fluctuations) that slightly shift the qubit frequency from shot to shot.</p><p><strong>The T2 ≤ 2T1 Inequality</strong></p><p>These two process — phase and energy relaxation — are not independent. T1 decay also causes T2 decay, because emitting an energy quantum from the excited state necessarily disrupts the superposition. The inequality T2 ≤ 2T1 precisely shows this.</p><p><strong>Step 1:</strong> Consider a qubit with only energy decay and no phase decay. Let’s consider <em>p = γdt</em> as the probability that <em>|1&gt;</em> falls to the ground state <em>|0&gt;</em>. Then the state of the qubit and its surrounding environment changes as follows:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fg39BNB8LkqzEHoybhbjiFv&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fg39BNB8LkqzEHoybhbjiFv" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7d994a8223599bc5bbcee5f2cbcb7aa1/href">https://medium.com/media/7d994a8223599bc5bbcee5f2cbcb7aa1/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FiZAfcrRwAnLKeRMFzZ7A52&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FiZAfcrRwAnLKeRMFzZ7A52" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c47bfb0ca3263b7c2205c8cc262b0151/href">https://medium.com/media/c47bfb0ca3263b7c2205c8cc262b0151/href</a></iframe><p>where |E_0&gt; is the initial state of the environment and |E_1&gt; is the state after the environment absorbed the qubit energy. Thus, an arbitrary qubit state evolves as follows:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F36Los6bhHxkcJTbGbun6vh&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F36Los6bhHxkcJTbGbun6vh" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6a35631d0385c8bcfe39da05aeddc376/href">https://medium.com/media/6a35631d0385c8bcfe39da05aeddc376/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FnVEn5GRbLmJ92aQsLMTrkc&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FnVEn5GRbLmJ92aQsLMTrkc" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/f85640ab33f92eddc7fa5aefef0da308/href">https://medium.com/media/f85640ab33f92eddc7fa5aefef0da308/href</a></iframe><p>Tracing out the environment factor of the total density matrix:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FpzzCEKG9N3iRupEXVf8pvm&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FpzzCEKG9N3iRupEXVf8pvm" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d6a3e6bbc3207465a258d96bb610457a/href">https://medium.com/media/d6a3e6bbc3207465a258d96bb610457a/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5krYung4xE2BeTKZHsLAMf&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5krYung4xE2BeTKZHsLAMf" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/cb6f4e52685cdf12bbc6a0c606b8abef/href">https://medium.com/media/cb6f4e52685cdf12bbc6a0c606b8abef/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqbBqre44SnPeyHYWFACiY8&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqbBqre44SnPeyHYWFACiY8" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d2ddfe11c6d8a29d1425d0ce251386e2/href">https://medium.com/media/d2ddfe11c6d8a29d1425d0ce251386e2/href</a></iframe><p>Let’s examine the excited state entry and coherence term:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FBevm72UADhb2DcfHQSk1e&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FBevm72UADhb2DcfHQSk1e" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/372851e8e87bd60727c64ecd9dbfffd8/href">https://medium.com/media/372851e8e87bd60727c64ecd9dbfffd8/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6enD567iVSTJDaX3jQ447j&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6enD567iVSTJDaX3jQ447j" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/1dac4e4a1647ed4b0a912650153f24bf/href">https://medium.com/media/1dac4e4a1647ed4b0a912650153f24bf/href</a></iframe><p>Thus:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fr1EhNFH7x3yNmpX8HLWxMy&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fr1EhNFH7x3yNmpX8HLWxMy" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e3000996ba5869abb4991ed8dd10cf58/href">https://medium.com/media/e3000996ba5869abb4991ed8dd10cf58/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkGYW563xgmo5ayWmatK2eR&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkGYW563xgmo5ayWmatK2eR" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/1da579451684604a8ad1172a7632f3c9/href">https://medium.com/media/1da579451684604a8ad1172a7632f3c9/href</a></iframe><p>Recall:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvmQzgKYCLg1qQNptkFpK83&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvmQzgKYCLg1qQNptkFpK83" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c8e18df576ae35546a9b10b4552bef88/href">https://medium.com/media/c8e18df576ae35546a9b10b4552bef88/href</a></iframe><p>Using this, we can convert the density matrix elements into Bloch vector components:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4F1svAUcNZx3jumGjLvdLr&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4F1svAUcNZx3jumGjLvdLr" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/543a9591364b33bd0d8d5a524633612d/href">https://medium.com/media/543a9591364b33bd0d8d5a524633612d/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvXYcPZd1p4CZy1E5YJEiKE&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvXYcPZd1p4CZy1E5YJEiKE" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/26e67a9aeaaffb9a440a8b4ba3087139/href">https://medium.com/media/26e67a9aeaaffb9a440a8b4ba3087139/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FsFZhb3u5G33YoJM7pt6ouC&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FsFZhb3u5G33YoJM7pt6ouC" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/62619929c817ea0481f0005ca8438615/href">https://medium.com/media/62619929c817ea0481f0005ca8438615/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FuHHbJRomrtsZKE5nwiWVQT&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FuHHbJRomrtsZKE5nwiWVQT" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8b0827c2da6eacb1dee8f04a63ebc4fa/href">https://medium.com/media/8b0827c2da6eacb1dee8f04a63ebc4fa/href</a></iframe><p>By simply replacing Re with Im, we ca obtain <em>r_y</em> in the same way as <em>r_x</em>. Therefore:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fwzs1YGhyTX8Eu8tZTofnRQ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fwzs1YGhyTX8Eu8tZTofnRQ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e5213da618159df5b657b72ad1be433f/href">https://medium.com/media/e5213da618159df5b657b72ad1be433f/href</a></iframe><p>In the absence of pure dephasing, coherences live exactly twice as long as populations.</p><p><strong>Step 2</strong>: Now let’s introduce additional phase noise that randomize the phase without exchanging energy. This adds an extra decay term to the transverse components:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5SUZQ8QEq7vnYrkrYCqPzj&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5SUZQ8QEq7vnYrkrYCqPzj" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7d17a6b371c0f416620fc0a3d814100e/href">https://medium.com/media/7d17a6b371c0f416620fc0a3d814100e/href</a></iframe><p>Here the <em>γ_</em>noise<em> ≥ 0 </em>always, since the environment can only add phase noise, never can remove it. Therefore, we get the final inequality:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4qbKERHH4XMpj543mL1dhb&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4qbKERHH4XMpj543mL1dhb" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ee408208a7d0c0bdede39ae5e4e2193d/href">https://medium.com/media/ee408208a7d0c0bdede39ae5e4e2193d/href</a></iframe><p><strong>Measuring T1 and T2</strong></p><p>Now we know what the T1 and T2 time represents. Then how can we measure it in actual quantum hardware?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*X_IkAGGzI1nhf1NwoUwYSQ.jpeg" /></figure><p>First, to measure T1, use the inversion-recovery sequence.</p><p>(1) Prepare the qubit in the excited state <em>|1&gt;, r_z (t=0) = -1</em></p><p>(2) Wait a time τ with no drive applied.</p><p>(3) Measure the probability of finding the qubit in <em>|1&gt;.</em></p><p>(4) Repeat for many <em>τ</em> and fit. Using the Bloch equation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkvNEEtMgcRNXRiVrkucN77&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkvNEEtMgcRNXRiVrkucN77" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/fe2401a0edc6ff96031265c69abc96f1/href">https://medium.com/media/fe2401a0edc6ff96031265c69abc96f1/href</a></iframe><p>The probability of finding |1&gt; is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F568y2zLGt6aQntv1Hh9wwU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F568y2zLGt6aQntv1Hh9wwU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2e6e5c426f8a452107dc3c7f0bc9066e/href">https://medium.com/media/2e6e5c426f8a452107dc3c7f0bc9066e/href</a></iframe><p>Sampling <em>P(τ)</em> at lots of <em>τ</em> and fitting to this exponential function gives T1.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PnP99KarRuqpFBFVIlB1wA.jpeg" /></figure><p>Next, to measure T2, use the Ramsey sequence.</p><p>(1) Apply a <em>π/2</em> pulse that rotates the Bloch vector from the north pole to the equator. Let’s say it rotates from <em>+z </em>→ <em>+x</em> (This depends on the choice of convention for the rotation axis).</p><p>(2) Allow free precession for time <em>τ. </em>The qubit precesses at the detuning <em>δ</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FuPwuJRmhp6zihzu53EdgeS&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FuPwuJRmhp6zihzu53EdgeS" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e57f3b60138cb4a96ce04ecaff3f32c9/href">https://medium.com/media/e57f3b60138cb4a96ce04ecaff3f32c9/href</a></iframe><p>which is the difference between the qubit frequency (intrinsic physical quantity of the qubit) and the drive frequency of the pulse (the frequency that we apply). The transverse amplitude decays:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F78LUUiTfngi1NCedTTw87J&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F78LUUiTfngi1NCedTTw87J" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/218c84560f2465668311a31bf89813b3/href">https://medium.com/media/218c84560f2465668311a31bf89813b3/href</a></iframe><p>(3) Apply a second <em>π/2</em> pulse that re-aligns the Bloch vector with z-axis. Bloch vector pointing <em>+x</em> will rotate to <em>-z</em>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6YmqjW6H1qfhwQvnu9nTBH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6YmqjW6H1qfhwQvnu9nTBH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6f0897ecace27e543f8edece152c0d76/href">https://medium.com/media/6f0897ecace27e543f8edece152c0d76/href</a></iframe><p>(4) Measure the probability of finding <em>|0&gt;</em>.</p><p>(5) Repeat for many <em>τ</em> and fit. The probability of finding <em>|0&gt;</em> is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FuhcfXqUEuGt9NHKsur9rMc&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FuhcfXqUEuGt9NHKsur9rMc" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6d17f30e9508de16156d23364e640078/href">https://medium.com/media/6d17f30e9508de16156d23364e640078/href</a></iframe><p>As a bonus, the oscillation frequency gives <em>δ</em> which we can use in calibration of the qubit frequency.</p><p><strong>T2 versus T*2</strong></p><p>In a real experiment, the Ramsey measurement does not give a simple exponential. The qubit frequency can fluctuate from run to run; for instance, due to slowly drifting background fields or defect charges. So each measurement sequence sees slightly different values of <em>ωq</em>. When we average out the results, the phase contributions from different runs partially cancel, and the envelope decays faster than true T2. This faster decay is characterized by the inhomogeneous dephasing time T*2.</p><p>However, if we insert a <em>π </em>pulse at the halfway point of the free precession, we can contribution from slowly varying noise, because the evolution before the <em>π </em>pulse is time-reversed by the evolution after it. The sequence thus recovers the true T2, and we call this the Hahn echo sequence. Here, T*2 ≤ T2 always.</p><p><strong>What the Bloch Equations Cannot Tell Us</strong></p><p>The Bloch equations are powerful because they require no microscopic model. But this is also their limitation: they tell us what T1 and T2 means, but not the actual values. To understand this, we need a model of how qubit couples to the environment. Throughout the next two columns, we will discuss about this topic.</p><p><strong>References</strong></p><ul><li><em>D. A. Lidar, “Lecture Notes on the Theory of Open Quantum Systems,” arXiv:1902.00967 (2019)</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c6ef1181a39e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 1 — The Density Matrix]]></title>
            <link>https://medium.com/@grhncgzsvb/column-1-the-density-matrix-1ac064f8e8e3?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/1ac064f8e8e3</guid>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[quantum-computer]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Sat, 01 Aug 2026 09:54:05 GMT</pubDate>
            <atom:updated>2026-08-01T09:54:05.922Z</atom:updated>
            <content:encoded><![CDATA[<p><em>An independent qubit can be described by a straightforward geometric object called the Bloch sphere. However, when the qubit interacts with its surrounding environment, we cannot simply assign a wavefunction. Instead, we must use the density matrix, which encodes the probability of the qubit being in certain state and the coherence between the basis states. Together, these tools are the foundation on which every qubit hardware system will rest.</em></p><p><strong>The Bloch sphere</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*uYVeQ03Ioo4ZsbwLFzXx3w.jpeg" /></figure><p>Any qubit system could be modeled as a two-level system: system consists of two basis states, namely <em>|0&gt;</em> and <em>|1&gt;</em>. Its state is characterized by the angles <em>θ</em> and <em>ϕ</em>, which describes inclination toward a specific state and its relative phase, respectively. This qubit lives on a sphere called Bloch sphere. Given any pure state:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FnWSCuamd8yYU5FLc1Vy6kY&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FnWSCuamd8yYU5FLc1Vy6kY" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4ac270e50d059add334bb451ef377316/href">https://medium.com/media/4ac270e50d059add334bb451ef377316/href</a></iframe><p>the two angles <em>θ</em> and <em>ϕ</em> indicates the direction of a state on the surface of the Bloch sphere. All the information about a pure qubit is encoded in a single geometric direction. The problem is that this picture assumes that we could write down a wavefunction of the qubit in the first place. When the qubit interacts with an external environment that we cannot observe (e.g., neighboring atoms, thermal radiation), it is impossible. The qubit no longer lives on the surface of the Bloch sphere. To describe this situation, we need a new mathematical object.</p><p><strong>The Density Matrix</strong></p><p>Think about a realistic situation: we know our system is in one of several pure states, but not sure which. Maybe our state-preparation procedure sometimes produces <em>|0&gt; </em>or sometimes <em>|+&gt; =</em> (<em>|0&gt;+|1&gt;)/√2</em>. This is not a quantum superposition; it is classical ignorance. The qubit is definitely in one state, but we just don’t know which. The density matrix encodes this ensemble as a weighted sum:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2LxDtGZTH7N2hUANKpqvB5&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2LxDtGZTH7N2hUANKpqvB5" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/bb2bd52c3db593951d9ab41845cfa716/href">https://medium.com/media/bb2bd52c3db593951d9ab41845cfa716/href</a></iframe><p>where <em>p_i</em> are the classical probabilities that sum up to 1.</p><p>Writing everything here in the <em>|0&gt;</em> and <em>|1&gt; </em>basis, it produces an explicit 2 × 2 matrix, whose diagonal entries are classical probabilities and off-diagonals are quantum coherence between those states.</p><p>There are some of the properties of the density matrix that deserve mentioning. First, since sum of the classical properties must equal to 1, the trace always equals to 1:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FimAydgHaHphcxUnWfmWemf&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FimAydgHaHphcxUnWfmWemf" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/23724dc86a89e4e3bfb2d4fa1ac6957f/href">https://medium.com/media/23724dc86a89e4e3bfb2d4fa1ac6957f/href</a></iframe><p>Second, since classical probability is always a real number, the density matrix is Hermitian:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F36Jtj5VExjaYsdpX34wWn1&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F36Jtj5VExjaYsdpX34wWn1" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c5851bde7e988c3fc141bd96609c4948/href">https://medium.com/media/c5851bde7e988c3fc141bd96609c4948/href</a></iframe><p>Third, <em>ρ</em> is positive semi-definite: for any vector |<em>ϕ&gt;</em>, the expectation value is not smaller than 0:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FtzSDv88twbA3oFJqn1VBzE&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FtzSDv88twbA3oFJqn1VBzE" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5cbd3e953551daaf0f1cb371cc0d257f/href">https://medium.com/media/5cbd3e953551daaf0f1cb371cc0d257f/href</a></iframe><p><strong>Purity and the Expectation Value</strong></p><p>When the classical ignorance of a qubit disappears — when we know the qubit is definitely in the state <em>|ψ&gt;</em> — then the density matrix has a single term, and its square equals itself:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FwUdJmTKA3fgXvtxbgrEDzT&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FwUdJmTKA3fgXvtxbgrEDzT" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/541f8a37c48616b7ab0d39c2c29e2e6a/href">https://medium.com/media/541f8a37c48616b7ab0d39c2c29e2e6a/href</a></iframe><p>We call this a pure state. We can quantify the purity of a single qubit state as:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FnyDfHkLrcYLNHrJhHBUyBi&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FnyDfHkLrcYLNHrJhHBUyBi" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d7eb8c65f581213979293e4d119eb865/href">https://medium.com/media/d7eb8c65f581213979293e4d119eb865/href</a></iframe><p>For a pure state:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FsfdvyV4SXmpTFXztsNr5Ee&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FsfdvyV4SXmpTFXztsNr5Ee" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/36878ab9f2dde2e650d92128aa148bd5/href">https://medium.com/media/36878ab9f2dde2e650d92128aa148bd5/href</a></iframe><p>and for any mixed state, <em>P </em>&lt; 1. For a <em>d</em>-dimensional system, the minimum purity is <em>P = 1/d</em> at the maximally mixed state <em>ρ = I / d</em>. For a qubit (<em>d</em> = 2), minimum is 1/2.