<?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 T E M I D A Y O on Medium]]></title>
        <description><![CDATA[Stories by T E M I D A Y O on Medium]]></description>
        <link>https://medium.com/@oyewodayo?source=rss-28541316e682------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*gb7OToxcVlXi9Sf2WxG2dQ.jpeg</url>
            <title>Stories by T E M I D A Y O on Medium</title>
            <link>https://medium.com/@oyewodayo?source=rss-28541316e682------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 22:30:18 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@oyewodayo/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[Quantum twin of a frustrated magnet: Simulating a Real Frustrated Magnet TmMgGaO₄ (thulium…]]></title>
            <link>https://medium.com/@oyewodayo/quantum-twin-of-a-frustrated-magnet-simulating-a-real-frustrated-magnet-tmmggao%E2%82%84-thulium-4c14e2c33dc8?source=rss-28541316e682------2</link>
            <guid isPermaLink="false">https://medium.com/p/4c14e2c33dc8</guid>
            <category><![CDATA[quantum-mechanics]]></category>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[quantum-computing]]></category>
            <category><![CDATA[condensed-matter-physics]]></category>
            <category><![CDATA[research]]></category>
            <dc:creator><![CDATA[T E M I D A Y O]]></dc:creator>
            <pubDate>Fri, 24 Jul 2026 19:58:34 GMT</pubDate>
            <atom:updated>2026-08-13T05:25:05.469Z</atom:updated>
            <content:encoded><![CDATA[<h3>Quantum twin of a frustrated magnet: Simulating a Real Frustrated Magnet TmMgGaO₄ (thulium magnesium gallium oxide) on a Neutral-Atom Processor</h3><p><strong>Challenge designed with / inspired by: </strong>the neutral-atom quantum-simulation programme of Pasqal SAS (Palaiseau, France), and in particular the 256-qubit one-to-one simulation of TmMgGaO₄ (<a href="https://arxiv.org/html/2603.20372v1">Leclerc <em>et al.</em>, arXiv:2603.20372</a>).</p><p>From a simple square-lattice antiferromagnet to a 49-atom stand-in for TmMgGaO₄, and the point where the classical emulator gives up</p><p><strong>Building a Quantum Twin of TmMgGaO₄</strong></p><p><em>A </em><a href="https://quantum.cern/"><em>CERN Quantum Technology Initiative</em></a><em> (QTI) hackathon.</em></p><p>Analogue quantum simulators are usually benchmarked against toy Hamiltonians — clean, generic models chosen because they’re easy to write down, not because they correspond to anything you could put under a microscope. The CERN QTI hackathon challenge we took on asked a sharper question: can a <a href="https://en.wikipedia.org/wiki/Neutral_atom_quantum_computer">neutral-atom quantum</a> processor act as a <strong>quantum twin</strong> of an actual material — reproducing real lab measurements, atom for atom?</p><p>The material in question is thulium magnesium gallium oxide (<strong>TmMgGaO₄)</strong>, a triangular-lattice frustrated magnet, and the inspiration was a genuinely striking result: <a href="https://www.pasqal.com/">Pasqal’s</a> 256-qubit Rydberg simulator recently reproduced the magnetisation curve of real TmMgGaO₄ crystals, one-to-one (Leclerc et al., <a href="https://arxiv.org/abs/2603.20372">https://arxiv.org/abs/2603.20372</a>). The 2D transverse-field Ising model is natively implemented by Rydberg arrays, without Trotterisation and without gate-synthesis overheads. The low-energy physics of TmMgGaO₄ is precisely that model on a triangular lattice. The device and the material are the same language.</p><p>The challenge walked us through that same path in three phases: warm up on a simple, well-understood square-lattice antiferromagnet, build the real quantum twin, then push into territory where classical simulation starts to break.</p><p>All code is on <a href="https://github.com/oyewodayo/tmmggao4_quantum_twin">GitHub</a> — three standalone Pulser notebooks under `<a href="https://github.com/oyewodayo/tmmggao4_quantum_twin/tree/main/notebooks">notebooks</a>/`.</p><p><strong>Why triangular lattices are interesting at all</strong></p><p>Put antiferromagnetically coupled spins on a square lattice and the ground state is straightforward: alternate up and down in a checkerboard pattern so every antiferromagnetic bond is satisfied. Put the same spins on a triangular lattice and that becomes impossible. Around any triangle, at most two of the three antiferromagnetic bonds can be satisfied simultaneously, leaving one bond inevitably frustrated. This is <strong>geometric frustration</strong>.</p><p>The key consequence isn’t just one unsatisfied bond — it is that there are enormously many ways to arrange the spins with exactly the same minimum energy. Trying to relieve frustration on one triangle simply shifts it to another. Instead of a unique ordered ground state, the system develops a highly degenerate manifold of competing states. That leaves weak perturbations — such as quantum fluctuations, longer-range interactions, or disorder — with an outsized role in determining the material’s behaviour, making triangular-lattice magnets fertile ground for phenomena such as spin liquids, competing magnetic orders, and the rich phase diagram observed in <strong>TmMgGaO₄</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/624/1*HPjQUtyCP74HExXh4ihjUg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tAzfK7zNAj2dBOY3SlrPMw.jpeg" /></figure><p>TmMgGaO₄’s Tm³⁺ ions occupy exactly this triangular lattice. Strong crystal-field anisotropy causes each ion to behave, at low energies, as an effective spin-1/2 with a preferred quantisation axis. This low-energy description is what allows a rare-earth oxide to be modelled using Pauli spin operators. The resulting Hamiltonian (Eq. 1 of Leclerc <em>et al.</em>) is:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/725/1*gLuuLP0X8l99Pn_oR7aZFg.jpeg" /></figure><p><strong>Each term has a clear physical meaning.