Physics Concepts Explainer

AI field note: In 2025, AWS data centers used 0.12 liters of water per kilowatt-hour, over 7x more water-efficient than the industry average of 0.84. That efficiency improved even as AI pushed compute demand higher. Here's how we did it.   Cooling a data center presents a three-way tradeoff: water use, energy use, and the temperature margin that keeps servers reliable. Push hard on one and pressure shows up somewhere else. Cool with little energy and you use more water. Cool with little water and you spend more energy on chillers, which draw 25 to 35% more electricity, often when the grid is most stressed. Keep both water and energy low and the servers run warmer, closer to their limits.   We asked if the cooling threshold we had treated as fixed actually had room to move. If the system can operate safely at a higher threshold before water-assisted cooling kicks in, you can keep water and energy low without sacrificing reliability.   So we tested it. Thousands of hours of operational data across campuses showed we could safely raise that threshold, within tested operating conditions, without increasing failure rates. Water-assisted cooling now starts only around 85°F. About 90% of the time, the data centers cool with outside air alone.   The results hold at scale, not just per unit of compute. In Northern Virginia, our largest region by load, water use fell 42% in a year while capacity grew. Across the sites we own and operate, total water withdrawn fell 2% from 2024 to 2025, even as the number of buildings rose. As per-unit efficiency improved, total use went down.   On the hottest hours, when air alone isn't enough, the systems use a small amount of evaporative water rather than switching to chillers that would spike electricity demand when the grid can least absorb it. A little water during peak heat is a lower total burden on the surrounding community than a lot of electricity at the same moment.   The savings for our most common data center designs came from a lot of systems innovation, and from proving that a constraint we'd long accepted as fixed could actually move. In this era, a lot of fixed constraints are worth re-testing.

Data centers now use 415 TWh of electricity a year. By 2030, that could hit 945 TWh. Cooling alone eats 30–40% of that energy. Fans and air can't keep up with AI chips anymore. Some are looking up. Google, NVIDIA, and startups like Starcloud are exploring data centers in orbit—where solar power is constant and the vacuum of space offers free cooling. No fans. No water. But the hurdles are steep: launch costs, radiation damage, latency, and radiators the size of buildings. Behnood Bazmi looked down instead. A grad student at the University of Illinois, he wasn't chasing AI. He was studying heat. And he kept asking one question: what if cooling is the real bottleneck? His team used algorithms to design copper cooling plates no engineer would sketch. Jagged, branching fins just 30–50 micrometers thick—thinner than a human hair. Too complex for machining. Too intricate for most 3D printing. Then they partnered with Fabric8Labs to build them using electrochemical additive manufacturing at room temperature. What they measured: ↳ 32% lower thermal resistance ↳ 68% less pumping power ↳ Cooling energy drops from ~550 MW to 11 MW in a 1 GW facility ↳ 98% less energy spent keeping chips cool Space data centers may come. But this works now—on Earth, at lab scale, with a path to manufacturing. Sometimes the answer isn't a moonshot. It's a grad student asking a question everyone else stopped asking. 1 question about heat. 10 researchers bridging design and manufacturing. 100 data centers running on a fraction of the energy. What problem have you stopped questioning because it felt too obvious? Follow me, Dr. Martha Boeckenfeld, for insights on thriving as AI rises while leaders stay human. Sources: IEA, Cell Reports Physical Science (May 2026), UIUC, Fabric8Labs https://lnkd.in/euNCgcGg

All of Electrical Engineering is one thing: managing the electromagnetic field. Most textbooks start with Ohm’s law. That’s backwards. ⚡The truth: there is only one field. From one perspective, we call it electric. From another, magnetic. Not separate — just two sides of the same reality. Here’s how Maxwell broke it down (with intuition): Divergence E = ρ / ε₀ (Charges create electric fields. Think of static charge building up on a balloon and the field radiating out.) Divergence B = 0 (No magnetic monopoles exist. You can’t isolate a north pole — cut a magnet and you just get two smaller dipoles.) Curl E = -∂B/∂t (A changing magnetic field creates an electric field. This is how generators and transformers work.) Curl B = μ₀ J + μ₀ε₀ ∂E/∂t (A current creates a magnetic field, and a changing electric field also creates a magnetic field. That’s why radio signals can travel through space.) And the Lorentz force ties it all together: F = q(E + v × B) (A moving charge feels a push from E, and gets bent sideways by B. That’s the principle behind motors and particle accelerators.) 👉 Resistors, capacitors, and inductors? They’re just ways of shaping and guiding this one field. This is the very first lesson in my High-Speed Digital Design course — because once you see EE this way, high-speed stops being mysterious. ➡️ Want to learn electronics from first principles? Sign up here: https://lnkd.in/guwp-iN9 Follow HaSofu where I'll be teaching electronics from scratch, since it's been requested multiple times. Yeah...PCB design is only one of my 5 domains of expertise in EE alone...more to come. #Electronics #Engineering #PCBDesign #SignalIntegrity #Maxwell #STEM #HighSpeedDesign