</p><p>How can we extract the expectation value of an observable using density matrix? For an ensemble system, we can write:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FoNEhEU2fUdhBtaYJm2F8zY&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FoNEhEU2fUdhBtaYJm2F8zY" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/9c552784b989d8a35173fa237f811ff2/href">https://medium.com/media/9c552784b989d8a35173fa237f811ff2/href</a></iframe><p>Using the equality:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FufRSJeka1ySYanDASdF8Xi&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FufRSJeka1ySYanDASdF8Xi" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/75c1448110a5b70f05239417f223735e/href">https://medium.com/media/75c1448110a5b70f05239417f223735e/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FfMQrhgAXHsN1MRvozDoTTH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FfMQrhgAXHsN1MRvozDoTTH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e03095f2984b8b24c045bc180dcd2e40/href">https://medium.com/media/e03095f2984b8b24c045bc180dcd2e40/href</a></iframe><p>Any quantum mechanical prediction that could be made about the system is encoded in <em>ρ</em> and extractable using this trace formula.</p><p><strong>The Bloch Vector</strong></p><p>Let’s go back to the Bloch sphere discussion. Any qubit density matrix can be expanded in the basis of identity matrix and Pauli matrices:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FwriWZhyuaLDe2RTPWwjkwk&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FwriWZhyuaLDe2RTPWwjkwk" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e18296692163d4949a750e0c8a987df1/href">https://medium.com/media/e18296692163d4949a750e0c8a987df1/href</a></iframe><p>where <em>r</em> is the Bloch vector. Explicitly in matrix form:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7sMLQGxkFaAB8hqzWMRaNN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7sMLQGxkFaAB8hqzWMRaNN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/1e5d974d225a8f1ba0ef8d95bf4b6eeb/href">https://medium.com/media/1e5d974d225a8f1ba0ef8d95bf4b6eeb/href</a></iframe><p>where the diagonal entries are the populations of |0&gt; and |1&gt; states, and the off-diagonal entries are the coherences.</p><p>Eigenvalues of the matrix above are:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fae2fM6ga5LoxTZ86RVxKMW&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fae2fM6ga5LoxTZ86RVxKMW" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8e6cfe01384b7bb04ae16e9c3890bb5b/href">https://medium.com/media/8e6cfe01384b7bb04ae16e9c3890bb5b/href</a></iframe><p>Since the eigenvalues must be non-negative, the norm of the Bloch vector <em>||r|| </em>≤ 1. Thus, the Bloch vector lives within the Bloch unit sphere.</p><p>If |<em>|r|| </em>= 1, the Bloch vector lies within the surface of the sphere and <em>λ</em> = 1. This corresponds to pure states. If |<em>|r||</em> &lt; 1, this corresponds to mixed states with two non-zero eigenvalues. If |<em>|r||</em> = 0, the Bloch vector points the center of the sphere, and this corresponds to maximally mixed state. We can also put this in another way:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4enyegfPbq9EZGa4QcDBy1&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4enyegfPbq9EZGa4QcDBy1" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4cffcedbe54a5995b7f4f501f2025c50/href">https://medium.com/media/4cffcedbe54a5995b7f4f501f2025c50/href</a></iframe><p>A unit Bloch vector gives <em>P</em> = 1, a shorter Bloch vector gives <em>P</em> &lt; 1. Therefore, the modulus of the Bloch vector is a measure of how pure the qubit state is.</p><p><strong>Partial Trace</strong></p><p>So far, we have studied examples where the uncertainty of the qubit state originates internally: it can be expressed using only the qubit basis |0&gt; and |1&gt;. Now let’s look at more subtle situation: the mixed state that arises from entanglement with an unobserved system.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*g7kPHoHApTIbRR3sp1n0EQ.jpeg" /></figure><p>Let’s think of a two-qubit system where qubit A is the system of interest and qubit B is the environment (anything beyond our control). The total system A+B evolves under the Schrödinger equation and has a well-defined density matrix <em>ρ_AB.</em> If the total density matrix is a simple tensor product:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fp9XTLXrE2tP7pV4a1nwiR2&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fp9XTLXrE2tP7pV4a1nwiR2" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a99921c860a437c7077a8791b1851b8d/href">https://medium.com/media/a99921c860a437c7077a8791b1851b8d/href</a></iframe><p>two sub-systems are uncorrelated, and qubit A is in a pure state. But generically, <em>ρ_AB </em>cannot be simply factorized. To figure out the state A alone, we need to average out all of the degrees of freedom of B.</p><p>We can do this through the operation called partial trace. The partial trace of B leaves the A matrix unchanged and traces over the B factor:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5Fs8vKEETG92GvSRtXJU5U&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5Fs8vKEETG92GvSRtXJU5U" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/140f3e5b4160a3fbddde66d8e3ee4112/href">https://medium.com/media/140f3e5b4160a3fbddde66d8e3ee4112/href</a></iframe><p>If we apply it on the <em>ρ_AB</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FZqjX4BQrhc57jJMB31Le3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FZqjX4BQrhc57jJMB31Le3" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/aead8fd9f1e5c4f9ab72282385cb4361/href">https://medium.com/media/aead8fd9f1e5c4f9ab72282385cb4361/href</a></iframe><p>where <em>ρ_A</em> is called the reduced density matrix of A. Expectation value of any observable acting on A can be computed using the reduced density matrix:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fn5AHazUFMPimTUwrtg8GpV&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fn5AHazUFMPimTUwrtg8GpV" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/1cb70305fa5b7bd8ceca68dc64e57b17/href">https://medium.com/media/1cb70305fa5b7bd8ceca68dc64e57b17/href</a></iframe><p>To explicitly look how the partial trace acts on the total density matrix, first we apply it to a simple product state:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FmAFsSUedsJc125CLircCN3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FmAFsSUedsJc125CLircCN3" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a26521bcbc7b0690d51ca764b96f6035/href">https://medium.com/media/a26521bcbc7b0690d51ca764b96f6035/href</a></iframe><p>which is trivial, because tracing out an unentangled partner cannot affect the system of interest.</p><p>Now consider the maximally entangled sate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4FXcYigzTrFoEhWtZri4Jd&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4FXcYigzTrFoEhWtZri4Jd" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/48623aece9bde234f9506f17745b14a1/href">https://medium.com/media/48623aece9bde234f9506f17745b14a1/href</a></iframe><p>Applying partial trace of B on the total density matrix:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fkm4P1wnaHZvdDRfU9GzdXU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fkm4P1wnaHZvdDRfU9GzdXU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/9e5147dcb38dcdfc15d4fef5ddd99e22/href">https://medium.com/media/9e5147dcb38dcdfc15d4fef5ddd99e22/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FaM6sZRGHc9sBSrpT9SGSs8&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FaM6sZRGHc9sBSrpT9SGSs8" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/3164d381c3914c44913f0c07db10827d/href">https://medium.com/media/3164d381c3914c44913f0c07db10827d/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqMGudNa7YNmSSrpm1nJEH3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqMGudNa7YNmSSrpm1nJEH3" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2d622b12f81e9b9cb9f6c215937ba4b7/href">https://medium.com/media/2d622b12f81e9b9cb9f6c215937ba4b7/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FeqMUvZUtw3hSkVbauwLozN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FeqMUvZUtw3hSkVbauwLozN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7038d9419732a90aabd7adef401ecd7b/href">https://medium.com/media/7038d9419732a90aabd7adef401ecd7b/href</a></iframe><p>We can see that the reduced density matrix of qubit A is the maximally mixed state. So the global two-qubit state was a pure state, however the qubit A alone cannot be expressed as a single wavefunction, no matter what basis or phase you choose. The density matrix is the only correct description.</p><p>Let’s wrap up. We have covered the foundational mathematical object to describe qubit and surrounding environment. What we lack is the description of how the density matrix changes when the qubit is immersed in its environment — when the Bloch vector shrinks and rotates due to leakage of energy and degradation of phase coherence. In the next column, we will cover this more systematically.</p><p><strong>References</strong></p><ul><li><em>D. A. Lidar, “Lecture Notes on the Theory of Open Quantum Systems,” arXiv:1902.00967 (2019)</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1ac064f8e8e3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 7 — Berry Phase in The Graphene System]]></title>
            <link>https://medium.com/@grhncgzsvb/column-7-berry-phase-in-the-graphene-system-8ad3195d68df?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/8ad3195d68df</guid>
            <category><![CDATA[condensed-matter-physics]]></category>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[quantum-physics]]></category>
            <category><![CDATA[physics]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Wed, 29 Jul 2026 00:53:13 GMT</pubDate>
            <atom:updated>2026-07-29T09:30:30.606Z</atom:updated>
            <content:encoded><![CDATA[<p><em>So far, we have discussed the Berry phase formalism and how it manifests as physical quantities in generic crystal systems. Now we will narrow our focus to a single sheet of carbon atoms: graphene.</em></p><p>In 2005, Geim, Novoselov, and other physicists placed a flake of graphene — a single atomic layer of carbon — in a strong magnetic field and measured its Hall conductance. What they found was so anomalous that it contributed directly to a Nobel Prize five years later. Instead of the integer sequence the ordinary quantum Hall effect produces, the Hall conductance of graphene stepped through …, -2, +2, +6, +10, … in units of conductance quantum <em>e² / h</em>. In this column, we will discuss the Berry phase formalism in the graphene system, which is at the very heart of this anomalous phenomenon.</p><p><strong>Two Atoms Per Unit Cell</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*le-hgpQlc7pNQ7ghMZckTA.jpeg" /></figure><p>Graphene is a single atomic layer of carbon atoms, forming the honeycomb lattice. It seems like a simple triangular lattice, but actually it is much more than that. The honeycomb lattice cannot be described by a single Bravais lattice; it requires a two-atom basis. Each unit cell contains two carbon atoms sitting on inequivalent sites, conventionally called sublattice A and B. Each carbon atoms has three nearest neighbours, all of them on the other sublattice.</p><p>Carbon atoms have four valence electrons, and three of them form strong covalent bonds with nearest neighbours: these are called σ bonds, caused by σ orbitals. That leaves one electron per atom alone, sitting in the <em>p_z</em> orbital that perpendicularly sticks out of the graphene plane: this is called π orbital. These π orbitals are the one that dominates the low-energy physics of the graphene.</p><p><strong>The Dirac Cone</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/904/1*QgiAOACBVyYGTouEJqEysw.jpeg" /></figure><p>Now let’s mathematically analyze the graphene lattice. We set the distance of the nearest neighbour as a_0 ( ≈ 1.42 Å) and express the primitive lattice vector:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FfCsRUwchqqpqjRgKkT3UgH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FfCsRUwchqqpqjRgKkT3UgH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/50e1bf354d94945835baf2cdad1e8320/href">https://medium.com/media/50e1bf354d94945835baf2cdad1e8320/href</a></iframe><p>The nearest neighbours of the sublattice A are:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FpUPRiDg9pHw7G1TiXfqAJY&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FpUPRiDg9pHw7G1TiXfqAJY" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4449494a9c2b49adf8ee463471b56cbd/href">https://medium.com/media/4449494a9c2b49adf8ee463471b56cbd/href</a></iframe><p>Constructing the reciprocal lattice vectors:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2MUkdnGQkjxu7JCTFKptmS&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2MUkdnGQkjxu7JCTFKptmS" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/eba47223774df6c4142684754f3ff868/href">https://medium.com/media/eba47223774df6c4142684754f3ff868/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F8eV19b3Ae1J55hYBMZAnGj&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F8eV19b3Ae1J55hYBMZAnGj" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ac380cdba9d47eeb1d0eaf38a3cc73bb/href">https://medium.com/media/ac380cdba9d47eeb1d0eaf38a3cc73bb/href</a></iframe><p>We can define a hexagonal Brillouin zone. The corner of the hexagon zone is called K, K’ valley:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fb175sKRNUMrfLH1EkvAXX8&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fb175sKRNUMrfLH1EkvAXX8" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c5ad4f1858f7c745d867ff1c630de67d/href">https://medium.com/media/c5ad4f1858f7c745d867ff1c630de67d/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9PMNFdzJgmshaSzhmDjRuf&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9PMNFdzJgmshaSzhmDjRuf" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/68d15f533aca13c66c01b6b42b937f62/href">https://medium.com/media/68d15f533aca13c66c01b6b42b937f62/href</a></iframe><p>While a hexagon has six corners, only two of them are inequivalent. Any other corners can be expressed using a combination of reciprocal lattice vectors.</p><p>Setting <em>|i&gt;</em> as the π state from <em>i</em>-th atom in sublattice A, we can construct the tight-binding Hamiltonian:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7GgvnuFTqn5yMHfMpa25Mh&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7GgvnuFTqn5yMHfMpa25Mh" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/3006d9385118c57608a899a1592eb5c6/href">https://medium.com/media/3006d9385118c57608a899a1592eb5c6/href</a></iframe><p>where <em>t</em> is a nearest-neighbour hopping energy (≈ 2.8 eV). Since we have two atoms per unit cell, there are two different states for momentum <em>q</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FiKQWvK1cmEnrzgqM1rzLgg&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FiKQWvK1cmEnrzgqM1rzLgg" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d2148b58f5e7a66d9fe49428249dc3df/href">https://medium.com/media/d2148b58f5e7a66d9fe49428249dc3df/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FsUDEGRn8Tj1qYs55HLZBgy&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FsUDEGRn8Tj1qYs55HLZBgy" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/155be47e2d3d9c72744113a017b3d197/href">https://medium.com/media/155be47e2d3d9c72744113a017b3d197/href</a></iframe><p>Applying Fourier transform again, we get</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FccwnyJSxYX9F5Xrcbj863s&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FccwnyJSxYX9F5Xrcbj863s" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5f328ea9818ea4cf4bf6c0c98a89e65b/href">https://medium.com/media/5f328ea9818ea4cf4bf6c0c98a89e65b/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fkmxv6kGiMpxWFLudr1CLve&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fkmxv6kGiMpxWFLudr1CLve" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6f66f614eb2512a5f99963a47925aef0/href">https://medium.com/media/6f66f614eb2512a5f99963a47925aef0/href</a></iframe><p>Putting this to Hamiltonian above,</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FuoEyWATLappgtGEBgKwXqL&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FuoEyWATLappgtGEBgKwXqL" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a961ef7639ccd05b84140023cf7f5fda/href">https://medium.com/media/a961ef7639ccd05b84140023cf7f5fda/href</a></iframe><p>Using the property:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9H6n3vDsWGPveqXkjvsGhZ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9H6n3vDsWGPveqXkjvsGhZ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/eed18112e99fb4ad05d002da02cbc8ac/href">https://medium.com/media/eed18112e99fb4ad05d002da02cbc8ac/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FjsBrc1pbJX8y7q5aMuch5L&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FjsBrc1pbJX8y7q5aMuch5L" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/3f9b115e99417d1468ce22f9f9981a1c/href">https://medium.com/media/3f9b115e99417d1468ce22f9f9981a1c/href</a></iframe><p>Define</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5PHyaX2P6Y4J6W1o1nEPxH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5PHyaX2P6Y4J6W1o1nEPxH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ba74a6345f675d4f8cc06c928b8faeb1/href">https://medium.com/media/ba74a6345f675d4f8cc06c928b8faeb1/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FhxMbnzYnMUDv2T66mUsGZK&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FhxMbnzYnMUDv2T66mUsGZK" width="474" height="159" frameborder="0" scrolling="no"><a