</strong> <strong>J₁</strong> is the frustrated nearest-neighbour antiferromagnetic interaction. A much smaller next-nearest-neighbour coupling, <strong>J₂ ≈ 0.05J₁</strong>, lifts the extensive classical degeneracy created by frustration and stabilises the characteristic <strong>1/3-magnetisation plateau</strong> (↑↑↓ order) rather than the simple Néel state found on an unfrustrated lattice. The term <strong>Δₓ ≈ 1.08J₁</strong> is an <em>intrinsic</em> transverse field arising from residual crystal-field mixing. Unlike the applied field, it is a fixed property of the material rather than an experimental control parameter, and it is far from negligible. Finally, <strong>Δ_z</strong> is the externally applied longitudinal magnetic field, related to the laboratory field through <strong>Δ_z/J₁ ≈ 1.543 μ₀H(T)</strong>. As the applied field is varied, the competition between frustration, exchange interactions, and quantum fluctuations drives the system through a quantum phase transition between the frustration-stabilised <strong>1/3-magnetisation plateau</strong> and the fully field-polarised paramagnetic phase.</p><p><strong>Phase 1 — Warm-up: proving the toolchain on a textbook antiferromagnet</strong></p><p>Before mapping a real material onto Rydberg atoms, we reproduced the cleaner landmark result this whole approach is built on: Scholl et al.’s adiabatic preparation of a 2D antiferromagnet on a square array (Nature 2021, <a href="https://arxiv.org/abs/2012.12268">https://arxiv.org/abs/2012.12268</a>).</p><p>The physical picture underneath all three phases starts here. Pasqal’s neutral-atom platform traps individual Rb atoms in optical tweezers and uses each one as an effective spin-1/2: the electronic ground state is spin down, and a laser-excited Rydberg state – enlarged and strongly interacting – is spin up. The global laser drive Ω competes with the Rydberg blockade, the fact that two nearby atoms can’t both be excited at once. Sweeping the detuning δ from negative to positive tips the energetic balance from “stay in the ground state” to “excite if you can”, and the blockade turns that competition into an alternating checkerboard pattern: exciting every atom is forbidden by blockade, but exciting every <em>other</em> atom isn’t.</p><p>The recipe: pick a blockade radius in the checkerboard window (a &lt; Rb &lt; a√2), ramp the Rabi frequency (Ω) up, sweep the detuning δ quasi-adiabatically from negative to positive, ramp back down to freeze the readout basis. Watch checkerboard order build.</p><p>![Phase 1: pulse controls, order parameters building during the sweep, residual fluctuations, and the final site-resolved spin pattern for a 3×3 array.](../notebooks/phase1_afm_result.png)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wBawruOw-fRBDnpm2lI1Jg.png" /></figure><p>Both the staggered magnetisation |m_stag| and the full structure factor Sᶻᶻ(π,π) climb together as the sweep crosses δ=0, and the final site-resolved ⟨σᵢᶻ⟩ pattern (rightmost panel) is a clean checkerboard. This was the deliberately unglamorous phase — the point was to prove the Pulser toolchain, the pulse-sequence logic, and the emulator round-trip all work end to end <em>before</em> spending any of that confidence on the harder mapping. It also surfaces something worth knowing early: the naive linear ramp that works cleanly at 3×3 does <strong>not</strong> trivially scale to 4×4 — the many-body gap shrinks, and this is flagged in the notebook as a guarded, optional cell rather than swept under the rug.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mUhpExljfOvHw9hUEP5vYw.png" /></figure><p>The notebook also carries a genuine escape hatch past that local ceiling: a guarded pipeline that submits the same detuning scan for a real 25-atom register to Pasqal Cloud’s EmuMPS backend at two bond dimensions, with persistent batch IDs so a re-run never silently resubmits duplicate jobs — the same safety pattern used everywhere we talk to the cloud in this project.</p><p><strong>Phase 2a — The real thing: a quantum twin of TmMgGaO₄</strong></p><p>This is the centrepiece. We map the material Hamiltonian above onto a Rydberg array via the paper’s Eq. 4:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/725/1*8UNVkELmROPEXpMuyNAVqw.jpeg" /></figure><p>Where δU is not an analytic shortcut but the register’s actual, numerically-computed interaction offset: the full pairwise van der Waals sum over <em>every</em> atom pair in the array (not just the 6 nearest neighbours), averaged over all N=49 sites — at our register’s spacing this comes out to δU ≈ 10.3·J₁. This isn’t a numerical convenience; it’s an operator-level identification H_QPU ≈ α_QPU·H_TMGO between the QPU’s native van-der-Waals-plus-drive Hamiltonian and the material’s, the actual content of the “quantum twin” claim.</p><p>We built the paper’s own smallest reported bulk size — a 7×7, N=49 register on Pulser’s `MockDevice` rather than claiming a one-to-one match to Pasqal’s real FM1 hardware: `MockDevice` gets the <em>dimensionless</em> ratio Δx/J₁=1.08 exactly right, but its C₆ coefficient differs from the physical Rydberg state, so the lattice spacing that comes out (~7.84 μm) isn’t the paper’s own ~9 μm. We state this explicitly in the notebook rather than accepting a seemingly plausible number as a hardware match. Similarly, we evolve all 49 atoms but read observables off the central 5×5=25-site region with a single buffer layer of protection: a larger, statistically friendlier analysis window than the paper’s own two-layer-buffer convention, at the cost of somewhat more boundary sensitivity. Both choices are flagged, not buried.</p><p><strong>Strategy.</strong> Rather than one continuous sweep, we ran the paper’s own point-by-point protocol: an independent quasi-adiabatic preparation (Rabi ramp-up, ~4 μs field sweep, fast ramp-down for readout) for each of 27 target fields spanning Δz/J₁ ∈ [1,10], at two bond dimensions (χ=50 for the full scan, χ=100 as a convergence check on points closest to the transition) — real jobs submitted to Pasqal Cloud’s EmuMPS backend, with persistent batch IDs so re-running the notebook never resubmits duplicates. Every reported point also carries a 68% bootstrap interval from resampling the finite measurement shots, so the error bars you see are honest finite-sampling uncertainty, not decoration.