"In the fall of 2022, a Princeton University graduate student named Carolina Figueiredo (opens a new tab) stumbled onto a massive coincidence. She calculated that collisions involving three different types of subatomic particles would all produce the same wreckage. It was like laying a grid over maps of London, Tokyo and New York and seeing that all three cities had train stations at the same coordinates.“They are very different [particle] theories. There’s no reason for them to be connected,” Figueiredo said. The coincidence soon revealed itself to be a conspiracy: The theories describing the three types of particles were, when viewed from the right perspective, essentially one. The conspiracy, Figueiredo and her colleagues realized, stems from the existence of a hidden structure, one that could potentially simplify the complex business of understanding what’s going on at the base level of reality. For nearly two decades, Figueiredo’s doctoral advisor, Nima Arkani-Hamed (opens a new tab), has been leading a hunt for a new way of doing physics. Many physicists believe they’ve reached the end of the road when it comes to conceptualizing reality in terms of quantum events that play out in space and time. Such language can’t easily describe the beginning of the universe, for instance, when the space-time fabric likely didn’t exist in its current form. Arkani-Hamed therefore suspects that the usual notion of quantum particles moving and interacting in space-time is an approximation of deeper, more abstract concepts, which, if found, could serve as a better language for talking about quantum gravity and the origin of the universe. A major development came in 2013, when Arkani-Hamed and his student at the time, Jaroslav Trnka, discovered a jewel-like geometric object that forecasts the outcome of certain particle interactions. They called the object the “amplituhedron.” However, the object didn’t apply to the particles of the real world. So Arkani-Hamed and his colleagues sought more such objects that would. Now Figueiredo’s conspiracy is another manifestation of abstract geometric structure that seems to underlie particle physics. Like the amplituhedron, the new geometrical method, known as “surfaceology,” streamlines quantum physics by sidestepping the traditional approach, which is to track the countless ways particles can move through space-time using “Feynman diagrams.” These depictions of particles’ possible collisions and trajectories translate into complicated equations. With surfaceology, physicists can get the same result more directly."https://lnkd.in/dSWp3W4a

The faster you move, the slower your time passes. This effect comes from Albert Einstein’s theory of special relativity, which revealed in 1905 that space and time are not separate things. Instead, they form a unified structure called spacetime. In this framework, every object in the universe is always moving through spacetime at the same overall rate. But that motion can be divided between motion through space and motion through time. If you are standing still relative to your surroundings, almost all of your motion through spacetime happens in the time direction. Your clock ticks normally. But when you begin moving through space, part of that motion shifts away from time. As your speed increases, the rate at which time passes for you slows compared with someone who is stationary. This phenomenon is called time dilation. The effect becomes noticeable only at extremely high speeds. The closer you travel to the speed of light, 186,000 miles per second (300,000 km/s), the stronger the slowdown of time. At speeds near light, time for the traveler can pass dramatically slower. A famous example is the “twin paradox.” If one twin travels on a spacecraft moving close to the speed of light while the other remains on Earth, the traveling twin would age more slowly. When they return, they would be younger than their sibling. This is not just theoretical. Experiments with atomic clocks have repeatedly confirmed the effect. In one famous experiment in 1971, scientists flew highly precise clocks on commercial airplanes. When the planes landed, the airborne clocks had lost tiny fractions of a second compared with clocks on Earth, exactly as relativity predicted. Even the satellites that power GPS must account for relativistic time differences caused by both speed and gravity. Without those corrections, navigation systems would accumulate errors of several miles each day. Einstein’s insight showed that time is not universal. It depends on how fast you move through space — making motion through the universe also a journey through time. Note: The information presented here is for general knowledge and discussion. #Connected #ConnectedCoach #SpaceTime