href="https://medium.com/media/aafac1827937ec1c35af8ec2424425a8/href">https://medium.com/media/aafac1827937ec1c35af8ec2424425a8/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F235JAqJ95f32gxQSzicEmU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F235JAqJ95f32gxQSzicEmU" width="494" height="167" frameborder="0" scrolling="no"><a href="https://medium.com/media/6c8e6228e5a66ae3c15db5e30f174fff/href">https://medium.com/media/6c8e6228e5a66ae3c15db5e30f174fff/href</a></iframe><p>If we solve for the eigenvalue of Hamiltonian:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fvzk2A5dLke9DYXzPSSqF87&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fvzk2A5dLke9DYXzPSSqF87" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/029729b5972319393098e5ab4aa97578/href">https://medium.com/media/029729b5972319393098e5ab4aa97578/href</a></iframe><p>Putting <em>K, K’</em> in <em>q</em>, we can easily find:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkT52fBAeL5A9RrtHUhZh8j&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkT52fBAeL5A9RrtHUhZh8j" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/109a8c4aae8b2a9039d392675a78dc88/href">https://medium.com/media/109a8c4aae8b2a9039d392675a78dc88/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FsVfPRtWH9u853W2uEaMepw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FsVfPRtWH9u853W2uEaMepw" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2784451d5be26af6dac172179a29ea41/href">https://medium.com/media/2784451d5be26af6dac172179a29ea41/href</a></iframe><p>It means that the conduction and valence bands are degenerate at the two points. We call these points the “Dirac points”. Expanding energy eigenvalue near K:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4d6i6xFJivMZ3vUj5nbx6F&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4d6i6xFJivMZ3vUj5nbx6F" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/1440c92c4dbd76ac7a5b232c0de1f10a/href">https://medium.com/media/1440c92c4dbd76ac7a5b232c0de1f10a/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FiCrgaq5LhpTmGBLAdDAt3e&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FiCrgaq5LhpTmGBLAdDAt3e" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a1ebeeb99474e027aecf8a2076602670/href">https://medium.com/media/a1ebeeb99474e027aecf8a2076602670/href</a></iframe><p>We can define the Fermi velocity as:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fsr8DhfSW36vYHhHGZYBwVB&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fsr8DhfSW36vYHhHGZYBwVB" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/dd7f809707f8e42d940d0699ef27e2b0/href">https://medium.com/media/dd7f809707f8e42d940d0699ef27e2b0/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FpdjLMwwY3Yh8zWFjL7E4FN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FpdjLMwwY3Yh8zWFjL7E4FN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2911043b9650be9989f3f3747c50268a/href">https://medium.com/media/2911043b9650be9989f3f3747c50268a/href</a></iframe><p>Thus, the dispersion is linear near the Dirac points (the same result is obtained near K’ too) and resembles that of massless Dirac fermions where the speed of light is replaced by the Fermi velocity. The negative energy eigenvalue has a concrete meaning. In the context of the tight-binding model, negative-energy states are valence-band states. The Dirac sea is the filled valence band. This will matter when we talk about the Landau levels.</p><p>Near the Dirac points, we can reformulate the Hamiltonian. For <em>K</em> valley:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FxkSCSEZ2epzVTZFJRM3mQk&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FxkSCSEZ2epzVTZFJRM3mQk" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4bf4da297d1ad08b44f600bd85859731/href">https://medium.com/media/4bf4da297d1ad08b44f600bd85859731/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FXCeBxnfTJjown8XG9gwqU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FXCeBxnfTJjown8XG9gwqU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5f2188a567ecd613bf7bcae2e0f40f6d/href">https://medium.com/media/5f2188a567ecd613bf7bcae2e0f40f6d/href</a></iframe><p>We can perform a similar analysis near K’:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9YCTchJks9jSMZUDyqf9VT&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9YCTchJks9jSMZUDyqf9VT" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5ef54c52ddc2c1eadc26033ce43a3e1c/href">https://medium.com/media/5ef54c52ddc2c1eadc26033ce43a3e1c/href</a></iframe><p>The two effective Hamiltonians can be unified by introducing a valley degree-of-freedom index:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqPmaJ1yJjEHmbd8zv5tGgr&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqPmaJ1yJjEHmbd8zv5tGgr" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6a349167b44b12a88371a0e15d6e3f3b/href">https://medium.com/media/6a349167b44b12a88371a0e15d6e3f3b/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkpPZ3JWNSnFqv8juDn9ggc&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkpPZ3JWNSnFqv8juDn9ggc" width="338" height="137" frameborder="0" scrolling="no"><a href="https://medium.com/media/4513dd6e909708cb2c3ec423ee9f87d5/href">https://medium.com/media/4513dd6e909708cb2c3ec423ee9f87d5/href</a></iframe><p>One critical point must be made explicitly before going further: the Pauli matrices in <em>H_K,K’</em> do not act on the electron’s real spin. They act on the sublattice indices A, B (it becomes clear when you look at the original version of Hamiltonian where the matrix elements are explicit). This is the pseudospin: a two-component degree of freedom arising from the two-atom unit cell. Real electron spin has not entered the discussion yet.</p><p><strong>The Berry Phase in Graphene</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/890/1*j7aRbe2EehF8PWAFOfjPAA.jpeg" /></figure><p>What Berry phase does an electron accumulate when its crystal momentum traces a closed loop around the Dirac point? The eigenstates of <em>H_K</em> are two component spinors. In the case of positive energy-eigenstate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fh5yn6HymmFnYGHkTGHCNVs&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fh5yn6HymmFnYGHkTGHCNVs" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/fa27e52fad2bd51a2d38ab5d13771f63/href">https://medium.com/media/fa27e52fad2bd51a2d38ab5d13771f63/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5ff6RdzgWB58RzpZGQ9LCn&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5ff6RdzgWB58RzpZGQ9LCn" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ce91628577b830846592a8c1c20460e6/href">https://medium.com/media/ce91628577b830846592a8c1c20460e6/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3XFVAuJC9ExnRjKZbe2AZH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3XFVAuJC9ExnRjKZbe2AZH" width="290" height="170" frameborder="0" scrolling="no"><a href="https://medium.com/media/ac5ab6a38e96f5d01a5c1564b20fa674/href">https://medium.com/media/ac5ab6a38e96f5d01a5c1564b20fa674/href</a></iframe><p>Now see what happens when <em>k</em> completes a full loop around the circle enclosing the Dirac point. The angle increases from 0 -&gt; 2π. Each component of the spinor becomes -1 after the full rotation. The wavefunction returns to itself with minus sign. Thus, we can say the geometric phase π is accumulated, and the Dirac point is the source.</p><p>Let’s consider what the Berry curvature looks like across the two valleys. The Berry curvature near K is nonzero and peaks at the Dirac point. It is concentrated at the tip of the cone, like a magnetic monopole. Near K’, the Berry curvature is equal in magnitude but opposite in sign. Time-reversal symmetry maps K to K’ and reverses the sign of the Berry curvature (recall K’ = -K), so the two cones carry opposite contributions. It means when you integrate the Berry curvature over the entire Brillouin zone, the contributions from K and K<em>’</em> valley cancel exactly, and the total is zero. This is why pristine, undoped graphene has no anomalous Hall effect: no Berry curvature, no net anomalous velocity. It requires a magnetic field to break this time-reversal symmetry, which leads to breaking this cancellation and produce a Hall signal. The integer quantum Hall effect of graphene is the signal.</p><p><strong>Klein Tunneling</strong></p><p>In undergraduate quantum mechanics course, we learn about what happens when the wavefunction meets the finite potential barrier. A wave of energy <em>E</em> meeting the barrier of height <em>V_0 &gt; E</em> leaks in as an evanescent tail that decays exponentially. A Dirac electron in graphene does the opposite — it crosses an arbitrary barrier with probability exactly one.</p><p>Recall the K-valley Hamiltonian. We can define the chirality (or helicity) operator as the pseudospin projected onto the direction of motion:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FjAfCkxFQEHkaAm6wqHNfeZ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FjAfCkxFQEHkaAm6wqHNfeZ" width="200" height="157" frameborder="0" scrolling="no"><a href="https://medium.com/media/96b2c301160058287d58815025a4e8d8/href">https://medium.com/media/96b2c301160058287d58815025a4e8d8/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9aiEYr1FzbHocEMKJDWk4w&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9aiEYr1FzbHocEMKJDWk4w" width="223" height="138" frameborder="0" scrolling="no"><a href="https://medium.com/media/6d97d6fbb4c8f3f324c3db87b9769dcf/href">https://medium.com/media/6d97d6fbb4c8f3f324c3db87b9769dcf/href</a></iframe><p>Since the Hamiltonian is constant times the chirality operator, its energy eigenstates are automatically eigenstates of the chirality with eigenvalue ±1. The conduction band state,</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvashNSR18JxZoeb87gKSrF&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvashNSR18JxZoeb87gKSrF" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c596f1e19b746bd674f9997bc782f2ba/href">https://medium.com/media/c596f1e19b746bd674f9997bc782f2ba/href</a></iframe><p>has pseudospin</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FtfJVtrxvsdFMYopnb6d63i&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FtfJVtrxvsdFMYopnb6d63i" width="259" height="137" frameborder="0" scrolling="no"><a href="https://medium.com/media/d4e1b70e51e2f658289f8cd84c5cee7d/href">https://medium.com/media/d4e1b70e51e2f658289f8cd84c5cee7d/href</a></iframe><p>pointing along the momentum, which has chirality +1. For the valence band state, it points exactly the opposite direction, having chirality -1. Electrons and holes in the same cone are locked into opposite pseudospin alignments. For states in K’ valley, the sign of <em>τ_z</em> flips, so the two valleys carry opposite chirality to K valley.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/904/1*8d6pZonQe03kR7g3dcj7Zw.jpeg" /></figure><p>Now let’s consider building a p-n junction with graphene. Applying gate voltages to two different regions of the graphene independently could do this. Suppose the gate generates an electrostatic potential:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FeBMTjdc6teQkSU4ruTDHky&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FeBMTjdc6teQkSU4ruTDHky" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/de0f074a92908531d8bd16ae81777bd0/href">https://medium.com/media/de0f074a92908531d8bd16ae81777bd0/href</a></iframe><p>and the potential is uniform along y-axis. This modifies the Hamiltonian of the electron in the K-valley:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F881njG2yxRozoi4jsHLy9a&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F881njG2yxRozoi4jsHLy9a" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c735143f9d8c822bfd346ae40bd2cb69/href">https://medium.com/media/c735143f9d8c822bfd346ae40bd2cb69/href</a></iframe><p>The crucial feature is that <em>V(x)</em> comes in with the identity multiplied. It can raise or lower the energy eigenvalue, but it cannot rotate the pseudospin.</p><p>Think of an electron coming from <em>x&lt;0 </em>hits the junction head-on: moving in <em>+x</em> direction, and its pseudospin pinned along <em>+x (θ_k = 0)</em>. For it to reflect, it would have to end up in the backward-moving conduction state at the same energy, which travels in <em>-x</em> and therefore carries pseudospin along <em>-x</em>. But the incident and reflected spinors are orthogonal:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FctGeKw4LPfuff2q7tSqiA&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FctGeKw4LPfuff2q7tSqiA" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b798bbf2d21cd63302ac52a4332f1770/href">https://medium.com/media/b798bbf2d21cd63302ac52a4332f1770/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FaCwiVzq2x9ZhELwU3eVEyn&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FaCwiVzq2x9ZhELwU3eVEyn" width="200" height="136" frameborder="0" scrolling="no"><a href="https://medium.com/media/22db1fb8013763c48cbfa3f9bb930bc4/href">https://medium.com/media/22db1fb8013763c48cbfa3f9bb930bc4/href</a></iframe><p>And the only term that could connect them, <em>V(x)</em>, cannot touch the pseudospin. Therefore, the backscattering amplitude is automatically zero. There is nothing to turn around the electron, so it is transmitted with coefficient <em>T = 1</em>, no matter how tall or wide the barrier is.</p><p>Then where does it go? Inside the <em>x&gt;0</em> region the electron’s energy sits below the Dirac point, so the only propagating state available is a hole state. In a case of hole, the group velocity is opposite to the momentum:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fd3GZ65rjHik3w7MUBmQiCy&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fd3GZ65rjHik3w7MUBmQiCy" width="378" height="151" frameborder="0" scrolling="no"><a href="https://medium.com/media/052251718e1de5e39c1da6ec2471e092/href">https://medium.com/media/052251718e1de5e39c1da6ec2471e092/href</a></iframe><p>Since momentum and pseudospin point in opposite directions for valence band states (as I mentioned earlier), a state carrying current along <em>+x</em> actually has momentum along <em>-x</em> and pseudospin along <em>+x</em>. The transmitted hole therefore carries the same pseudospin as the incident electron. The two match smoothy across the junction, and the wavefunction transitions from electron to hole without reversing its pseudospin, thus the chirality is perfectly conserved.</p><p><strong>Closing</strong></p><p>Let’s wrap up. We started with the carbon atoms on a honeycomb, constructed the tight-binding Hamiltonian and arrived at the relativistic Dirac fermion equation. And then we computed the effective Hamiltonian of the graphene near Dirac points, complete with pseudospin and a geometric phase of π picked up every loop around the Dirac point. Finally, we explored Klein tunneling, which is a bizarre manifestation of chirality in graphene.</p><p>The graphene Dirac cone is the prototype for an entire family of 2D materials where the low-energy physics is controlled by valley structure. The next column is about a broader question: what does it mean for a material to be two-dimensional? The answer sets the stage for every quantum phase the rest of this series will explore.</p><p><strong>References</strong></p><ul><li><em>Girvin &amp; Yang, Modern Condensed Matter Physics (2019)</em></li><li><em>Novoselov, K. S., et al. “Two-Dimensional Gas of Massless Dirac Fermions in Graphene.” Nature, vol. 438, no. 7065, 2005, pp. 197–200, </em><a href="https://doi.org/10.1038/nature04233."><em>https://doi.org/10.1038/nature04233.</em></a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8ad3195d68df" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 6 — Topology in Crystal Band Structures]]></title>
            <link>https://medium.com/@grhncgzsvb/column-6-topology-in-crystal-band-structures-8df5dcc1f653?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/8df5dcc1f653</guid>
            <category><![CDATA[condensed-matter-physics]]></category>
            <category><![CDATA[quantum-physics]]></category>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[physics]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Tue, 23 Jun 2026 00:24:19 GMT</pubDate>