</p><p><strong>The quantum-twin magnetisation curve</strong></p><p>Before bringing the real material into it at all, the twin’s own curve is the first deliverable in its own right: bulk Mᶻ against Δz/J₁, at both bond dimensions, with 68% bootstrap error bars on every point.</p><p>![Quantum-twin bulk magnetisation vs. Δz/J₁, χ=50 (main scan) and χ=100 (convergence check), with 68% bootstrap error bars.](../notebooks/phase2a_magnetisation_curve.png)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/980/1*uYT_hTjxq5JiatAbUg3Y_g.png" /></figure><p>The curve runs from Mᶻ ≈ 1 deep in the paramagnet down toward the 1/3-plateau as Δz/J₁ decreases, with χ=50 and χ=100 tracking each other closely wherever both were run — the first, on-its-own-terms convergence check, before the curve is judged against anything external.</p><p><strong>Comparing against the real material</strong></p><p>The genuinely satisfying part: we digitised the actual experimental AC-susceptibility curve from Fig. 1e of the paper’s own PDF (pixel-calibrated against the plot’s own axis tick marks — not fabricated placeholder data), and overlaid it directly against the bootstrap-resampled magnetisation curve.</p><p>![Our N=49 quantum twin against the real, digitised TmMgGaO₄ AC-susceptibility curve. Both curves agree closely at the two classical endpoints and diverge through the genuinely quantum-critical middle.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/690/1*P7zUJPI7x3I95TT24U77Dw.png" /></figure><p>As a sanity check on the digitisation itself: the extracted curve’s elbow sits at Δz/J₁ ≈ 4 and its inflection near Δz/J₁ ≈ 6 — exactly where the paper’s own text places the quantum critical point and the classical crossover, respectively. That’s independent confirmation the digitisation is trustworthy before we even use it. Reading the comparison honestly means reading it alongside the two caveats above: any residual gap between our twin and the real crystal in the intermediate, quantum-critical region is exactly where finite size, the single buffer layer, and the `MockDevice` approximation are expected to matter most — the two classical endpoints, where those effects bite least, are where we’d expect (and see) the closest agreement.</p><p>We also computed the connected structure-factor diagnostic Sᶻᶻ(q_1/3) (the paper’s Fig. 2B) as a secondary check, and pushed further than a qualitative look: a systematic varying-window cubic-fit procedure (17 candidate field windows, each fit accepted only if it clears a minimum-point count and its extremum isn’t sitting on the window’s edge) extracts an N=49 <strong>pseudo-critical crossover </strong>— explicitly not the thermodynamic critical point, which the paper itself only claims at its largest system, N=256, where “bulk behaviour dominates&quot;. The spread across accepted windows becomes a fit-window-sensitivity band, kept visually and numerically distinct from the bootstrap sampling error bars — two different uncertainties that are easy to conflate and important not to.</p><p>Connected 1/3-order structure factor vs. Δz/J₁, at two bond dimensions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/690/1*3FG_b9Au2eT9Zr5DVZPBlQ.png" /></figure><p>Reporting a crossover <em>with its window-sensitivity band attached</em>, rather than a single clean-looking critical field, is the more honest deliverable: it tells a reader exactly how much of that number is physics and how much is which fitting window you happened to pick.</p><p><strong>Phase 2b — Probing the classical frontier with a quench</strong></p><p>Phase 2a is a fidelity check — DMRG and QMC also prepare ground states well, so equilibrium reproduction alone doesn’t prove a quantum device was <em>*necessary*</em>. The open-exploration phase of the challenge asks a different question: where does a quantum simulator stop being merely convenient and start being the only tool that works?</p><p>The paper’s answer is post-quench dynamics (their Fig. 4). Tensor-network emulators like `emu_mps` compress a many-body state using a bond dimension χ; after a sudden quench, entanglement grows in time, so the χ needed for an accurate description grows with it — until no achievable χ is enough, and the classical emulation simply fails. A quantum device compresses nothing, so it keeps going exactly where the emulator can’t.</p><p><strong>Protocol</strong> (mirroring the paper’s Ext. Dat. Fig. 4b/S2B): prepare near the paramagnetic product state |↑…↑⟩ at large Δz/J₁, then abruptly square-pulse-quench Δz/J₁ toward or inside the 1/3-ordered phase, and hold — tracking the nearest-neighbour correlator</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/725/1*q61JmR5p-KOVlaZMhU9LNA.jpeg" /></figure><p>exactly as the paper defines it. One subtlety worth flagging for anyone reusing this code: `emu_mps.CorrelationMatrix` returns <em>occupation</em> correlations ⟨nᵢnⱼ⟩, not ⟨σᵢᶻσⱼᶻ⟩ directly. Since σᶻ = 1−2n, the correct non-connected conversion is ⟨σᵢᶻσⱼᶻ⟩ = 1 − 2⟨nᵢ⟩ − 2⟨nⱼ⟩ + 4⟨nᵢnⱼ⟩ — not the simpler factor-of-4 shortcut, which is only valid for the <em>connected</em> covariance used in Phase 2a’s structure factor. Small distinction, easy to get quietly wrong.</p><p>The notebook splits this into two complementary parts, deliberately at two different system sizes: <strong>Part A</strong> goes small enough (N=9) that we can check thermalisation against an <em>exact</em> reference with zero stochastic error; <strong>Part B</strong> goes to the paper’s own N=49 specifically because N=9 is too small to show a classical method struggling at all.