𝗠𝗮𝗶𝗻𝘁𝗮𝗶𝗻𝗶𝗻𝗴 𝗰𝗼𝗵𝗲𝗿𝗲𝗻𝗰𝗲 𝗶𝗻 𝘀𝘂𝗽𝗲𝗿𝗰𝗼𝗻𝗱𝘂𝗰𝘁𝗶𝗻𝗴 𝗾𝘂𝗮𝗻𝘁𝘂𝗺 𝗽𝗿𝗼𝗰𝗲𝘀𝘀𝗼𝗿𝘀 𝗶𝘀 𝗮 𝗰𝗼𝗻𝘀𝘁𝗮𝗻𝘁 𝗯𝗮𝘁𝘁𝗹𝗲. While many factors contribute to qubit decoherence, 𝗧𝘄𝗼-𝗟𝗲𝘃𝗲𝗹 𝗦𝘆𝘀𝘁𝗲𝗺 (𝗧𝗟𝗦) 𝗱𝗲𝗳𝗲𝗰𝘁𝘀 remain among the most 𝗳𝗿𝘂𝘀𝘁𝗿𝗮𝘁𝗶𝗻𝗴 𝗰𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲𝘀. 🔹 𝗧𝗵𝗲 𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝗧𝗟𝗦 𝗱𝗲𝗳𝗲𝗰𝘁𝘀, typically found in the surfaces and interfaces of superconducting circuits, can r𝗲𝘀𝗼𝗻𝗮𝗻𝘁𝗹𝘆 𝗰𝗼𝘂𝗽𝗹𝗲 𝘄𝗶𝘁𝗵 𝗾𝘂𝗯𝗶𝘁𝘀, leading to 𝗶𝗻𝗰𝗿𝗲𝗮𝘀𝗲𝗱 𝗱𝗲𝗰𝗼𝗵𝗲𝗿𝗲𝗻𝗰𝗲 𝗮𝗻𝗱 𝗴𝗮𝘁𝗲 𝗲𝗿𝗿𝗼𝗿𝘀. These defects are particularly problematic due to their spatial and temporal instability, causing 𝘂𝗻𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗮𝗯𝗹𝗲 "𝗱𝗿𝗼𝗽𝗼𝘂𝘁𝘀" 𝗶𝗻 𝗾𝘂𝗯𝗶𝘁 𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲. When it comes to mitigating TLS noise, several approaches exist: 🔹𝗛𝗮𝗿𝗱𝘄𝗮𝗿𝗲-𝗟𝗲𝘃𝗲𝗹 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗲𝘀 - 𝗠𝗮𝘁𝗲𝗿𝗶𝗮𝗹 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴: High-purity materials and advanced fabrication techniques to reduce TLS density. - 𝗦𝘂𝗿𝗳𝗮𝗰𝗲 𝗧𝗿𝗲𝗮𝘁𝗺𝗲𝗻𝘁𝘀: Minimizing lossy interfaces where TLSs often reside. - 𝗖𝗶𝗿𝗰𝘂𝗶𝘁 𝗗𝗲𝘀𝗶𝗴𝗻: Engineering qubit circuits to minimize coupling to TLSs. 🔹𝗖𝗼𝗻𝘁𝗿𝗼𝗹 & 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗧𝗲𝗰𝗵𝗻𝗶𝗾𝘂𝗲𝘀 - 𝗤𝘂𝗯𝗶𝘁 𝗙𝗿𝗲𝗾𝘂𝗲𝗻𝗰𝘆 𝗧𝘂𝗻𝗶𝗻𝗴: Shifting qubit frequencies away from TLS resonances, widely used in tunable transmon architectures. - 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗗𝗲𝗰𝗼𝘂𝗽𝗹𝗶𝗻𝗴: Pulse sequences that average out the effect of TLS noise. - 𝗔𝗰𝘁𝗶𝘃𝗲 𝗙𝗲𝗲𝗱𝗯𝗮𝗰𝗸: Real-time monitoring and adaptive qubit control. While some of these techniques come with considerable overhead, new approaches are emerging to address the TLS challenge more efficiently: 🔹𝗧𝗵𝗲 𝗧𝗜𝗖-𝗧𝗔𝗤 𝗔𝗽𝗽𝗿𝗼𝗮𝗰𝗵: 𝗔 𝗡𝗲𝘄 𝗖𝗼𝗻𝘁𝗿𝗼𝗹 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝘆 The Siddiqi group just introduced a new technique called 𝗧𝗜𝗖-𝗧𝗔𝗤 (Targeted In-situ Control of TLS and Qubits): - 𝗦𝗶𝗻𝗴𝗹𝗲 𝗖𝗼𝗻𝘁𝗿𝗼𝗹 𝗟𝗶𝗻𝗲: Provides local and independent control of each qubit’s noise environment with a single on-chip control line. - 𝗘𝗹𝗲𝗰𝘁𝗿𝗶𝗰 𝗙𝗶𝗲𝗹𝗱 𝗧𝘂𝗻𝗶𝗻𝗴: Instead of shifting the qubit frequency, TIC-TAQ dynamically tunes TLSs away from the qubit frequency by applying a local electric field. - 𝗖𝗼𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁𝗮𝗿𝘆 𝗧𝗲𝗰𝗵𝗻𝗶𝗾𝘂𝗲: Expected to enhance existing strategies for managing TLS-induced errors. 𝗧𝗜𝗖-𝗧𝗔𝗤 𝘀𝗵𝗼𝘄𝘀 𝗽𝗿𝗼𝗺𝗶𝘀𝗶𝗻𝗴 𝗿𝗲𝘀𝘂𝗹𝘁𝘀: - 36% Improvement in single-qubit error rates. - 17% Increase in qubit relaxation times (T₁). - 4x Suppression in TLS-induced performance outliers. 𝗪𝗵𝘆 𝗗𝗼𝗲𝘀 𝗧𝗵𝗶𝘀 𝗠𝗮𝘁𝘁𝗲𝗿? TLS defects are a roadblock on the path to fault-tolerant quantum computing. It’s great to see how hardware innovations and smart control techniques make a measurable impact. Are you more optimistic about hardware-based or control-based solutions for mitigating TLS noise? 📸 Image Credits: Larry Chen, Kan-Heng Lee et al. (arXiv, 2025)