            <atom:updated>2026-06-23T00:27:32.697Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Last column introduced the Berry phase — the geometric phase a quantum state picks up when its parameters complete a closed loop. In this column, we will discover how the resulting Berry curvature field produces physical consequences for electrons in crystal. The symbolic quantity within this mechanism is the Chern number — an integer firmly stamped into each energy band by topology, and that cannot be erased without tearing the band structure.</em></p><p><strong>A “Magnetic Field” in Momentum Space</strong></p><p>In column 5 we saw that a quantum state undergoing adiabatic evolution accumulates a geometric phase when its parameters trace a closed loop. We introduced the Berry connection <em>A</em> and derived the Berry phase by performing a line integral of <em>A</em> along any closed loop in the parameter space. Converting line integral into area integral gave the Berry curvature.</p><p>Relationship between the Berry connection and the Berry curvature is exactly equal to the relationship between the vector potential and the magnetic field. In the classical electrodynamics, the vector potential is gauge-dependent: you can add the gradient of any scalar function to the vector potential without changing the measurable physical quantity. The Berry connection has the same gauge dependence. If you add phase to the eigenstate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FbrBWBbgHQrYyEz8SngcgkH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FbrBWBbgHQrYyEz8SngcgkH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/24fefddc0dc4f28ffd220e7d7bd2864a/href">https://medium.com/media/24fefddc0dc4f28ffd220e7d7bd2864a/href</a></iframe><p>the Berry connection shifts by:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F5o4svCbdHkTV8or9HxoyD9&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F5o4svCbdHkTV8or9HxoyD9" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/833e5dc8e2840571cff8c53f81006fc6/href">https://medium.com/media/833e5dc8e2840571cff8c53f81006fc6/href</a></iframe><p>Taking the curl of <em>A</em> gives the Berry curvature:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FcnsHkKRC2raLZk2wgW1tKJ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FcnsHkKRC2raLZk2wgW1tKJ" width="262" height="136" frameborder="0" scrolling="no"><a href="https://medium.com/media/8c1e2f09c702cd21efb678ac24b297e5/href">https://medium.com/media/8c1e2f09c702cd21efb678ac24b297e5/href</a></iframe><p>Since the curl of gradient is always zero, the Berry curvature is evidently a gauge-invariant quantity. Thus, unlike the Berry connection, the Berry curvature is a physical, observable field, playing the role of a “magnetic field” within the parameter space of the Hamiltonian.</p><p>The Berry phase around a closed loop is the flux of the Berry curvature through the surface bounded by the loop:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FrmzHp7ufhLtXqWromywMDr&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FrmzHp7ufhLtXqWromywMDr" width="345" height="156" frameborder="0" scrolling="no"><a href="https://medium.com/media/34378a572c4e2d929e3459668cd2aaf7/href">https://medium.com/media/34378a572c4e2d929e3459668cd2aaf7/href</a></iframe><p>Hence, the phase is a “Berry flux”, analogous to the magnetic flux Φ.</p><p>Also, in three-dimensional parameter space, the Berry curvature can be written as a rank-2 tensor:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FhoXoKbtq3zmAU9C8RZcmTj&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FhoXoKbtq3zmAU9C8RZcmTj" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/d4f78ddcad54d69ed7c4d5ebb5e318b1/href">https://medium.com/media/d4f78ddcad54d69ed7c4d5ebb5e318b1/href</a></iframe><p>, a direct parallel to the electromagnetic field tensor <em>F_μν</em>.</p><p><strong>Into the Brillouin Zone</strong></p><p>Recall the k-dependent Schrödinger equation we derived:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7HtNZjx2fAgfAZxp5DPRyC&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7HtNZjx2fAgfAZxp5DPRyC" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/12ddfa4681e2a2c6e27f6153a81e7b8e/href">https://medium.com/media/12ddfa4681e2a2c6e27f6153a81e7b8e/href</a></iframe><p>Here the crystal momentum k enters the Hamiltonian as a parameter. In the Berry phase formalism:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FsyzgzrxNBAzG5mdWjRa7yc&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FsyzgzrxNBAzG5mdWjRa7yc" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/cf8d71c5349fbd3f8db4760fbbb48317/href">https://medium.com/media/cf8d71c5349fbd3f8db4760fbbb48317/href</a></iframe><p>The Berry curvature of the <em>n</em>-th Bloch band is therefore:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FuKkq2wUeekCNaXfZGP1383&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FuKkq2wUeekCNaXfZGP1383" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a15570d036f7a368b293c1d07100645a/href">https://medium.com/media/a15570d036f7a368b293c1d07100645a/href</a></iframe><p>This is a gauge-invariant “magnetic field” in the Brillouin zone.</p><p>There is a different way to write the same quantity. The Berry curvature</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FaAuuAbCSbaZsF8cBno7ZVL&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FaAuuAbCSbaZsF8cBno7ZVL" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/61beeb8f14554e64c7a187689f0cd062/href">https://medium.com/media/61beeb8f14554e64c7a187689f0cd062/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7k3ypZg7jU2CNCfrD5UVQa&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7k3ypZg7jU2CNCfrD5UVQa" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/bfc1ae02fd068c9d0687998068aac26c/href">https://medium.com/media/bfc1ae02fd068c9d0687998068aac26c/href</a></iframe><p>Since</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fu1ZzAZozWvs5zP9GVL9SfL&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fu1ZzAZozWvs5zP9GVL9SfL" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7e76e75a550215cd24cfc657f1391b35/href">https://medium.com/media/7e76e75a550215cd24cfc657f1391b35/href</a></iframe><p>the second term of the Berry curvature formula is zero and it reduces to:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2QtvZCcqeXpNkVQnQqrCdN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2QtvZCcqeXpNkVQnQqrCdN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7467e103597db5e1f1b810b8703064e6/href">https://medium.com/media/7467e103597db5e1f1b810b8703064e6/href</a></iframe><p>using the completeness of the basis:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F9555k2RjZLauSbETV51kdB&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F9555k2RjZLauSbETV51kdB" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/35ab5dc5674e7553874fce2521218845/href">https://medium.com/media/35ab5dc5674e7553874fce2521218845/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkZseRWLR3WRoabrHAABWqK&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkZseRWLR3WRoabrHAABWqK" width="277" height="157" frameborder="0" scrolling="no"><a href="https://medium.com/media/9603e231698ad60d49e643d2ad81499d/href">https://medium.com/media/9603e231698ad60d49e643d2ad81499d/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FmJMF8BKcKFQQB2Bg7FZcTx&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FmJMF8BKcKFQQB2Bg7FZcTx" width="342" height="158" frameborder="0" scrolling="no"><a href="https://medium.com/media/75475762935ce4ff78186f8337d70bd0/href">https://medium.com/media/75475762935ce4ff78186f8337d70bd0/href</a></iframe><p>The <em>n’=n</em> term cancels out in the bracket, so:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4SVwWo68vLDbUYSHic51W4&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4SVwWo68vLDbUYSHic51W4" width="417" height="158" frameborder="0" scrolling="no"><a href="https://medium.com/media/4f58c53478ded4be53b2953f52e7f067/href">https://medium.com/media/4f58c53478ded4be53b2953f52e7f067/href</a></iframe><p>Now let’s use the Schrödinger equation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FcUKbjV7Zaj8rmqXtxP7bA5&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FcUKbjV7Zaj8rmqXtxP7bA5" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/573d12e3899cc8a0ac4c4a4b114da2ba/href">https://medium.com/media/573d12e3899cc8a0ac4c4a4b114da2ba/href</a></iframe><p>Applying the bra-vector:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FnXFP6DVopktwZjGzZNznBd&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FnXFP6DVopktwZjGzZNznBd" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/cc4acf8b656482d6f12ae569fc528891/href">https://medium.com/media/cc4acf8b656482d6f12ae569fc528891/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Ftco3FAv9kV7EGDScan72TY&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Ftco3FAv9kV7EGDScan72TY" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e287940ad721fccc82c20fe0b945bddc/href">https://medium.com/media/e287940ad721fccc82c20fe0b945bddc/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FaMaknG9idbujjPdMmATib9&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FaMaknG9idbujjPdMmATib9" width="286" height="156" frameborder="0" scrolling="no"><a href="https://medium.com/media/6a37c73653cba04b46fec6be26fb017c/href">https://medium.com/media/6a37c73653cba04b46fec6be26fb017c/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fg9yCtPYwJeTrbNSeLSLtt9&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fg9yCtPYwJeTrbNSeLSLtt9" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/f5a781aedd7ed9215868eb5c5bdecf1c/href">https://medium.com/media/f5a781aedd7ed9215868eb5c5bdecf1c/href</a></iframe><p>Therefore,</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F8Zeu43F5vS8NwqcwHfdC9d&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F8Zeu43F5vS8NwqcwHfdC9d" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/bb88656a0ef91f86d223a92d109d06b4/href">https://medium.com/media/bb88656a0ef91f86d223a92d109d06b4/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqxML3GzPVQHaJ4HA5txcgU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqxML3GzPVQHaJ4HA5txcgU" width="542" height="165" frameborder="0" scrolling="no"><a href="https://medium.com/media/7f0704f0ef43302bf12687eca1ede818/href">https://medium.com/media/7f0704f0ef43302bf12687eca1ede818/href</a></iframe><p>The denominator of the Berry curvature is the square of the energy gap between the bands. If the two bands are close in energy, the Berry curvature becomes large. Whenever the energy eigenvalues become degenerate, the Berry curvature diverges. Degeneracy points are sources of Berry curvature, where it radiates outward into <em>k</em>-space like magnetic monopoles.</p><p><strong>Anomalous Velocity</strong></p><p>Using the semiclassical theory of Bloch electron gives us the average velocity:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FdQhffikmBYjeSVSEmQBvnY&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FdQhffikmBYjeSVSEmQBvnY" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c7a12b4be33f26d39b16f6c85a311dc2/href">https://medium.com/media/c7a12b4be33f26d39b16f6c85a311dc2/href</a></iframe><p>As we saw in column 5, when the Berry curvature is non-zero, there is an additional contribution:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FcET5f86xTs8M1ZTPxa9t4z&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FcET5f86xTs8M1ZTPxa9t4z" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/43ab9f27536aa6f5ea868e7d954aa0da/href">https://medium.com/media/43ab9f27536aa6f5ea868e7d954aa0da/href</a></iframe><p>where the anomalous velocity:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FbJVuYeRAE7qN9pL3USuTdb&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FbJVuYeRAE7qN9pL3USuTdb" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/dcd290e764b10b72886cb0c734f1c515/href">https://medium.com/media/dcd290e764b10b72886cb0c734f1c515/href</a></iframe><p>The anomalous velocity is perpendicular to the electric field, thus contributes a current in the transverse direction. This is the mechanism of the anomalous Hall effect — where a Hall voltage is generated without any external magnetic field.</p><p>The anomalous velocity is k-dependent, which means electrons at different crystal momenta are deflected by different amounts. For a fully occupied bands, summing the contributions of all electrons gives a Hall conductance:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Ffo3BP3eJo5E3q4wZuPHe5M&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Ffo3BP3eJo5E3q4wZuPHe5M" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c1669c221812bb05e2cf7f37f8547406/href">https://medium.com/media/c1669c221812bb05e2cf7f37f8547406/href</a></iframe><p>where <em>e²/h</em> is the conductance quantum. The Hall conductance of a filled band is proportional to the Berry curvature integrated over the 1st Brillouin zone. At first glance, this may seem like nothing more than a mere intermediate calculation, but it plays a crucial role in our discussion.</p><p><strong>The Chern Number</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*r9xJ6ONmIM9iEhL4jCQQSw.jpeg" /><figcaption>Believe it or not, this is actually a drawing of the World Cup match ball…</figcaption></figure><p>First, let’s step back and think about what kind of object we are calculating in the integral. Topology is the study of properties that do not change under continuous deformations. Two objects are topologically equivalent if one can be smoothly deformed into the other one without cutting or gluing. A coffee cup and a donut are equivalent (both have one handle), but neither of them can be continuously deformed into a football. The number of handles is a topological invariant: an integer quantity that cannot change unless you tear the surface and reconnect it. It means that you cannot smoothly deform one integer to another.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PLN-WFH2RYPT7IFw8vrJdw.jpeg" /></figure><p>If we integrate the Berry curvature over the Brillouin zone — which is a torus — we can get a quantity called the Chern number:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvnbNsBKsg7cRzLH26hxM4K&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvnbNsBKsg7cRzLH26hxM4K" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2b73f71fd8902566d4965576a39c9f3e/href">https://medium.com/media/2b73f71fd8902566d4965576a39c9f3e/href</a></iframe><p>As long as the <em>n</em>-th band does not touch any other band within the Brillouin zone — as long as it is isolated by an energy gap in the k-space — the Chern number is an integer. Using this definition, we can rewrite the Hall conductance formula:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FmpkLd77XJRMyqkkFsgkhei&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FmpkLd77XJRMyqkkFsgkhei" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/073cf4d2f8dacede17744330c7c94e57/href">https://medium.com/media/073cf4d2f8dacede17744330c7c94e57/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4wbxMLE2w7XrE9qivmS1zb&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4wbxMLE2w7XrE9qivmS1zb" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/43d23bd52ab6fe12acfb56387d658abc/href">https://medium.com/media/43d23bd52ab6fe12acfb56387d658abc/href</a></iframe><p>This is the TKNN formula, explaining the quantized Hall conductance in units of the conductance quantum.</p><p>There is also a geometric picture. As we discussed before, the points that different energy bands become degenerate acts as a magnetic monopoles. The Chern number thus counts the number of monopoles within the Brillouin zone.</p><p><strong>Why the Chern Number Must Be an Integer</strong></p><p>The claim that the Chern number is always an integer stems from differential geometry. The Gauss-Bonnet theorem states that for any smooth, closed 2D surface <em>M</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FrMULaoDtft5LEctUEo4WRR&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FrMULaoDtft5LEctUEo4WRR" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ae1710fc277ac8fec60ee906f42111fc/href">https://medium.com/media/ae1710fc277ac8fec60ee906f42111fc/href</a></iframe><p>where <em>K</em> is the Gaussian curvature and <em>g_M</em> is the genus — the number of handles. For a sphere:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FucQbkAWeoMZNJSEEAHzuUN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FucQbkAWeoMZNJSEEAHzuUN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7dbaf96445a38820e01067e725ad88aa/href">https://medium.com/media/7dbaf96445a38820e01067e725ad88aa/href</a></iframe><p>we can obtain the equality:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqZtYWUn821kv7ojfkMWLtC&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqZtYWUn821kv7ojfkMWLtC" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ce8cff35b75164d924f82dc0221fc7ee/href">https://medium.com/media/ce8cff35b75164d924f82dc0221fc7ee/href</a></iframe><p>In the case of donut, the integral equals to 0. The total curvature is a topological invariant: no matter how you stretch, squeeze, or deform it into any shape, the evaluated integral remains unchanged. The genus cannot change under continuous deformation.</p><p>The Berry curvature plays the role of <em>K</em>, the Brillouin zone torus plays the role of <em>M</em>, and the Chern number plays the role of <em>2–2g_M</em>. Smooth changes in the band structure cannot change the total integral over the torus. The Chern number is thus topologically protected, just like a genus of a donut. The only thing that can change the Chern number is a band-closing event. If the energy gap between two separate bands closes, the singularity of the Berry curvature emerges, which leads to a change in the Chern number.</p><p>In addition, the Chern number integral can be rewritten as a line integral of the Berry connection around the zone boundary. Since the Brillouin zone in the torus — surface with opposite edges glued together — the boundary is a closed loop, and as <em>k</em> traverses the loop, the Bloch wavefunction returns to itself, picking up the phase γ. For the wavefunction to be single-valued, γ must be an integer multiple of 2π. Therefore, the Chern number is exactly the winding number of phase factor:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7QPJWS6xPYvTJjKernniQg&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7QPJWS6xPYvTJjKernniQg" width="298" height="144" frameborder="0" scrolling="no"><a href="https://medium.com/media/67c2c2dd810afc14a8258af93b9b2ef7/href">https://medium.com/media/67c2c2dd810afc14a8258af93b9b2ef7/href</a></iframe><p><strong>A Glimpse of the Chern Insulator</strong></p><p>The Chern number does actually provide a bizarre phenomenon at any boundary of the material. For instance, think of an insulator whose occupied bands carry a total non-zero Chern number <em>C</em>. Threading a magnetic flux quantum through the system adiabatically forces exactly <em>C</em> electrons from one edge of the sample to the opposite side. In order to make the electrons mobile, they should cross the Fermi energy. However the Fermi energy sits in the bulk gap, so no bulk states are available. Hence, the electrons that move are necessarily at the edges. This is the origin of the gapless edge states: conducting channels at the boundary of a topologically non-trivial insulator, which is called a Chern insulator.</p><p>The Berry curvature is the source of all of this. It is the “magnetic field” in the Brillouin zone, and the Chern number is the total flux through the closed Brillouin zone torus. So the analogy of a Berry connection as vector potential and Berry curvature as magnetic field, is not just a stretch. It is a core viewpoint of the topology of the band theory, and it will keep reappearing throughout the upcoming columns.</p><p><strong>References</strong></p><ul><li><em>Girvin &amp; Yang, Modern Condensed Matter Physics (2019)</em></li><li><em>Sprinkart, Scheer &amp; Di Bernardo, “Tutorial: From Topology to Hall Effects — Implications of Berry Phase Physics” (2024)</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8df5dcc1f653" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 5 — The Berry Phase]]></title>