</p><p><strong>Part A — An exact thermal reference, for free (N=9)</strong></p><p>The paper compares long-time dynamics against a thermal ensemble at an effective temperature fixed by energy conservation (their Eq. 8), computed via a stochastic-series-expansion (QMC-SSE) sampler — out of scope to build in hackathon time. But at our system size (N=9 for the small demo), we could do something strictly better: solve for the exact thermal ensemble via full diagonalisation of H_QPU, with zero stochastic error, and compare directly. Bisecting in β=1/T correctly finds <strong>negative effective temperatures**</strong> when the quench lands inside the 1/3-ordered phase — the same regime the paper itself flags, and our T_eff/J₁ = −1.55 matches their reported kBT/(ħJ₁) = −1.25 at a similar quench target in sign and order of magnitude.</p><p>![Post-quench C₁ᶻᶻ(t) at three bond dimensions (visually identical — expected at N=9, where the Hilbert space is small enough that D=16 is already numerically exact) against the exact thermal reference line.](../notebooks/phase2b_quench.png)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*z7IiCpWueAGBh3M2WENDtQ.png" /></figure><p>Immediately after the quench, C₁ᶻᶻ drops and oscillates in a band that straddles the thermal reference line rather than sitting far above it or collapsing cleanly onto it — a legitimate intermediate case: the correlator is visibly relaxing toward the thermal value, but at N=9 and tJ₁ ≲ 12 we’re watching coherent oscillations decay, not a settled late-time plateau. Extending the hold time on <strong>Pasqal Cloud</strong> out to tJ₁≈36 answers the obvious next question: the oscillations <em>don’t </em>damp out — they persist with roughly constant amplitude, in a band that sits mostly below the thermal line rather than shrinking onto it. That’s not a bug either. Eigenstate thermalisation is a statement about systems large enough to act as their own heat bath; at N=9, unbuffered, there simply aren’t enough degrees of freedom for that, so the closed system shows persistent quasi-periodic revivals instead. This is exactly why the paper makes its thermalisation claim at N=100–256, not here — thermalisation is an emergent, large-N phenomenon, and a genuine test of it needs a system big enough to exhibit it.</p><p><strong>Part B — Where does the classical emulator actually break? (N=49)</strong></p><p>Part A is reassuring but not yet the point: at N=9 the Hilbert space is small enough that exact diagonalisation <em>*is*</em> the reference, which is precisely why it can’t show a classical method failing. Part B scales to the paper’s own N=49 register (reusing Phase 2a’s device/mapping cells, so this notebook stays standalone) and sweeps both quench hardness (Δz/J₁) and bond dimension χ ∈ {50,100,200,300}, using the <em>*disagreement between*</em> χ values as a measurable proxy for where the classical description stops being trustworthy.</p><p>The result: curves overlap at early times (weak entanglement, every χ still accurate), separate through an intermediate window, then reconverge at late times. The widest pairs — |χ300−χ50| and |χ300−χ100| — peak well above a divergence threshold through tJ₁≈5–10, while the narrowest high-χ pair, |χ300−χ200|, barely grazes it, meaning χ≥200 is close to converged and the low-χ divergence is a genuine departure from a trustworthy reference, not two poor approximations disagreeing with each other. The rise is the real signal — the onset of classical intractability, scaling cleanly with the χ gap. The rapid decay afterward is a <strong>finite-size recurrence, not classical catch-up</strong>: at 49 atoms the state partially returns toward its starting point, entanglement drops, and even χ=50 works again. At the paper’s 256 qubits, that recurrence is pushed far outside the observation window and the breakdown is sustained rather than transient.</p><p>We went in expecting the classical breakdown to peak near the quantum critical point (Δz/J₁≈3.9), since entanglement generation should be strongest there. The sweep didn’t show that cleanly — divergence crosses threshold at low field <em>*and*</em> again near 4.0, but not monotonically in between. We don’t have a clean explanation, and we said so rather than smoothing it over: a likely confound is that at 49 atoms, each field value peaks and recurs at a different time, so whether a given quench clears threshold within our observation window depends partly on recurrence timing, not purely on how much entanglement it generates. Disentangling a genuine field-dependent trend from this finite-size artifact needs larger systems, a natural next step and one squarely beyond a single-quench hackathon demo.</p><p><strong>What we’d do next</strong></p><p><strong>System size.</strong> Push `l_bulk` to 6/9 (N=100/169) on GPU to watch the Phase 2a pseudo-critical crossover sharpen toward the paper’s own N=256 estimate and to separate genuine field-dependence from finite-size recurrence in Phase 2b Part B.</p><p><strong>Bond dimension.</strong> The paper still finds D=256 not fully converged at their largest system — always report a D-scan alongside any headline curve, not just a single “converged-looking” run.</p><p><strong>The real FM1 device geometry.</strong> Swapping `MockDevice` for the paper’s actual Rydberg level/C₆ would close the gap between our dimensionless-ratio match and a genuine one-to-one hardware reproduction.</p><p><strong>A real QMC-SSE thermal sampler</strong> would remove the N≲18 ceiling that exact diagonalisation imposes on the Phase 2b Part A thermal reference, and is the paper’s own approach at scale.</p><p><strong>Why this matters beyond one hackathon</strong></p><p>The interesting claim here isn’t “we ran Pulser.” It’s that a neutral-atom processor’s <em>native</em> Hamiltonian, van der Waals interactions plus a Rabi drive and detuning, maps almost operator-for-operator onto the microscopic physics of a real, characterised crystal. That’s a qualitatively different kind of benchmark than reproducing a generic textbook model: it’s a check against independent magnetometry on an actual material, with a clear, falsifiable target. And the Phase 2b result — honest about where it worked, where it didn’t, and why — is the more useful kind of hackathon deliverable: not a number forced to look clean, but a documented account of exactly which physics is a real signal and which is a finite-size artifact still waiting on a bigger system to resolve.