The Story of Maxwell’s Equations – Easy Mode Maxwell’s equations so simple that your students can visualize them without drowning in calculus. Think of it as “the 4 laws of electromagnetism in plain language. 1. Gauss’s Law for Electricity Electric field lines start from positive charges and end at negative charges. Meaning: Charges create electric fields. More charge = more field lines coming out (positive) or going in (negative). Mathematically: Electric flux through a closed surface = total charge inside / permittivity. Example: Rub a balloon on your hair — it gets charged and an electric field radiates out. 2. Gauss’s Law for Magnetism Magnetic field lines always form closed loops — no start, no end. Meaning: No magnetic monopoles exist (so far). Every magnet has a north and south pole — cut it in half, you still get both. Magnetic field lines loop from north to south outside the magnet, and back inside. Example: Sprinkle iron filings around a bar magnet — the lines are loops. 3. Faraday’s Law of Induction A changing magnetic field creates an electric field. Meaning: Move a magnet near a coil → current flows. This is how generators work: spinning magnets change the magnetic field, producing electricity. Mathematically: The faster the magnetic change, the bigger the induced voltage. Example: Bicycles with dynamo lights — wheel spins → magnet moves → light glows. 4. Ampère–Maxwell Law A changing electric field or electric current creates a magnetic field. Meaning: Current in a wire produces a magnetic field around it. Even without a current, a changing electric field (like in a capacitor charging) also creates magnetism. This “Maxwell fix” is what predicted electromagnetic waves. Example: Electromagnets — wrap wire around iron, pass current, and you get strong magnetism. Electricity and magnetism are two sides of the same coin — change one, and you get the other. This dance gives us light, radio waves, and all of modern wireless tech. Integral form → great for explaining “flux through a surface” with diagrams. Differential form → useful for advanced students who can understand field variation at a point. You can make students remember the “curl” ones (Faraday & Ampère) as the “changing field creates the other field” laws.

Keeping a Schrödinger’s cat alive for 23 minutes does not mean a real cat was placed in a box. In quantum physics, the term refers to a superposition, a system existing in multiple states at once. In this case, scientists created a controlled quantum state and managed to preserve it far longer than usual. The achievement is about coherence. Quantum states are extremely fragile and usually collapse quickly due to interaction with the environment, a process called decoherence. Maintaining such a state for 23 minutes is a major technical breakthrough because it requires isolating the system from noise, temperature changes, and external disturbances with extreme precision. These experiments often involve systems like photons, ions, or superconducting circuits. Researchers use advanced cooling, shielding, and error correction methods to keep the system stable. The longer a quantum state can survive, the more useful it becomes for real applications in computation, sensing, and communication technologies. This progress is especially important for quantum computing. Qubits rely on superposition to perform calculations that classical computers cannot easily handle. Extending coherence time means more complex operations can be performed before errors take over, bringing practical quantum machines closer to reality in the coming years. The deeper insight is that quantum physics is moving from theory into engineering. What once sounded like a thought experiment is now something scientists can create, control, and extend in the lab. Keeping a quantum state alive for minutes shows how far precision physics has come in turning abstract ideas into measurable reality. #quantum #physics #technology #science