            <link>https://medium.com/@grhncgzsvb/column-5-the-berry-phase-34589ed2ed58?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/34589ed2ed58</guid>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[quantum-physics]]></category>
            <category><![CDATA[condensed-matter-physics]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Sat, 20 Jun 2026 10:30:58 GMT</pubDate>
            <atom:updated>2026-06-23T00:27:58.245Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Every quantum state carries a phase. Most of them vanish when you take a measurement, but some cannot be extinguished if you close the right loop. This column is about that specific kind of phase; a phase that depends only on the shape of a path.</em></p><p><strong>The Phase That We Didn’t Know Was There</strong></p><p>Let’s start with a general setup. Suppose you have a Hamiltonian that depends on some external parameter <em>R</em>. It could be anything such as nuclear separation in the molecule or the direction of a magnetic field. We can write down the Schrödinger’s equation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FfXCctPBUCN1rpnzjZAjycT&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FfXCctPBUCN1rpnzjZAjycT" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6144b917e5fc1abc3aeef3b94b109623/href">https://medium.com/media/6144b917e5fc1abc3aeef3b94b109623/href</a></iframe><p>For a specific time <em>t</em>, we can define the energy eigenvalue as:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2NQaVwZLbep97m1sHxS2MN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2NQaVwZLbep97m1sHxS2MN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/aadea7bee355c0956951b95a1b486ded/href">https://medium.com/media/aadea7bee355c0956951b95a1b486ded/href</a></iframe><p>As the parameters change slowly, a system starting in the <em>n</em>-th eigenstate stays in the corresponding <em>n</em>-th eigenstate, according to the adiabatic approximation. So the time-evolved state has the form:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fc6g2RQ8fVZhcFQQ4NiUYdG&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fc6g2RQ8fVZhcFQQ4NiUYdG" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8d88043d4a1305f6688ae7418f691011/href">https://medium.com/media/8d88043d4a1305f6688ae7418f691011/href</a></iframe><p>Applying Hamiltonian operator:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fna2AJhzT6PfJskRKU7wnhX&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fna2AJhzT6PfJskRKU7wnhX" width="653" height="158" frameborder="0" scrolling="no"><a href="https://medium.com/media/4f7a122f209f511e6a73ce04b9ef70ee/href">https://medium.com/media/4f7a122f209f511e6a73ce04b9ef70ee/href</a></iframe><p>Projecting this to the bra-vector:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FghjQ4kxbdNM7y4i4tj62LP&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FghjQ4kxbdNM7y4i4tj62LP" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/32ada9e69d6501d2f0badf06bd6645ce/href">https://medium.com/media/32ada9e69d6501d2f0badf06bd6645ce/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fc6RHoLofdtU5EfYThD1XgU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fc6RHoLofdtU5EfYThD1XgU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/70dc6c3c57482af577d924c2a41a483b/href">https://medium.com/media/70dc6c3c57482af577d924c2a41a483b/href</a></iframe><p>If we solve this differential equation of the coefficient C:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fp2jzUmooehinaspZAuwckH&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fp2jzUmooehinaspZAuwckH" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6157399cf493ba01ca68e04f2005b061/href">https://medium.com/media/6157399cf493ba01ca68e04f2005b061/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fae7MnY9bezyBPDjE1iBTyN&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fae7MnY9bezyBPDjE1iBTyN" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/f47f7d3f87c934eee268d75c61153ef2/href">https://medium.com/media/f47f7d3f87c934eee268d75c61153ef2/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvsnaAioogmRZUw8hxAgJuz&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvsnaAioogmRZUw8hxAgJuz" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/da037967ff0c52d57e6034f02f38883e/href">https://medium.com/media/da037967ff0c52d57e6034f02f38883e/href</a></iframe><p>The first exponential factor is the dynamical phase. This is trivial result that we often have seen when solving the simple Schrödinger’s equation. What we should focus on is the second one. If we define:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FrcgfgWC2PnbxocoN1ntAe5&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FrcgfgWC2PnbxocoN1ntAe5" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/38d1211afc6a3903887b08b6bc829453/href">https://medium.com/media/38d1211afc6a3903887b08b6bc829453/href</a></iframe><p>then</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fn57LWdcA1gC6WqPvpqwHwb&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fn57LWdcA1gC6WqPvpqwHwb" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8ceb7b78c17312019ad658ef3f67d358/href">https://medium.com/media/8ceb7b78c17312019ad658ef3f67d358/href</a></iframe><p>If we convert the time-derivative to a parameter-derivative using the chain rule,</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvxUcPavrRuCnhdoHfRt1pS&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvxUcPavrRuCnhdoHfRt1pS" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/410dc90d1d1e71ae99571c7bd7500690/href">https://medium.com/media/410dc90d1d1e71ae99571c7bd7500690/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fha72NLoitYZRATnn6GSxV1&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fha72NLoitYZRATnn6GSxV1" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/fd381eaa00ebdac43792fcc3c5f0ea8b/href">https://medium.com/media/fd381eaa00ebdac43792fcc3c5f0ea8b/href</a></iframe><p>Here, the <em>A</em> is called the Berry connection, and γ is the geometric phase — the Berry phase. The important thing is that the phase depends only on the path <em>C</em> that <em>R</em> evolves through parameter space; no other dependencies such as time, energy or speed of the parameter changes. This is the key difference from the dynamical phase.</p><p><strong>The Geometric Phase Cannot Be Gauged Away</strong></p><p>When physicists first encountered the geometric phase, they did not much care about this. Because for an open path in parameter space, we can always choose the phase of the energy eigenstate through gauge transformation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3cQH8jKb5ESJZzk2zccJ5W&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3cQH8jKb5ESJZzk2zccJ5W" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e3d5c6131aa734f9c21a5995ba7f5e2f/href">https://medium.com/media/e3d5c6131aa734f9c21a5995ba7f5e2f/href</a></iframe><p>Here the Berry connection shifts by:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fi95Bxa8CUyPbBcnVBFGv8N&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fi95Bxa8CUyPbBcnVBFGv8N" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/925d33cd27d197f1cf024ea6428dfd2a/href">https://medium.com/media/925d33cd27d197f1cf024ea6428dfd2a/href</a></iframe><p>and the accumulated phase changes as:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fsq9AVeKz1x6C8mr1S9xNNk&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fsq9AVeKz1x6C8mr1S9xNNk" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/bc4bb4e10d3cd1a6be785001b58e08dd/href">https://medium.com/media/bc4bb4e10d3cd1a6be785001b58e08dd/href</a></iframe><p>So for the open path such that</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fe5RYe4tbKpSA7KUp1RhEmW&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fe5RYe4tbKpSA7KUp1RhEmW" width="245" height="137" frameborder="0" scrolling="no"><a href="https://medium.com/media/11ba129e7663dc03e86411dc4dd77cb0/href">https://medium.com/media/11ba129e7663dc03e86411dc4dd77cb0/href</a></iframe><p>we can shift γ by any amount we want by choosing appropriate ζ. So the phase is gauge-dependent, thus is not a physical observable.</p><p>However, if we consider a closed loop where the parameter always satisfies</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FmHiQqqV5BmgAfNGCB6Z2M3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FmHiQqqV5BmgAfNGCB6Z2M3" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/ba70dae36d49d6122f4efde79a82bfa6/href">https://medium.com/media/ba70dae36d49d6122f4efde79a82bfa6/href</a></iframe><p>the phase shift from the gauge transformation vanishes, and</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fg4MK3YSCVZSQEFFyriBLr1&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fg4MK3YSCVZSQEFFyriBLr1" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b6bbf6251d9e8b83033a27bbc6761ac1/href">https://medium.com/media/b6bbf6251d9e8b83033a27bbc6761ac1/href</a></iframe><p>Here, no matter what ζ we choose, the closed-loop integral of Berry connection is unchanged. What left here is a gauge-invariant physical quantity. This is Berry’s discovery: the geometric phase accumulated through a closed loop in parameter space is a genuine physical observable, independent of any phase convention.</p><p>In addition, if we use the Stokes’ theorem on the Berry phase formula:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FjH8u8kMwvEQkkPLmkbyxc6&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FjH8u8kMwvEQkkPLmkbyxc6" width="525" height="156" frameborder="0" scrolling="no"><a href="https://medium.com/media/0b977294e857cb577a60f9f0d8f6c2f8/href">https://medium.com/media/0b977294e857cb577a60f9f0d8f6c2f8/href</a></iframe><p>where Ω is the Berry curvature. The Berry phase can thus be interpreted as the flux of the Berry curvature through the enclosed surface <em>S</em>.</p><p><strong>Spin-1/2 particle in a Rotating Field</strong></p><p>To make this physically straightforward, let’s consider the simplest possible system that exhibits a Berry phase: a spin-1/2 particle in a magnetic field which slowly rotates. The Hamiltonian of this system is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7G22Ao6b1DpxGDWQMBu4YU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7G22Ao6b1DpxGDWQMBu4YU" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/859f2346d638a5a1097fca8f6173e16b/href">https://medium.com/media/859f2346d638a5a1097fca8f6173e16b/href</a></iframe><p>where:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvKZ5DbzhCBmVdLhJib8aYU&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvKZ5DbzhCBmVdLhJib8aYU" width="298" height="137" frameborder="0" scrolling="no"><a href="https://medium.com/media/a7545d493db1fc7d48b3d0c444123e37/href">https://medium.com/media/a7545d493db1fc7d48b3d0c444123e37/href</a></iframe><p>If we solve for the eigenvalues and eigenvectors:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FnKX8hGbtxKD9WfSEJK9yF1&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FnKX8hGbtxKD9WfSEJK9yF1" width="240" height="160" frameborder="0" scrolling="no"><a href="https://medium.com/media/dcc9b0c342929242d63581326a80cb86/href">https://medium.com/media/dcc9b0c342929242d63581326a80cb86/href</a></iframe><p>We can see that the eigenstates depend on the orientation of the external magnetic field. Thus, as the field direction traces a closed path on the unit sphere, these eigenstates rotate continuously. We can compute Berry connection explicitly for the spin eigenstates. For the lower eigenstate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FhumsQYJmeKCuJDFZbTMdDD&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FhumsQYJmeKCuJDFZbTMdDD" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b40da1695a510a6262fa3dc53c1e3e57/href">https://medium.com/media/b40da1695a510a6262fa3dc53c1e3e57/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FeTqnzNutpc8S6rhLS9A4ry&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FeTqnzNutpc8S6rhLS9A4ry" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b988c5847fe7e84226e96fb211079f4f/href">https://medium.com/media/b988c5847fe7e84226e96fb211079f4f/href</a></iframe><p>Integrating around a closed loop gives the Berry phase:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F69BCC2iUz7pb3s6VQ1KWeY&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F69BCC2iUz7pb3s6VQ1KWeY" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0129cb9f8a56ef2b3723f448951f1198/href">https://medium.com/media/0129cb9f8a56ef2b3723f448951f1198/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FkA4dUtrS2Aw2XBzr7Z9gK9&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FkA4dUtrS2Aw2XBzr7Z9gK9" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/45b7d869fe3c1140f1210eab6a76734f/href">https://medium.com/media/45b7d869fe3c1140f1210eab6a76734f/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fnb793osPXLJMDMYXFPFRME&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fnb793osPXLJMDMYXFPFRME" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/cab28358b27ae2d5cc109a2ca561114e/href">https://medium.com/media/cab28358b27ae2d5cc109a2ca561114e/href</a></iframe><p>where <em>F</em> is a Berry curvature tensor component and Ω<em>s</em> is the solid angle subtended by the path <em>C</em> from the origin. The same result holds for spin-up state, with the opposite sign.</p><p>Thus, the Berry phase is half the solid angle enclosed by the trajectory on the unit sphere. There is no energy, no time, no speed; only the area of the loop on <em>S</em>. A path that encircles the hemisphere (Ω<em>s</em> = 2<em>π</em>) accumulates a phase of <em>π</em>, regardless of whether it takes a second or a year.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*yowL8jKuvBST6FkWD-sCNQ.jpeg" /><figcaption>3-Dimensional parameter sphere</figcaption></figure><p>There is one more thing worth mentioning. <em>F</em> is the Berry curvature component on the unit sphere of field directions, ignoring the magnitude of the field <em>h</em>. If we consider Berry phase in the full parameter space:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fh493Zd7r14sJ5e5SqDkjrq&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fh493Zd7r14sJ5e5SqDkjrq" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/3f628b5c1581637048b8fe1f2d2c3cad/href">https://medium.com/media/3f628b5c1581637048b8fe1f2d2c3cad/href</a></iframe><p>where <em>b</em> is the curvature seen as a field in 3-dimensional parameter space and <em>dS</em> is the area element of the sphere of radius <em>h</em>. Comparing with the value we already obtained:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2QPKDkwdfqtaSmsCcgJfvw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2QPKDkwdfqtaSmsCcgJfvw" width="640" height="158" frameborder="0" scrolling="no"><a href="https://medium.com/media/3a9a7b1dd8e39eb2f8cbb5649581ae06/href">https://medium.com/media/3a9a7b1dd8e39eb2f8cbb5649581ae06/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FuzrwSpZezPmRPsdZMAr9gi&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FuzrwSpZezPmRPsdZMAr9gi" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/cdd8208ad8275894ce6ed8919bc4d5e8/href">https://medium.com/media/cdd8208ad8275894ce6ed8919bc4d5e8/href</a></iframe><p>This is exactly the field of a magnetic monopole sitting at the origin, where the two eigenstates become degenerate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fhf5Qhhv5j4urBCmSBK7Ukp&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fhf5Qhhv5j4urBCmSBK7Ukp" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8347b6b8c716ff55a37152a538cca1a1/href">https://medium.com/media/8347b6b8c716ff55a37152a538cca1a1/href</a></iframe><p>Therefore, we can say that the degeneracy point radiates the Berry curvature.</p><p><strong>Berry Phase in Crystal Structure</strong></p><p>Discussion so far has been about a spin in a rotating magnetic field. But the Berry phase formalism assumed no specific system or parameter. It needs only two things: a Hamiltonian parameterized by some external parameter, and a closed loop in the parameter space.</p><p>According to the Bloch’s theorem, the wavefunction of an electron in a periodic potential is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FoHzrf1ijVwZKx5iDGk67wt&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FoHzrf1ijVwZKx5iDGk67wt" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/07c9d74ed8e19491d3bc1b7d64d60d30/href">https://medium.com/media/07c9d74ed8e19491d3bc1b7d64d60d30/href</a></iframe><p>and the Hamiltonian gives the energy eigenvalue:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FgbKZ9dZNXnfjQWUY5SUy7q&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FgbKZ9dZNXnfjQWUY5SUy7q" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/917cde011dd63bf297b4f9f6953cd164/href">https://medium.com/media/917cde011dd63bf297b4f9f6953cd164/href</a></iframe><p>if we massage the Hamiltonian equation a bit, we can get a parameterized Hamiltonian:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fm8cH1nYeFMLybGfmvJvJ7c&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fm8cH1nYeFMLybGfmvJvJ7c" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/c3672b4630598adc46d0911d7782b7dc/href">https://medium.com/media/c3672b4630598adc46d0911d7782b7dc/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F4qYj6goC6tQq8BmNyVuMja&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F4qYj6goC6tQq8BmNyVuMja" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/503d55268a0b253b0490607b6d0ffb5f/href">https://medium.com/media/503d55268a0b253b0490607b6d0ffb5f/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FbjwEpzv5ZZ6reuBfEUZDNQ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FbjwEpzv5ZZ6reuBfEUZDNQ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/6b32b5017c418455d66b69124ec2d869/href">https://medium.com/media/6b32b5017c418455d66b69124ec2d869/href</a></iframe><p>Here the crystal momentum <em>k</em> plays the role of external parameter.