</p><p><em>Code, </em><a href="https://github.com/oyewodayo/tmmggao4_quantum_twin/tree/main/notebooks"><em>notebooks</em></a><em>, and full derivations: see the </em><a href="https://github.com/oyewodayo/tmmggao4_quantum_twin"><em>repository</em></a><em>. Key references: Leclerc et al., </em><a href="https://arxiv.org/abs/2603.20372"><em>arXiv:2603.20372;</em></a><em> Scholl et al., </em><a href="https://www.nature.com/articles/s41586-021-03585-1"><em>Nature 595, 233 (2021)</em></a><em>; Vovrosh et al., </em><a href="https://arxiv.org/abs/2511.19340"><em>arXiv:2511.19340 </em></a><em>and arXiv:2511.20388; </em><a href="https://arxiv.org/abs/2104.15044"><em>Silvério et al., Pulser, Quantum 6, 629 (2022).</em></a></p><p><strong>References</strong></p><p>Reference material for simulating the frustrated triangular-lattice antiferromagnet TmMgGaO₄ on a Rydberg atom array.</p><ol><li>Cevallos, Stolze, Kong &amp; Cava, <a href="https://arxiv.org/abs/1710.07707"><em>Anisotropic magnetic properties of the triangular plane lattice material TmMgGaO₄</em></a>, Mater. Res. Bull. <strong>105</strong>, 154 (2018).</li></ol><p>Discovery and synthesis. Crystal structure; Tm³⁺ on a triangular net.</p><p>2. Shen et al., <em>Intertwined dipolar and multipolar order in the triangular-lattice magnet TmMgGaO₄</em>, Nat. Commun. <strong>10</strong>, 4530 (2019). <a href="https://arxiv.org/abs/1810.05054">arXiv:1810.05054</a></p><p>Neutron scattering. Three-sublattice order gives a magnetic Bragg peak at the <strong>K point</strong> — the order parameter the simulation must reproduce.</p><p>3. Li et al., <a href="https://arxiv.org/abs/1912.02344"><em>Partial up-up-down order with the continuously distributed order parameter in the triangular antiferromagnet TmMgGaO₄</em></a>, Phys. Rev. X <strong>10</strong>, 011007 (2020).</p><p>Magnetisation data for comparison against emulation.</p><p><strong>The model</strong></p><p>4. Chen, <a href="https://journals.aps.org/prresearch/abstract/10.1103/PhysRevResearch.1.033141"><em>Intrinsic transverse field in frustrated quantum Ising magnets</em></a>, Phys. Rev. Research <strong>1</strong>, 033141 (2019).</p><p>Why a non-Kramers ion’s two-singlet ground state acts as an intrinsic transverse field.</p><p>5. Liu, Huang &amp; Chen, <em>Intrinsic quantum Ising model on a triangular lattice magnet TmMgGaO₄</em>, Phys. Rev. Research <strong>2</strong>, 043013 (2020). <a href="https://arxiv.org/abs/1909.03608">arXiv:1909.03608</a></p><p>Establishes TmMgGaO₄ as the transverse-field Ising antiferromagnet on a triangular lattice. <strong>The basis of the whole mapping.**</strong></p><p><strong>Frustration and order by disorder</strong></p><p>6. Wannier, <em>Antiferromagnetism. </em><a href="https://journals.aps.org/pr/abstract/10.1103/PhysRev.79.357"><em>The triangular Ising net</em></a>, Phys. Rev. <strong>79</strong>, 357 (1950). Erratum: Phys. Rev. B <strong>7</strong>, 5017 (1973).</p><p>Extensive ground-state degeneracy; residual entropy ≈ 0.323 k_B/site. The classical model never orders.</p><p>7. Moessner &amp; Sondhi, <a href="https://journals.aps.org/prb/abstract/10.1103/PhysRevB.63.224401"><em>Ising models of quantum frustration</em></a>, Phys. Rev. B <strong>63</strong>, 224401 (2001).</p><p>8. Isakov &amp; Moessner, <a href="https://arxiv.org/abs/cond-mat/0302105"><em>Interplay of quantum and thermal fluctuations in a frustrated magnet</em></a>, Phys. Rev. B <strong>68</strong>, 104409 (2003).</p><p>[7]+[8]: how a transverse field selects three-sublattice order via quantum order by disorder.</p><p>9. Li et al., <em>Kosterlitz–Thouless melting of magnetic order in the triangular quantum Ising material TmMgGaO₄</em>, <a href="https://www.nature.com/articles/s41467-020-14907-8">Nat. Commun. <strong>11</strong>, 1111 (2020)</a>.</p><p>10. Hu et al., <em>Evidence of the Berezinskii–Kosterlitz–Thouless phase in a frustrated magnet</em>, Nat. Commun. <strong>11</strong>, 5631 (2020).</p><p>Experimental counterpart to [9].</p><p>11. Dun et al., <em>Neutron scattering investigation of proposed Kosterlitz–Thouless transitions in the triangular-lattice Ising antiferromagnet TmMgGaO₄</em>, Phys. Rev. B <strong>103</strong>, 064424 (2021). <a href="https://arxiv.org/abs/2011.00541">arXiv:2011.00541</a></p><p>Inelastic neutron scattering fixes the effective Hamiltonian parameters. <strong>Source for numerical values of J and Γ.</strong></p><p><strong>Rydberg platform</strong></p><p>12. Scholl et al., <em>Quantum simulation of 2D antiferromagnets with hundreds of Rydberg atoms</em>, Nature <strong>595</strong>, 233 (2021). <a href="https://arxiv.org/abs/2012.12268">arXiv:2012.12268</a></p><p>196 atoms; square <strong>and</strong> triangular arrays. The warm-up target.</p><p>13. Leclerc et al. (PASQAL), <em>One-to-one quantum simulation of the low-dimensional frustrated quantum magnet TmMgGaO₄ with 256 qubits</em>, arXiv:2603.20372 (2026). <a href="https://arxiv.org/abs/2603.20372">arXiv:2603.20372</a></p><p>Closest existing work. 256-qubit Rydberg simulator of the TmMgGaO₄ Hamiltonian; magnetisation matches single-crystal susceptibility. Keeps both J₁ and J₂; probes the paramagnet → 1/3-order (√3×√3) transition; adds snapshot analysis and quench dynamics.</p><p><strong>Tools</strong></p><p>- <a href="https://github.com/pasqal-io/Pulser">Pulser</a> — pulse-level programming <a href="https://pulser.readthedocs.io/">docs</a></p><p>- emu-mps / emu-sv — tensor-network and state-vector emulators</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4c14e2c33dc8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When Things Break, Physics Begins]]></title>
            <link>https://medium.com/@oyewodayo/when-things-break-physics-begins-b79f1cc4903a?source=rss-28541316e682------2</link>
            <guid isPermaLink="false">https://medium.com/p/b79f1cc4903a</guid>