</p><p>Now the topology of the Brillouin zone becomes crucial. Periodic boundary condition in reciprocal space makes the opposite endpoints of the first Brillouin zone equal:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FtQSZFrXX1kHJMkqWAVany4&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FtQSZFrXX1kHJMkqWAVany4" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/712bf3de0a587c028a6d91d9135d2a46/href">https://medium.com/media/712bf3de0a587c028a6d91d9135d2a46/href</a></iframe><p>producing the torus-shaped Brillouin zone in 2 dimension. As <em>k</em> traverses the torus, it is always travelling through a closed surface.</p><p>In consequence, for any closed path in the Brillouin zone, the Bloch wavefunction accumulates a Berry phase:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6f6a5jUhE6DpgsWao4k4HD&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6f6a5jUhE6DpgsWao4k4HD" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0369a6b8a2f9cf5bec7579c71360b429/href">https://medium.com/media/0369a6b8a2f9cf5bec7579c71360b429/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FtRmXVohchdiM3gEdwMP3Rw&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FtRmXVohchdiM3gEdwMP3Rw" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/da21a8da4ad5e00266afbb25cfd4cdf3/href">https://medium.com/media/da21a8da4ad5e00266afbb25cfd4cdf3/href</a></iframe><p>The Berry curvature — the local field whose flux gives the Berry phase — is defined at every point in the Brillouin zone:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2g2Hd8UB5FNzhzMbX1fjpj&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2g2Hd8UB5FNzhzMbX1fjpj" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/47d060f305a51de593d0c000ea7a0a0b/href">https://medium.com/media/47d060f305a51de593d0c000ea7a0a0b/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fm5y79dW7PKFdCnt51G9ebJ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fm5y79dW7PKFdCnt51G9ebJ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/5a81bb458c94c4f84e97f09817958a69/href">https://medium.com/media/5a81bb458c94c4f84e97f09817958a69/href</a></iframe><p>given that:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FipzgSsdAZg1Lia46zjTeMd&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FipzgSsdAZg1Lia46zjTeMd" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/48ab2b43692bc113c4860a6750cc8b23/href">https://medium.com/media/48ab2b43692bc113c4860a6750cc8b23/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FbkvCVqGRH997RsxeY8xSK9&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FbkvCVqGRH997RsxeY8xSK9" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/059eafdd7f3268aeee5023088035675c/href">https://medium.com/media/059eafdd7f3268aeee5023088035675c/href</a></iframe><p>we can conclude:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fw63XiZ81V7pXdD8HQ2YkKT&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fw63XiZ81V7pXdD8HQ2YkKT" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4eba0bcb9506009ed1872277f0efb0a9/href">https://medium.com/media/4eba0bcb9506009ed1872277f0efb0a9/href</a></iframe><p>and thus, the Berry curvature:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F8jA9MK7j7BkEB37sqMgQDq&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F8jA9MK7j7BkEB37sqMgQDq" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/f7c668b7fb6703c4853be83233ac7b7c/href">https://medium.com/media/f7c668b7fb6703c4853be83233ac7b7c/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F8Kic6eaovb7n23mSoEXuWv&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F8Kic6eaovb7n23mSoEXuWv" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e9b279ef3fc63ee9ebd6b4313dd58212/href">https://medium.com/media/e9b279ef3fc63ee9ebd6b4313dd58212/href</a></iframe><p>In analogy with electrodynamics, the Berry connection <em>A</em> is a vector potential in k-space, and the Berry curvature is a magnetic field in the Brillouin zone. It is not a handy interpretation; it has a physical consequences. For instance, when an electric field is applied to a crystal, the Berry curvature generates an anomalous velocity:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FgsAiPB8XzzASSaEdc6FCFM&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FgsAiPB8XzzASSaEdc6FCFM" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/df718b70fd7538a9064dfd32ea0bdaee/href">https://medium.com/media/df718b70fd7538a9064dfd32ea0bdaee/href</a></iframe><p>The first term is group velocity, and the second term deflects electrons sideways, perpendicular to both the E-field and the Berry curvature, without presence of an external magnetic field. This is called the anomalous Hall velocity.</p><p>Let’s wrap up. We started from the adiabatic approximation to derive the Berry phase, and defined key physical quantities along the way, such as the Berry connection and Berry curvature. We then got a more intuitive understanding of the Berry phase formalism through the example of the spin-1/2 particle. Finally, we applied this concept to a crystal structure. In the next column, we will further dive into various physical consequences that stem from the Berry curvature, including the definition of the Chern number, quantized Hall conductance, and more.</p><p><strong>References</strong></p><ul><li><em>Girvin &amp; Yang, Modern Condensed Matter Physics (2019)</em></li><li><em>Sakurai &amp; Napolitano, Modern Quantum Mechanics (2nd ed.)</em></li><li><em>Sprinkart, Scheer &amp; Di Bernardo, “Tutorial: From Topology to Hall Effects — Implications of Berry Phase Physics” (2024)</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=34589ed2ed58" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 4 — Second Quantization and the Hubbard Model]]></title>
            <link>https://medium.com/@grhncgzsvb/column-4-second-quantization-and-the-hubbard-model-6084857f2778?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/6084857f2778</guid>
            <category><![CDATA[condensed-matter-physics]]></category>
            <category><![CDATA[quantum-physics]]></category>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[quantum-mechanics]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Tue, 02 Jun 2026 04:19:01 GMT</pubDate>
            <atom:updated>2026-06-02T07:03:43.460Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Discussions of the band theory treats electrons as non-interacting particles moving through a periodic potential. This column introduces a new language, second quantization, that makes electron-electron interactions tractable, and deploys it to build the Hubbard model: the simplest model that captures the tension between kinetic energy and repulsion.</em></p><p><strong>Why Particle Labels Are the Problem</strong></p><p>In undergraduate quantum physics course, we learn how to write down a many body wavefunction. For <em>N</em> electrons, we form a Slater determinant: an anti-symmetrized product of <em>N</em> single-particle wavefunctions, each labeled by a quantum number. It takes the form:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FoZZj7o87avZVhxzcWh2n4n&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FoZZj7o87avZVhxzcWh2n4n" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/0b576c29be49506e49cfd5bf9de8e9b2/href">https://medium.com/media/0b576c29be49506e49cfd5bf9de8e9b2/href</a></iframe><p>which depends on N position variables and encodes the Pauli exclusion principle through its anti-symmetry under particle exchange.</p><p>This works beautifully for <em>N</em> = 2 or 3. However, for realistic solid with <em>N</em> an order of Avogadro number (<em>N</em> ~ 10²³), it is not just computationally intractable — it is conceptually cumbersome. Every time you want to describe a physical process, you have to track which particle in which state, even though the particles are physically indistinguishable. The labels you attach to electrons are arbitrary and meaningless, but yet the formalism forces you to carry them every step.</p><p>There is a deeper problem too. In first quantization, the Hilbert space is fixed by specifying <em>N</em> initially. You cannot even write down a system whose number of particles is uncertain or variable.</p><p>Second quantization solves both problems by abandoning particle labels entirely. The key conceptual question is not about “which electron is in which state?”, but “how many electrons occupy each state?” The particles disappear, the slots remain.</p><p><strong>Fock Space and the Occupation Number Representation</strong></p><p>To make this precise, we need a special kind of Hilbert space: a <strong>Fock space</strong>. The standard Hilbert space for <em>N</em> particles does not have room for states with different particle numbers. The Fock space includes states with every possible number of particles, from 0 to infinity.</p><p>A basis for Fock space is built from occupation numbers. For fermions like electrons, we assign a number <em>n_r ∈ {0, 1}</em> to each single-particle state <em>r</em>: the state is either empty or occupied. A complete Fock space basis state is then the set of <em>{n_r}</em>, which tells you the occupation of every single-particle state simultaneously. The total number of electrons in system is simply</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2wCdsKfRCCsT5eyAgi2ocD&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2wCdsKfRCCsT5eyAgi2ocD" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b2b08a68c0b8c510f992d4b1cb4a14d1/href">https://medium.com/media/b2b08a68c0b8c510f992d4b1cb4a14d1/href</a></iframe><p><strong>Creation and Annihilation Operators</strong></p><p>The machinery that operates on Fock space consists of two types of operators for each single-particle state <em>r</em>. The <strong>creation operator</strong> increases the occupation of state <em>r</em> from 0 to 1, or gives zero if the state is already occupied. The <strong>annihilation operator</strong> does the reverse: it removes an electron from state <em>r</em>, or gives zero if the state is already empty. Starting from the vacuum state |0&gt;, every Fock space basis state can be derived by applying creation operators:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FwAfqGPtAHXa9gmW9TL7S2v&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FwAfqGPtAHXa9gmW9TL7S2v" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/091e398d35362d021f964e6040e5df21/href">https://medium.com/media/091e398d35362d021f964e6040e5df21/href</a></iframe><p>If you think about the harmonic oscillator ladder operators, this structure should seem familiar. What is different for fermions is the algebra these operators satisfy. Two fermionic creation operators anti-commute.</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fm2QQY1gp4tTNZ7KSipNzou&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fm2QQY1gp4tTNZ7KSipNzou" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/9c3b389bfb68af29d47869c7d7d544c1/href">https://medium.com/media/9c3b389bfb68af29d47869c7d7d544c1/href</a></iframe><p>The special case <em>r = s</em> gives:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F3df5JFNaGPGRF2SGGoHc2U&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F3df5JFNaGPGRF2SGGoHc2U" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/dce3aee494d06a50c69bef8a22b550ba/href">https://medium.com/media/dce3aee494d06a50c69bef8a22b550ba/href</a></iframe><p>which indicates you cannot create two electrons in the same state. Pauli exclusion is well embodied in this structure. The fundamental relation between a creation and annihilation operator is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fq9XvGu9rD5DL6AUiCkbego&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fq9XvGu9rD5DL6AUiCkbego" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/034e1b2d753970cf6ff84e5f830e2b80/href">https://medium.com/media/034e1b2d753970cf6ff84e5f830e2b80/href</a></iframe><p>Here, the anti-commutator implements the fermionic exchange mechanism automatically.</p><p>The occupation number operator</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvdSX1N9TCsNrRkjfGvEpjv&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvdSX1N9TCsNrRkjfGvEpjv" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/e1a0e9e80d168bdcf89e0d8fd37a61d2/href">https://medium.com/media/e1a0e9e80d168bdcf89e0d8fd37a61d2/href</a></iframe><p>has eigenvalues of 0 or 1, consistent with fermionic occupation numbers. The total particle number operator is then:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FmoFqzXvsGVkodSVqKDApAX&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FmoFqzXvsGVkodSVqKDApAX" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7478448ba5f4a2013f87e860f3b503de/href">https://medium.com/media/7478448ba5f4a2013f87e860f3b503de/href</a></iframe><p><strong>From Tight-Binding model to the Hubbard Model</strong></p><p>The second quantization provides a general recipe for converting any quantum mechanical operator into Fock-space form. For any arbitrary one body operator, we can construct a Fock-space version:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F26GofZHJYTcf27wiaGXof3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F26GofZHJYTcf27wiaGXof3" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/2df1ae05004b85da7b9e2d46c8e5b5c1/href">https://medium.com/media/2df1ae05004b85da7b9e2d46c8e5b5c1/href</a></iframe><p>Let’s think of an example: a tight-binding model. In tight-binding model, electrons can hop between localized orbitals on neighboring lattice sites with certain hopping amplitude <em>t</em>, which is set by the overlap of orbitals on adjacent sites. When we apply the one-body recipe above only in the nearest-neighbor elements — which is precisely the tight-binding approximation — the result is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FsyKqPzRmCUAM3wBFpKeiUT&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FsyKqPzRmCUAM3wBFpKeiUT" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/fa673333c41b0e4eb87b2fc03496bcca/href">https://medium.com/media/fa673333c41b0e4eb87b2fc03496bcca/href</a></iframe><p>The first one is the tight-binding term, and the second term counts doubly-occupied sites and charges an energy penalty <em>U</em>. The product:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6mATVup1nWSqRsSxxGzGA7&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6mATVup1nWSqRsSxxGzGA7" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/7a40deed60b8a3b4317007277b204439/href">https://medium.com/media/7a40deed60b8a3b4317007277b204439/href</a></iframe><p>Here, two parameters govern the physics: the hopping amplitude <em>t</em>, which sets the bandwidth and drives electrons to delocalize, and the on-site repulsion <em>U</em>, which penalized double occupancy and drives electrons to localize.</p><p><strong>The Competition Between Delocalization against Localization</strong></p><p>To see what the competition results in, it helps to think about two clean limits.</p><p>First, set <em>U</em> = 0. The model reduces to pure tight-binding model. The eigenstates are then the Bloch waves. When the bands are half-filled — one electron per site on average, filling the lower half of the band — the ground state is a metallic Fermi sea. The system conducts freely. Band theory works perfectly.</p><p>Now set <em>t</em> = 0 and <em>U</em> very large. Hopping is entirely forbidden. Each electron is pinned to its site. At half-filling there is exactly one electron per site, and every configuration of spin orientations (up or down at each site) is a valid ground state. So there are 2^<em>N</em> ground states, all degenerate. The system clearly cannot conduct, because there is no kinetic energy for electrons to pay the energy cost <em>U</em> to move into any other occupied state. The system is an insulator. But as I mentioned earlier in Column 2, the band theory sees nothing. A half-filled band in the independent-electron picture is a metal. The insulation has nothing to do with band gaps — it is caused entirely by electron-electron repulsion.</p><p>What happens if both <em>t</em> and <em>U</em> term are present? Despite its simplicity of the model, we cannot obtain exact analytical solution. The Hubbard model is the minimal setting in which these competing tendencies can both be active, but already it is too hard to solve exactly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1011/1*nX77ZD4tkmxqB9muWzXErA.jpeg" /></figure><p>The physics of the competition has a magnetic effect as well. At large <em>U</em>, where electrons are largely localized, they can temporarily move to a neighboring site (which called a virtual hop) — allowed by uncertainty principle — and creates a short-living doubly occupied site. This costs energy <em>U</em>, but for a pair of electrons on the same site, it lowers the kinetic energy. The resulting effect is an antiferromagnetic coupling between neighboring spins, with exchange constant:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fvh2A4n2jNMKfVgjmYweiif&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fvh2A4n2jNMKfVgjmYweiif" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/92ec06ee14a62e90bcd091e8c6eaa0ea/href">https://medium.com/media/92ec06ee14a62e90bcd091e8c6eaa0ea/href</a></iframe><p>So the antiferromagnetism has a kinetic-energy origin. It is the way of gaining some hopping energy without paying the full double-occupancy cost. This is why antiferromagnetism is the characteristic magnetic order of the Mott insulating phase.</p><p><strong>The Mott Insulator</strong></p><p>The strongly interacting half-filled state has a name called Mott insulator. To appreciate what is new here, recall how the two familiar insulating mechanisms work. A band insulator has a completely filled valence band and an empty conduction band. The chemical potential sits in a gap, and adding an extra electron requires promoting it across the band gap. The compressibility <em>dn/dμ</em> is zero because there are no available states at the Fermi level. A disorder-induced insulator — which is known as Anderson insulator — has states available at the Fermi level (so <em>dn/dμ &gt; 0</em>), but the eigenstates are all spatially localized by disorder, resulting in a diffusion constant to vanish: <em>D</em> = 0.</p><p>The compressibility <em>dn/dμ</em> is also zero, like a band insulator, but the gap is not caused by band structure but is entirely an effect from interactions. Many transition-metal oxides fall into this category — materials with partially filled d-bands that conventional band theory classifies as metals, are experimentally proved to be insulators with large gaps and antiferromagnetic order.</p><p><strong>What the Language Brought to Us</strong></p><p>Let’s wrap up. We started from the observation that particle labels are nonsense, went through the construction of Fock space and the operator algebra, to a compact two-term Hubbard Hamiltonian that implies rich physics. In the next column, I will introduce the Berry phase — a geometric object that hides the quantum state of any system with slowly varying parameters within.</p><p><strong>References</strong></p><ul><li><em>Girvin &amp; Yang, Modern Condensed Matter Physics (2019)</em></li><li><em>Ashcroft &amp; Mermin, Solid State Physics (1976)</em></li><li><em>Sakurai &amp; Napolitano, Modern Quantum Mechanics (2nd ed.)</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6084857f2778" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 3 — Symmetry Breaking and Order: The Landau Framework]]></title>
            <link>https://medium.com/@grhncgzsvb/column-3-symmetry-breaking-and-order-the-landau-framework-464d7f03cea2?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/464d7f03cea2</guid>
            <category><![CDATA[quantum-physics]]></category>
            <category><![CDATA[condensed-matter-physics]]></category>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[physics]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Fri, 22 May 2026 04:35:10 GMT</pubDate>
            <atom:updated>2026-05-22T04:35:10.926Z</atom:updated>
            <content:encoded><![CDATA[<p><em>This column introduces one of the most powerful organizing ideas in all of physics: that the phases of matter can be classified by which symmetries they break. Developed by Lev Landau in 1930s, this framework unifies phenomena as different as ferromagnetism and crystallization under a single conceptual umbrella — and sets the stage for understanding why superconductivity is, in a subtle but profound sense, entirely something else.</em></p><p>There is a common experience in physics where a pattern, once you see it, becomes impossible to unsee. You realize that two phenomena you thought had nothing to do with each other are actually the same phenomenon in different clothing. That experience is what Landau’s theory of phase transitions offers. The magnet on your fridge door and the ice in your glass are, from the right vantage point, doing the same thing.</p><p><strong>What Does It Mean to Break a Symmetry?</strong></p><p>Let’s start with a fundamental question: what is a symmetry of a physical system? A symmetry is a transformation — rotation, translation, reflection, reversal of time — that leaves the Hamiltonian invariant. For example, rotating a ferromagnet’s Hamiltonian by any angle in 3D doesn’t change the energy. The Hamiltonian does not care about the orientation of the spins or the position of the origin. It is agnostic.</p><p>But the <em>state</em> of the system — the actual configuration the material settles into — is often not agnostic. Below the Curie temperature, a ferromagnet spontaneously develops a magnetization pointing in specific direction. Above the temperature, the spins are disordered and have no preferred direction. If we rotate the low-temperature state, the spins point different direction. Hamiltonian is still unchanged but state is not. Thus, we can say that the low-temperature phase has low symmetry than the Hamiltonian. This is what physicists mean by a spontaneous symmetry breaking.</p><p>The term “spontaneous” plays an actual role. The Hamiltonian does not choose the direction of magnetization. There is no built-in preferred axis. The choice, in a precise sense, is an accident: it is not a fundamental law but a consequence of the system’s history. Similarly, when a liquid crystalizes, the Hamiltonian has a total translational symmetry — the energy is invariant under shifting all the particles by any arbitrary vector. The resulting crystal is not. Atoms end up sitting in particular lattice site not because of the Hamiltonian’s preference but because the crystal happened to nucleate.</p><p>There is one important caveat. Spontaneous symmetry breaking, in strict sense, can only occur in the thermodynamic limit — that is, for a system with infinitely many degrees of freedom. A finite magnet, if we wait long enough and only with small disturbance, will tunnel between the states pointing opposite directions. In infinite system, the barrier between these states is infinite, and the tunneling time is infinity. As a result, the system is fixed into a single symmetry-broken state. This is why macroscopic objects have definite, stable properties.</p><p><strong>The Order Parameter</strong></p><p>Once we accept that the phase can have less symmetry than its Hamiltonian, we need a method to quantify which kind of symmetry is broken and how much. This is the role of the order parameter.</p><p>The order parameter is zero in disordered, symmetric phase and measurable non-zero quantity in well-ordered, symmetry-broken phase. It must also transform non-trivially under the symmetry that is broken: the fact that its value changes when we perform the broken symmetry operation is precisely what makes it a marker of the broken symmetry.</p><p>Two typical kinds of example precisely show this.</p><p>For a ferromagnet, the order parameter is magnetization, the net magnetic moment per unit volume. Above Curie temperature, the spins are randomly oriented thus the net magnetic moment is zero. Below it, they are ordered in certain direction, producing the finite magnetization. The magnetization is odd in time — reversing all the currents flips all the magnetic moments — and this is the symmetry that ferromagnet breaks.</p><p>For a crystal, the order parameter is less intuitive but equally precise. The liquid has complete translational symmetry. The average density is same everywhere. In crystal, on the other hand, is periodic: high in the lattice sites and low in between. A natural way to quantify this is the Fourier transform of the density: take average density in every point in space, compute its Fourier components, and look at the amplitude at wavevectors matching the crystal’s periodicity. In liquid, those Fourier components are zero for any nonzero wavevector. In the crystal, they are not. Moreover, the phase of these Fourier components — which tells you where exactly the density peaks are located — shifts when you translate the system. The fact that the order parameter is not invariant under translations, even though the Hamiltonian is, is the telltale signature of broken translational symmetry.</p><p>These two cases look physically very different. A ferromagnet is characterized by aligned spins, whereas a crystal by periodic density modulations. Yet they share a deep structural similarity: in both cases, a phase of matter possesses less symmetry than its microscopic Hamiltonian, and in both cases there is a well-defined quantity that measures the degree of ordering.</p><p><strong>The Landau Free Energy</strong></p><p>The Landau framework provides a systematic way to describe how a system transitions between the symmetric and broken-symmetry phases. The central concept is a free energy written as a function of the order parameter.</p><p>For a ferromagnet, denote the order parameter — which is local magnetization — by <em>m</em>. Near the critical temperature, <em>m</em> is small, so we can expand the free energy in powers of <em>m</em>. The time reversal symmetry of the Hamiltonian forbids odd powers — the free energy must look the same if we flip <em>m</em> to <em>-m</em>, since the two states are energetically degenerate. The free energy takes the form:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FaM44n4y3i63zxY1f3SvZgz&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FaM44n4y3i63zxY1f3SvZgz" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/73425bb248d09a603b5a74b0f215d32a/href">https://medium.com/media/73425bb248d09a603b5a74b0f215d32a/href</a></iframe><p>Let’s truncate the free energy at the fourth power and impose <em>b &gt; 0</em> (f should go to infinity as m goes to infinity). The crucial feature is the temperature dependence of <em>a</em>: it changes sign at the critical temperature <em>Tc</em>. When a is positive, the only minimum of <em>f</em> is at <em>m = 0 </em>— the system is disordered. When a turns negative, two symmetric minima appear at nonzero m — the system spontaneously selects one of them, breaking the symmetry.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wKBjTq2d542YFJKXLjwBow.png" /></figure><p>Minimizing the free energy below <em>Tc</em> gives the spontaneous magnetization:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7VD6EmW8RpeambgbSPmwZF&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7VD6EmW8RpeambgbSPmwZF" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/247e1e2bd0efe7553e656edec76c9acd/href">https://medium.com/media/247e1e2bd0efe7553e656edec76c9acd/href</a></iframe><p>The 1/2 term is called a critical exponent. Real ferromagnets, measured in the laboratory, have an exponent closer to 1/3. The discrepancy signals that fluctuations matter near the critical point — mean-field theory treats every spin as feeling the same average environment, which is accurate far from <em>Tc</em> but breaks down when fluctuations become large. Nonetheless, the Landau framework captures the right qualitative structure.</p><p>The Landau paradigm’s power lies in what it unifies. The same double-well landscape, the same temperature-dependent coefficient, the same mean-field exponent — all of these appear not just in ferromagnets but in crystallization, in liquid-vapor transitions, in the superfluid transition of helium, and in many other systems that look physically unrelated. All these systems turn out to be doing the same thing: a Hamiltonian with a certain symmetry is settling into a phase that breaks that symmetry, and the free energy near the transition can be expressed by the same expansion.</p><p><strong>Goldstone Modes</strong></p><p>The Landau framework has a striking implication whenever the broken symmetry is <em>continuous</em> — when the order parameter can vary smoothly, like a magnetization vector that can point in any direction on a sphere.</p><p>In such cases, the broken-symmetry state comes with a family of degenerate ground states related by the symmetry transformation. For example, we can smoothly rotate a ferromagnet’s magnetization through a continuous range of directions, and each orientations costs no energy (at least without applied fields). A long-wavelength oscillation that slowly rotates the magnetization direction — a spin wave — therefore requires vanishingly little energy as the wavelength grows. This is the Goldstone theorem: whenever a continuous symmetry is spontaneously broken, there must be at least one gapless collective mode, a “Goldstone mode”. The Goldstone mode is nothing but the long-wavelength fluctuation of the order parameter, and spin wave in a ferromagnet are exactly this.</p><p>The same story applies to the crystal. Translational symmetry is continuous, and it is broken when the crystal forms. The corresponding Goldstone modes are phonons, the long-wavelength sound waves of the lattice.</p><p>One important caveat: the Goldstone theorem applies only to broken continuous symmetries. The Ising ferromagnet breaks only a discrete symmetry — you can either flip all spins or not, with no continuous rotation in between — and it has no Goldstone mode.</p><p><strong>Ginzburg-Landau theory</strong></p><p>In 1950, Ginzburg and Landau proposed a description of superconductors which was reminiscent of ferromagnetism. Superconducting order parameter <em>ψ</em> is a complex-valued quantity which vanishes above <em>Tc</em> and is nonzero below it. Its magnitude measures the degree of superconducting order, just as the magnetization magnitude measures the degree of ferromagnetic order. From the BCS theory, which came several years later, <em>ψ </em>can be understood as a one-particle wave function describing the center-of-mass motion of a Cooper pair.</p><p>The Ginzburg-Landau (GL) free energy for the superconductor is structurally a Landau expansion in <em>ψ</em>:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FegBH6e2mmcz2CxC1Qk6F9T&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FegBH6e2mmcz2CxC1Qk6F9T" width="298" height="160" frameborder="0" scrolling="no"><a href="https://medium.com/media/478906216605f5328f29822fa202e9ef/href">https://medium.com/media/478906216605f5328f29822fa202e9ef/href</a></iframe><p>Here <strong><em>A</em></strong> is the electromagnetic vector potential, <em>m*(= 2m)</em> and <em>e*(= 2e)</em> are the effective mass and charge of the Cooper pair, and <strong><em>B</em></strong><em> </em>is the magnetic field. This looks like a natural generalization of the ferromagnetic free energy at first glance — a quadratic term, a quartic term. This parallel is real and useful. The GL theory makes a quantitative predictions — the penetration depth of magnetic fields into a superconductor, the coherence length over which <em>ψ </em>varies — that agree well with experiments.</p><p>One of the most striking consequences concerns the complex phase of <em>ψ. </em>We can write the order parameter <em>ψ </em>as</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6kZyghVnsU8xGZu3Ed5wdJ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6kZyghVnsU8xGZu3Ed5wdJ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b4692ff8df3f57a423300daefe3b61a5/href">https://medium.com/media/b4692ff8df3f57a423300daefe3b61a5/href</a></iframe><p>For any closed loop in the superconductor, <em>ψ</em> in the initial and final point should be the same, which requires the phase <em>φ</em> to be an integer multiple of <em>2π</em>. This implies that the magnetic flux through any closed loop must be quantized: the flux can only take values that are integer multiples of fundamental constant. Flux quantization has been observed experimentally with great precision and provides compelling evidence for the validity of the complex order parameter description of the superconductor.</p><p><strong>The Critical Distinction Between Ferromagnet and Superconductor</strong></p><p>While the GL free energy appears formally similar to the ferromagnetic case, it does not strictly map onto it. According to the Appendix, the ferromagnetic free energy is:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FntzvRuik6oc9shCvTCEJQg&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FntzvRuik6oc9shCvTCEJQg" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/a3c0750e2259c7dca6a0b245f319aa86/href">https://medium.com/media/a3c0750e2259c7dca6a0b245f319aa86/href</a></iframe><p>Here, the gradient term is simply an ordinary squared gradient of the order parameter. In the GL free energy for the superconductor, it is not that simple. The gradient term of GL free energy:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FvfjE41hwfj73FC49zWLYSD&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FvfjE41hwfj73FC49zWLYSD" width="298" height="192" frameborder="0" scrolling="no"><a href="https://medium.com/media/11c361056f4b99d37d69a9b57d376486/href">https://medium.com/media/11c361056f4b99d37d69a9b57d376486/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FgxqUgNqeGpZEqi9mweiv4V&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FgxqUgNqeGpZEqi9mweiv4V" width="298" height="160" frameborder="0" scrolling="no"><a href="https://medium.com/media/c0e26d444285847781f395ba27bcd1d0/href">https://medium.com/media/c0e26d444285847781f395ba27bcd1d0/href</a></iframe><p>Putting</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6kZyghVnsU8xGZu3Ed5wdJ&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6kZyghVnsU8xGZu3Ed5wdJ" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/b4692ff8df3f57a423300daefe3b61a5/href">https://medium.com/media/b4692ff8df3f57a423300daefe3b61a5/href</a></iframe><p>into the formula yields:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FsNnReMAXw86Qu8dMGAEq3s&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FsNnReMAXw86Qu8dMGAEq3s" width="298" height="160" frameborder="0" scrolling="no"><a href="https://medium.com/media/8ebc5012be3fafa6c8937de034ee3ade/href">https://medium.com/media/8ebc5012be3fafa6c8937de034ee3ade/href</a></iframe><p>Thus, in the GL free energy, the gradient is coupled with the electromagnetic vector potential <strong><em>A</em></strong>. This coupling is not a cosmetic detail. Suppose the phase of the order parameter is shifted by arbitrary amount:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fv2h7a4LGXrZ9obWXJuBjJy&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fv2h7a4LGXrZ9obWXJuBjJy" width="298" height="112" frameborder="0" scrolling="no"><a href="https://medium.com/media/9ddf878c284d45148efb8b6026d736ee/href">https://medium.com/media/9ddf878c284d45148efb8b6026d736ee/href</a></iframe><p>Although the gradient of Λ seemingly changes the free energy at first sight, it can be canceled out through an appropriate gauge transformation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F77gbiMB6bpPivaTHFyCJvx&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F77gbiMB6bpPivaTHFyCJvx" width="298" height="112" frameborder="0" scrolling="no"><a href="https://medium.com/media/6a2ca87a4013cceeebded4bb423ebb77/href">https://medium.com/media/6a2ca87a4013cceeebded4bb423ebb77/href</a></iframe><p>Consequently, we can arbitrarily shift the phase of the order parameter locally without changing the free energy. Compare this to the ferromagnet. Rotating spins locally makes the difference in the free energy. The only way to perform rotation without changing it is to do it globally.