            <category><![CDATA[innovation]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[physics]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[history-of-science]]></category>
            <dc:creator><![CDATA[T E M I D A Y O]]></dc:creator>
            <pubDate>Sun, 24 May 2026 07:01:01 GMT</pubDate>
            <atom:updated>2026-05-24T07:01:01.213Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LXQOAY18_Gq2PV41cF0GYw.png" /></figure><p>We often view physics through the prism of its victories, the clean lines of textbooks, the elegant symmetry of equations, and the absolute predictability of a satellite launch. It gives the impression that physics is a grand architecture of perfect order, built to conquer a messy universe.</p><p>But that gets the story entirely backward. Physics doesn’t exist to eliminate chaos. It is fueled by it.</p><p>If the universe were perfectly tidy, physics would have died in 1900. Back then, the prevailing sentiment in science was dangerously close to “we’re basically done here.” The laws of Newton and Maxwell seemed to have wrapped reality up in a neat bow. Lord Kelvin famously remarked that only “two small clouds” remained on the horizon of physics, which were the minor anomalies that couldn’t quite be explained.</p><p>Those two small clouds turned out to be hurricanes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*26DI3K44ipY2YA7zOC-vNw.png" /></figure><p><strong>The Beautiful Disasters that Rebuilt Reality</strong></p><p>The first was a <a href="https://en.wikipedia.org/wiki/Catastrophe_theory">mathematical catastrophe</a>. Classical physics dictated that a simple heated object should radiate infinite energy at higher frequencies—predicting that turning on an oven would blast you with lethal X-rays. Not only did the math fall short, it fell so absurdly short it got its own name, the ultraviolet catastrophe. When faced with a broken equation and no clean way out, Max Planck proposed something radical: that energy doesn’t flow like a smooth river but arrives in chunky, discrete packets called quanta. Quantum mechanics did not arise out of inspiration. It was born from desperation.</p><p>The second anomaly was the speed of light. Experiments showed it didn’t care how fast you were moving; it always traveled at exactly the same speed, which broke Newtonian mechanics completely. Einstein did not dismiss this as an error that should be ignored. He regarded it as an absolute truth, enjoyed the confusion, and dismantled our absolute concepts of space and time. In reality, physics was saved by what appeared to be its collapse.</p><p>This pattern holds throughout the history of the discipline. Atomic structure shouldn’t have been stable; classical electromagnetism predicted orbiting electrons would spiral inward and collapse. They don’t. That impossibility demanded an answer, and quantum mechanics delivered one. Every era of physics has its own version of this story: a result that refuses to fit, a contradiction that won’t go away, a discovery that breaks everything and rebuilds it better.</p><p><strong>The Deterministic Wild</strong></p><p>Physics didn’t just use chaos as a stepping stone; it eventually had to study chaos as a feature of nature itself. This is the central irony of <a href="https://en.wikipedia.org/wiki/Chaos_theory">chaos theory</a>: perfectly strict, deterministic laws can yield entirely unpredictable results.</p><p>The <a href="https://en.wikipedia.org/wiki/Butterfly_effect">butterfly effect</a> is more than just a metaphor for time travel movies. It’s a mathematical reality for nonlinear systems. A weather model can have the correct equations, but changing a starting value by a fraction of a decimal place, the equivalent of a butterfly flapping its wings, changes the forecast from a sunny afternoon to a snowfall. Unpredictability here isn’t a failure of intelligence or a gap in data. It is a fundamental design feature of complex systems, from ocean currents to the rhythm of a human heart. Understanding it has quietly transformed the practical world: better aircraft design, more secure communications, sharper climate models.</p><h3>The Modern Engines of Anxiety</h3><p>Today, physics is alive and productive precisely because it is deeply challenged. We live in a world with a fundamental, unresolved contradiction: Our best theory of the very large, General Relativity, cannot communicate with our best theory of the very small, Quantum Mechanics. This is a stark and unresolved contradiction that we live with. They speak different languages, employ different mathematical systems, and openly disagree about the basic principles of how the universe functions. <a href="https://en.wikipedia.org/wiki/String_theory">String theory</a>, loop quantum gravity, and many approaches yet to be named are the result of decades of work in search of reconciliation.</p><p>Add to that the universe’s most humbling fact: everything we can see, touch, and detect accounts for only 5% of all that exists. The remaining 95% is a placeholder for what we call Dark Matter and Dark Energy. We don’t actually know what they are. We only know they’re there because the universe keeps refusing to behave according to our current models.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bfygfdUq-o3-D35Ekf-zTA.png" /></figure><p>These aren’t failures. These are the front lines.</p><h3>The Mess Behind the Method</h3><p>The way this feels on the inside is something that is rarely covered in textbooks. It is common for physicists to spend years, even careers, pursuing concepts that eventually fail. Results from experiments don’t match any predictions. Equations that should yield answers become unresolvable. Progress isn’t a steady climb; it’s a long, disorienting walk through uncertainty, with clarity arriving in unpredictable flashes. The discipline advances not because researchers calmly fill in expected blanks but because they’re willing to remain in the confusion long enough and stubborn enough to question the assumptions everyone else stopped questioning. Creativity in physics isn’t a product of comfort. It’s a product of sustained intellectual pressure.