</p><p>This contrast — local vs. global — has a deep consequence. In BCS theory, the mean-field ground state is conventionally described as breaking U(1) symmetry — the phase symmetry. This seems similar to the ferromagnet case. However, we saw that the gradient of the superconductor’s order parameter <em>ψ</em> is coupled with <strong><em>A</em></strong>, and any local phase variation of <em>ψ</em> could be compensated by gauge transformation of <strong><em>A</em></strong><em>. </em>This means that any spatially varying phase variation is physically unobservable. In conclusion, the states in the U(1) space are not physically distinct; they are the same state, just described with different gauge conventions. And according to Elitzur’s theorem, which was developed in 1975, the gauge redundancy cannot be spontaneously broken. Therefore, the superconductor cannot be treated as spontaneous symmetry breaking, and thus there is not corresponding Goldstone mode.</p><p><strong>Beyond The Landau Framework</strong></p><p>The Landau framework is powerful. After it was developed, it was widely believed that every distinct phase of matter could be identified by the symmetry it breaks — that symmetry was the complete classification. Then, Elitzur’s theorem seems to crack this seemingly impenetrable paradigm. Moreover, in 1982, the fractional quantum Hall effect was discovered, and the absolute authority of the Landau framework was shattered.</p><p>The FQHE ground state, as we will discuss in later columns, has no broken symmetry; rather it has same symmetry as its disordered counterpart. And yet it is unmistakably a distinct phase of matter with exotic properties that have no analog in any Landau-ordered phase. What organizes these phases is topology rather than symmetry. A completely new organizing principle — topological order — is needed to resolve this issue.</p><p>For now, the takeaway is this: Landau gave us a language — symmetry, order parameter, spontaneous symmetry breaking — that has been one of the most productive organizing frameworks in the history of physics. Understanding it deeply, as this column has tried to show, is not merely historical. It is the prerequisite for recognizing when that language fails, and for appreciating the new language that must replace it.</p><p><strong>Appendix</strong></p><p>We know that the ferromagnetic Landau free energy is written as:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FaM44n4y3i63zxY1f3SvZgz&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FaM44n4y3i63zxY1f3SvZgz" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/73425bb248d09a603b5a74b0f215d32a/href">https://medium.com/media/73425bb248d09a603b5a74b0f215d32a/href</a></iframe><p>But this is actually the free energy of a homogeneous system, where <em>m</em> is same everywhere. Let’s now consider the fluctuation of <em>m </em>over the system.</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FiehRWpruwtucCL4AH8HC3B&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FiehRWpruwtucCL4AH8HC3B" width="298" height="128" frameborder="0" scrolling="no"><a href="https://medium.com/media/40708c89ee4722338dc3d1ef5ca92b68/href">https://medium.com/media/40708c89ee4722338dc3d1ef5ca92b68/href</a></iframe><p>Suppose that fluctuation is small, so we can ignore the high-order derivative term.</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FwKMJ9ANjgmXWKX8BZAS54T&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FwKMJ9ANjgmXWKX8BZAS54T" width="298" height="224" frameborder="0" scrolling="no"><a href="https://medium.com/media/0b75ce2d006f564945661cbce6a4ba17/href">https://medium.com/media/0b75ce2d006f564945661cbce6a4ba17/href</a></iframe><p>A Taylor expansion on <em>mo</em> yields:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fs7jr7mfhgREtHnwakY5Md1&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fs7jr7mfhgREtHnwakY5Md1" width="298" height="144" frameborder="0" scrolling="no"><a href="https://medium.com/media/571308dd00e75d83c8e74f1e19f98e55/href">https://medium.com/media/571308dd00e75d83c8e74f1e19f98e55/href</a></iframe><p>i) The first term is simply a value of <em>f(m=mo)</em></p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F8e3o2biQimQ6Zdto12xVqS&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F8e3o2biQimQ6Zdto12xVqS" width="298" height="112" frameborder="0" scrolling="no"><a href="https://medium.com/media/ffe453900ce982b34a631f8ff7f3d9d9/href">https://medium.com/media/ffe453900ce982b34a631f8ff7f3d9d9/href</a></iframe><p>ii) Free energy has inversion symmetry — if we flip the position vector <em>r </em>to <em>-r</em>, then ∇ flips to -∇. Under this inversion operation, the free energy changes if the value of <em>α</em> is non-zero. Therefore, the second term vanishes.</p><p>iii) Let’s think of <em>γ</em> as energy cost when <em>m</em> changes along certain direction. The <em>j=i</em> terms are directional stiffness, which means the energy cost when m changes along <em>i</em> axis. So it is straightforward that these terms must be equal for all of the axis — there is no preferred magnetization direction.</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F6zvTKbuQ2gCakFjn5bxJpT&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F6zvTKbuQ2gCakFjn5bxJpT" width="298" height="112" frameborder="0" scrolling="no"><a href="https://medium.com/media/affe252d32c6bbfc76bc00935f103d10/href">https://medium.com/media/affe252d32c6bbfc76bc00935f103d10/href</a></iframe><p>For i ≠ j terms, we can get hint from the mirror symmetry. When we flip the direction of the x axis,</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fuw4orDnXB12yyyEMnpkzB3&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fuw4orDnXB12yyyEMnpkzB3" width="298" height="224" frameborder="0" scrolling="no"><a href="https://medium.com/media/71a0ffc44ce8d235b3f86b99d720c482/href">https://medium.com/media/71a0ffc44ce8d235b3f86b99d720c482/href</a></iframe><p>We can see the value of stiffness became negative while the absolute value is unchanged. If the value of <em>γ</em> is non-zero, then it means the energy cost is changed just by flipping coordinate. Thus:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2Fnq1ri4JaAy9BVeVrbVegsF&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2Fnq1ri4JaAy9BVeVrbVegsF" width="271" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/4e61f57244465f92992a8b296e789e8e/href">https://medium.com/media/4e61f57244465f92992a8b296e789e8e/href</a></iframe><p>Therefore, the Landau free energy of ferromagnet:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F2TbWDBP5TXsdcNhyEdb93s&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F2TbWDBP5TXsdcNhyEdb93s" width="298" height="192" frameborder="0" scrolling="no"><a href="https://medium.com/media/bcdf0e8ea5eea28f2e8c35f50ef0650e/href">https://medium.com/media/bcdf0e8ea5eea28f2e8c35f50ef0650e/href</a></iframe><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2F7UdzCjdnFfruzkvW86v8U6&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2F7UdzCjdnFfruzkvW86v8U6" width="298" height="128" frameborder="0" scrolling="no"><a href="https://medium.com/media/b7223a74bc451c381dbe12727892366b/href">https://medium.com/media/b7223a74bc451c381dbe12727892366b/href</a></iframe><p><strong>References</strong></p><ul><li><em>Girvin &amp; Yang, Modern Condensed Matter Physics (2019)</em></li><li><em>Ashcroft &amp; Mermin, Solid State Physics (1976)</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=464d7f03cea2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Column 2 — Limitations of the Band Theory]]></title>
            <link>https://medium.com/@grhncgzsvb/column-2-limitations-of-the-band-theory-e02530331d5e?source=rss-b40ef157b96d------2</link>
            <guid isPermaLink="false">https://medium.com/p/e02530331d5e</guid>
            <category><![CDATA[condensed-matter-physics]]></category>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[quantum-physics]]></category>
            <dc:creator><![CDATA[Changbin Bae]]></dc:creator>
            <pubDate>Wed, 06 May 2026 00:51:04 GMT</pubDate>
            <atom:updated>2026-05-06T00:51:04.581Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Band theory is one of the great successes of solid-state physics — built on the radical premise that electrons ignore one another. Landau’s Fermi liquid theory justified that it works well even in the presence of electron-electron interactions, by showing how the interactions merely ‘dress’ electrons into quasiparticles. This column traces that justification, then turns to where it fails: Mott insulators, superconductors, and stranger phases beyond.</em></p><h4><strong>Mott Insulator</strong></h4><p>I have mentioned before in column 1 that the band theory has one critical assumption: it does not take into account the interactions between electrons, treating them as independent. Let’s think of a situation that directly challenges this assumption. In a material with strongly localized orbitals and large electron-electron repulsion, the dominant energy is not the kinetic energy that delocalizes electrons into Bloch waves — it is the electrostatic cost (<em>U</em>) of putting two electrons on the same site. If <em>U</em> is large enough compared to the bandwidth, — which is a measure of how much kinetic energy the electron gains by hopping between sites — then the electrons prefer to sit one per site, locked in place by mutual repulsion, rather than delocalize into a metallic state.</p><p>This physics is captured cleanly by the Hubbard model, which is the minimal model for this competition:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FdEMqNt8oPy94pjPjpV6JqW&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FdEMqNt8oPy94pjPjpV6JqW" width="200" height="96" frameborder="0" scrolling="no"><a href="https://medium.com/media/8acfbd766e703cf300efacf1f5088aa4/href">https://medium.com/media/8acfbd766e703cf300efacf1f5088aa4/href</a></iframe><p>The first term describes nearest-neighbor hopping with amplitude <em>t</em> — the kinetic energy that drives delocalization. The second term penalizes double occupancy with energy <em>U</em>. The ratio <em>U/t</em> controls which tendency wins. When <em>U/t</em> is small, the hopping dominates and the system is metallic. When <em>U/t</em> is large, the repulsion wins, and each electrons freezes onto its site. Charge excitations — moving electron from a singly occupied site to a doubly occupied one — cost an energy comparable to <em>U</em>, producing an effective insulating gap. This is a Mott insulator.</p><p>Mott insulator cannot be understood with the single-particle framework. The Mott insulating gap is not in the band structure — it arises from correlations between electrons, from the many-body physics that band theory explicitly ignores.</p><h4><strong>The Real Question: Is the Band Theory Useless?</strong></h4><p>How strong, then, must the electron-electron interactions be to break down the single-particle framework? Landau made a point that what matters is whether the ground state of the interacting system can be continuously connected (adiabatically connected) to the non-interacting ground state as you slowly turn on the interactions.</p><p>Imagine starting with a non-interacting Fermi sea and smoothly dialing up the interaction strength from zero to its physical value. If no phase transition occurs along the way — no level crossing, no sudden rearrangement of the ground state — then there is a one-to-one correspondence between the states of the interacting system and those of the non-interacting system. The labels we put on free-electron states (e.g., quantum number k, the band index, spin) can still be used for the interacting states. This interacting system is then called as a Fermi liquid.</p><p>Landau called these objects that carry the labels quasiparticles. A quasiparticle is not a bare electron — it is an electron dressed by a cloud of interactions with all the other electrons around it. It is a collective excitation of the many-body system. Still, it carries well-defined quantum numbers (momentum, spin), it obeys the Pauli exclusion principle.</p><h4><strong>How Long Does a Quasiparticle Last?</strong></h4><p>The quasiparticle picture only makes sense if quasiparticles have a long enough lifetime to be well-defined objects. If an electron excitation immediately decays into a complicated many-body mess, labeling it and tracking it would be useless. It turns out that quasiparticles near the Fermi surface have fairly long lifetimes in the low-temperature regime due to the Pauli exclusion principle.</p><p>For an electron in state <em>k</em> above the Fermi surface to scatter, it must find a second electron to scatter with, and both electrons must scatter into empty final states. At T = 0K, all states below the Fermi surface are filled and all states above are empty. If the initial electron has energy only slightly above the Fermi energy <em>E_F</em>, the phase space available for all of these constraints to be satisfies simultaneously — both of the initial states occupied, final states empty, and total energy and momentum conserved — shrinks to essentially none. (Both of the final states must lie above the Fermi surface, while the scattering partner must be below. From the perspective of energy conservation, an electron with small excitation energy simply lacks the energy to complete this process.)</p><p>At the Fermi surface, the available phase space is exactly zero: no scattering is possible. The lifetime of a quasiparticle right at <em>E_F </em>is infinite. Moving slightly above the Fermi surface to energy <em>E_1 &gt; E_F</em> opens up phase space proportional to the square of energy difference, giving a scattering rate:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fmath.embed.fun%2FqWDw1WRUGFW6XvWds7JWi9&amp;type=text%2Fhtml&amp;schema=embed&amp;display_name=Embed+Fun&amp;src=https%3A%2F%2Fmath.embed.fun%2Fembed%2FqWDw1WRUGFW6XvWds7JWi9" width="213" height="151" frameborder="0" scrolling="no"><a href="https://medium.com/media/962e89e0a6388ffc9714bcefbdd418a1/href">https://medium.com/media/962e89e0a6388ffc9714bcefbdd418a1/href</a></iframe><p>This is the key result. The scattering rate goes to zero quadratically as you approach the Fermi surface in energy. And crucially, this happens regardless of how strong the interactions are — the phase-space suppression is a kinematic consequence of the Fermi surface, not a statement about interaction strength. This is why band theory, with its independent electron approximation, is self-consistently justified for states near <em>E_F</em>.</p><p>To wrap up, the band theory works not because interactions are small but because the Fermi surface geometry kinematically suppresses scattering for low-energy excitations. The theory knows this implicitly when it focuses on electrons near <em>E_F</em>. Landau made it explicit.</p><h4><strong>When Even Fermi Liquid Theory Breaks Down</strong></h4><p>Fermi liquid theory is a powerful framework, but it carries a condition: the ground state of the interacting system must be adiabatically connected to the non-interacting ground state. When that connection is broken — when a phase transition intervenes as interactions are turned on — the quasiparticle description fails entirely, and something new and often more interesting takes its place.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CziTyFMnlZyYBFSA8QWVXg.jpeg" /></figure><p>One classic example is superconductivity. In a Fermi liquid with even a weak net attractive interaction between electrons, the Fermi surface is unstable against the formation of Cooper pairs — bound electron pairs with opposite momenta and spins. The pairing instability changes the ground state discontinuously: an excitation gap opens at the Fermi surface, and the concept of individual quasiparticles moving through the system is replaced by a condensate of pairs. The superconducting state cannot be built perturbatively from the Fermi liquid, no matter how carefully you add corrections.</p><h4><strong>The Question the Rest of the Series Answers</strong></h4><p>Take stock of where we are. Band theory, built on Bloch’s theorem, correctly classifies most materials and gives a quantitative account of their electronic properties — an achievement that still impresses. Fermi liquid theory provides the deep reason this works: even in the presence of strong interactions, phase-space constraints near the Fermi surface protect quasiparticles from rapid decay, and the independent-electron picture remains valid as long as the ground state is adiabatically connected to the Fermi sea.</p><p>But that connection breaks in materials where correlations are strong enough to drive a phase transition: Mott insulators, superconductors, and more exotic phases where the entire quasiparticle concept fails. These materials are not edge cases. They include some of the most technologically important and scientifically fascinating systems known.</p><p>The rest of this series examines what happens in those materials. What are the theoretical tools that replace band theory when it fails? What new phases of matter emerge from strong correlations, and what are their signatures? And what does it mean for a quantum state of matter to be fundamentally different from a free electron gas?</p><p>The band theory story is the starting point. The interesting physics begins where it ends.</p><p><strong>References</strong></p><ul><li><em>Girvin &amp; Yang, Modern Condensed Matter Physics (2019)</em></li><li><em>Ashcroft &amp; Mermin, Solid State Physics (1976)</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e02530331d5e" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>