</p><p>The universe is not a clockwork machine waiting to be wound up. It is dynamic, turbulent, and beautifully uncooperative. The moment physics runs out of problems, it stops. Until then, it will keep growing not by finding comfort in what is certain, but by standing right at the edge of the breaking point and asking what comes next.</p><p>Without chaos, there is no curiosity. Without problems, there is no progress. The history of physics makes this undeniable: the greatest discoveries have always emerged from the deepest confusion. Chaos is not the enemy of physics. It is its most reliable engine.</p><p>Follow <a href="https://x.com/oyewodayo">Temidayo</a> on <a href="https://x.com/oyewodayo">X</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b79f1cc4903a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Quantum Handshake: Forces, Spin, and the Secret Architecture of Matter]]></title>
            <link>https://medium.com/@oyewodayo/the-quantum-handshake-forces-spin-and-the-secret-architecture-of-matter-b862f2e9d0b1?source=rss-28541316e682------2</link>
            <guid isPermaLink="false">https://medium.com/p/b862f2e9d0b1</guid>
            <dc:creator><![CDATA[T E M I D A Y O]]></dc:creator>
            <pubDate>Sun, 08 Feb 2026 05:43:58 GMT</pubDate>
            <atom:updated>2026-02-08T05:43:58.419Z</atom:updated>
            <content:encoded><![CDATA[<h3>Why doesn’t the universe fall apart?</h3><p>Right now, as you read this, trillions upon trillions of protons in your body are crammed together in atomic nuclei. Each one is screaming with positive charge. Each one desperately “wants” to flee from its neighbors. And yet, here you are. Solid. Stable. Alive.</p><blockquote>At the deepest level, the answer lies in quantum physics, specifically the strong nuclear force and the quantum mechanical property known as spin.</blockquote><p>The reason has nothing to do with luck. It’s because of two profound quantum phenomena working in perfect concert: an invisible force so powerful it defies everyday physics and an intrinsic property of matter so strange that it determines the very architecture of reality.</p><p>This is the story of the <a href="https://en.wikipedia.org/wiki/Hideki_Yukawa">Yukawa</a> force and quantum <a href="https://en.wikipedia.org/wiki/Spin_(physics)">spin</a>. The handshake that holds the universe together.</p><h3>Part I: The Crisis at the Heart of the Atom</h3><h3>The Problem Nobody Could Solve</h3><p>By the early 1930s, physicists had reached an impasse. They knew atoms had nuclei. They knew these nuclei contained protons (particles with positive electric charge) packed into a space smaller than a trillionth of a millimeter.</p><p>And therein lay the paradox.</p><p>Positive charges repel each other. Violently. The electromagnetic force should have torn every atomic nucleus apart the instant it formed. Yet atoms existed. Stars burned. Matter persisted.</p><p>Something was holding the nucleus together with a grip far stronger than electromagnetism could break. But what?</p><h3>Yukawa’s Audacious Solution</h3><p>In 1935, a young Japanese physicist named <a href="https://en.wikipedia.org/wiki/Hideki_Yukawa">Hideki Yukawa</a> proposed something radical: a new force of nature, one that had somehow escaped detection because it only worked at impossibly short distances.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/568/1*avZg1Zs848Z79m87YkTnGQ.jpeg" /></figure><p>Think of it as a quantum <a href="https://en.wikipedia.org/wiki/Bungee_cord">bungee cord</a>. Yukawa theorized that protons and neutrons stay bound together by constantly exchanging a mediator particle (what he called the “meson,” later identified as the pion). This exchange creates an attractive force, but one with a crucial twist:</p><p><strong>It only works when particles are almost touching.</strong></p><p>The mathematical heart of Yukawa’s force looks like this:</p><p>$$V(r) = -g² \frac{e^{-m_{\pi}r}}{r}$$</p><p>That exponential term (e raised to a negative power) is the key. It causes the force to plummet to near-zero once particles separate beyond about one femtometer. That’s 10⁻¹⁵ meters, roughly the width of a proton.</p><h3>Why This Matters</h3><p>This is profoundly different from gravity or electromagnetism, which reach across the cosmos. The Yukawa force is local, intense, and possessive. It’s the reason atomic nuclei can exist at all. It’s also the reason you can’t crush atoms together with your bare hands.</p><p>Without this “shy” but powerful force, matter as we know it would dissolve into a formless cloud of repelling particles. There would be no chemistry. No stars. No you.</p><p>But binding matter together is only half the story. Stability also requires rules about how particles arrange themselves once they are bound</p><h3>Part II: The Dance Card of the Universe</h3><h3>If Force is the Glue, Spin is the Protocol</h3><p>The Yukawa force explains why particles <em>can</em> stick together. But it doesn’t explain <em>how</em> they organize themselves. Why do some particles crowd together while others refuse to share space? Why do atoms build themselves in the specific patterns that give us the periodic table?</p><p>For that, we need to understand <strong>quantum spin</strong>.</p><h3>What Spin Actually Is (And Isn’t)</h3><p>Here’s where things get genuinely weird.</p><p>When we say a particle has “spin,” we don’t mean it’s physically rotating like a planet or a top. An electron isn’t a tiny sphere spinning on an axis. Instead, spin is an <strong>intrinsic property</strong>. It’s as fundamental to a particle’s identity as its mass or charge.</p><p>It’s a mathematical feature baked into the quantum state of a particle. And here’s the kicker: it never changes. An electron always has spin ½. A photon always has spin 1. This value is permanent, unchangeable. It <em>is</em> the particle.</p><h3>Two Tribes: Fermions and Bosons</h3><p>This intrinsic spin divides all particles in the universe into two tribes:</p><p><strong>Fermions</strong> have half-integer spins (½, 3/2, 5/2, etc.). This includes electrons, protons, and neutrons. The building blocks of matter.</p><p><strong>Bosons</strong> have integer spins (0, 1, 2, etc.). This includes photons, gluons, and the Higgs boson. The carriers of forces.</p><p>This distinction isn’t arbitrary. It determines everything about how these particles behave.</p><h3>Part III: The Experiment That Proved It</h3><h3>Stern and Gerlach’s Silver Surprise</h3><p>In 1922 (before Yukawa, before quantum mechanics was fully formulated), Otto Stern and Walther Gerlach performed an experiment that would change physics forever.</p><p>They heated silver atoms until they vaporized, then fired a beam of these atoms through a strong, non-uniform magnetic field. Classical physics predicted the beam would smear out in a continuous gradient as the atoms’ “spinning” orientations randomized.</p><p><strong>That’s not what happened.</strong></p><p>Instead, the beam split cleanly into two distinct paths—up and down. No middle ground. No gradient. Just two discrete options.</p><p>This was the first direct evidence that spin is <strong>quantized</strong>. That it comes in fixed, discrete values rather than a continuous spectrum. Particles don’t “choose” their spin orientation smoothly; they snap into one of a limited set of allowed states.</p><p>The universe, it turned out, runs on binary code.</p><h3>Part IV: The Social Rules of Quantum Matter</h3><h3>The Pauli Exclusion Principle: No Doubles Allowed</h3><p>Because of their half-integer spin, fermions follow a strict rule discovered by Wolfgang Pauli: <strong>no two identical fermions can occupy the same quantum state.</strong></p><p>This single rule is why matter has structure. This is why you don’t fall through your chair. Think about that for a second.</p><p>When two electrons want to occupy the same orbital in an atom, they <em>must</em> have opposite spins: one “up,” one “down.” Once those two slots are filled, any additional electron must go to a higher energy level. This is the foundation of chemistry. The reason atoms bond. The reason elements have distinct properties.</p><p>Without the Pauli Exclusion Principle, all electrons would collapse into the lowest energy state, and all matter would be an undifferentiated goop.</p><h3>Hund’s Rule: The Energy Efficiency Protocol</h3><p>In larger atoms with multiple orbitals at the same energy level, electrons follow another rule: they prefer to occupy different orbitals with the <em>same</em> spin first, before pairing up with opposite spins.</p><p>Why? Because aligned spins lead to quantum states that reduce effective repulsion and lower the atom’s energy. By staying in separate orbitals with aligned spins, they minimize their mutual repulsion and keep the atom’s energy as low as possible.</p><p>Think of it as quantum social distancing. Particles prefer their own space when they can get it.</p><h3>The Exotic Frontier: Anyons and Beyond</h3><p>In three-dimensional space, particles are either fermions or bosons. But in <strong>two-dimensional systems</strong> (like electrons confined to a flat plane), something stranger becomes possible.</p><p>Enter <strong>anyons</strong>: particles with exotic spin values that don’t fit the fermion/boson binary. When <a href="https://en.wikipedia.org/wiki/Anyon">anyons</a> wind around each other, they accumulate phase shifts that depend on their path. Some anyons exhibit what is called non-Abelian statistics.</p><p>This isn’t just theoretical curiosity. <a href="https://en.wikipedia.org/wiki/Anyon">Anyons</a> are the leading candidates for building topological quantum computers. Machines that could revolutionize computing by storing information in the braids and knots of particle trajectories.</p><h3>Conclusion: The Invisible Architecture</h3><p>The universe is an act of balance.</p><p>The Yukawa force provides the crushing, intimate grip that binds protons and neutrons into nuclei, while the deeper strong interaction (quantum chromodynamics via gluon exchange) binds quarks into those particles in the first place. Quantum spin provides the organizational logic. The strict protocols that tell particles where they can and cannot exist, how they must align, and what structures they can build.</p><p>From Yukawa’s predicted pion to the split beams of Stern and Gerlach, from the elegant mathematics of Pauli to the exotic anyons trapped in 2D quantum systems, these “invisible rules” aren’t optional features of reality.</p><p><strong>They are the reason matter exists at all.</strong></p><p>The next time you touch a solid object, remember you’re feeling the Yukawa force at work. And when you look at the periodic table, remember you’re seeing the Pauli Exclusion Principle laid bare.</p><p>The universe doesn’t fall apart because of a handshake. A quantum handshake between force and spin, between binding and organization, between chaos and structure.</p><p>And that handshake is happening in every atom of your body, right now.</p><p><strong>Further reading for curious minds:</strong></p><ul><li>H. Yukawa, <em>On the Interaction of Elementary Particles</em> (1935)</li><li>R. P. Feynman, <em>The Character of Physical Law</em></li><li>J. J. Sakurai, <em>Modern Quantum Mechanics</em></li><li>D. J. Griffiths, <em>Introduction to Elementary Particles</em></li><li>A. Zee, <em>Quantum Field Theory in a Nutshell</em></li><li>Nobel Prize lectures on the strong interaction and quantum statistics</li></ul><p>For readers interested in anyons and topological quantum computing, introductory reviews by Alexei Kitaev and Xiao-Gang Wen provide accessible starting points.</p><p><em>Thanks for reading. If you enjoyed this exploration of quantum physics, I’d love to hear your thoughts in the comments.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b862f2e9d0b1" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>