Enthalpy

The insulation of electric wire is usually PVC, which an alkali would not deteriorate quickly. Though, I'd prefer ETFE or PVDF insulation, which exist, among others for satellites. I'd worry more about the electrodes than the insulator. Is your goal to separate Sn from Pb in an alloy? Electricity is expensive, electrolysis is slow. Did you consider distillation instead? Pb boils at 1649°C (1atm) Sn boils at 2602°C (1atm) http://www.webelements.com/tin/index.html that makes an efficient spread, easy to use (reduce the pressure), faster and costing far less than the electrolysis, avoiding chlorine.

An electron-positron pair can annihilate precisely because they overlap. This is the fundamental requirement of any interaction. Annihilation doesn't require small dimensions nor points. Think of two photons which have a locally destructive interference: they do that over some extension, in some experiments wide enough to be seen with the naked eye. No need for points there. Yes, particles are created. This may start in a small volume, for instance a photon created by an atom is initially as small as the atom, about 10,000 times smaller than the wavelength, and the wave extension is even much bigger. But why should this need a point photon? It needs the capability for the photon to be small, as any wave can. More generally, an interaction extends as much as the overlapping of both particles, that is often the smaller particle. This provides no means to tell if a particle is a point. It does tell - and this is a historical reason for QM - that an interaction can occur in a volume much smaller than a particle has been, with some properties of this particle fully available in the small volume, albeit with a saller probability for a small volume. You write: "finding where the point particle was", and, no, an interaction does not reduce the position uncertainty to a point. "Collapse" is often misinterpreted, I'd prefer "reduction" or "adaptation". ---------- A beta electron is hugely wider than an atom. After hitting an atom and ejecting an electron from there, the beta continues its path from there. If the collision is at small parameter impact, the beta deviates, but this is scarce because the cross section is much smaller than an atom. Simplified: at one hydogen radius, the electrostatic energy is two Ryberg or 26eV. When a collision implies 260keV the distance between both electrons is 1/10,000 of an atom radius, so such a collision happens every 100,000,000 atoms. Most encounters hence happen at a bigger impact parameter, so they imply a much smaller energy, and the deflection of the beta is much smaller - the trajectory is about straight. As the ionizing particle loses energy, its path makes more zigzag since both (a) efficient collisions have a bigger section (b) the particle have less momentum - and this is observed. This is not a reason for a point particle. It does need that some interactions be small, here 1/10,000 of an atom radius for electrons and 260keV, but this is still an extension and not a point. Far better: if an interaction could happen in a point instead of a volume, the particles would leave in all directions as a result. The very relationship between the energy and the angle after a collision tells that every collision has a volume. This is necessary to limit the diffraction. By the way, the scarce efficient collisions, which imply a big momentum change, don't localize both particles more than the smallest was before. Only their relative positions are known if we observe the deflection, but the position of the collision is as uncertain as before - for instance "within the atom" for a target electron and "much larger that that for a beta, resulting in "somewhere in the atom" for the collision. Entanglement, yes. Only some subsequent event, say one collision with a betetr localised nucleus, would refine the position of both particles. ---------- You put "an alpha has a few fm" and this is only the extension of its constituents. The delocalization of the alpha is huge: it's emitted in any direction before some interaction reduces the uncertainty, for instance when the alpha meets a first atom, or the daughter atom hits an other one. These are different dimensions. Yet an other dimension would be the collision cross-section (sqrt as you wish) with, for instance, a helium nucleus. Still an other, the distance at which the electric field can create virtual electron-positron pairs. The only difference with an electron is that the alpha, the proton, a nucleus, an atom... are composite particles, where the constituents define this extra radius, area or volume where they're located. This extra volume can be smaller or bigger than the delocalisation - extremely delocalised atoms in a Bose-Einstein condensate for instance. Is this any reason for a point electron? I've seen no reason in interactions for a point electron. But in the lack of "self-interaction", yes: the electron doesn't screen itself against other particles.

Thanks! You imagine things waaaay more subtle than I've done! I'm a coarse engineer, you know: when a 16mm diameter screw breaks I put a 24mm one. I've corrected the energy as the mass, that is in a first-order approximation, and only to compare with the departure from the Z2 law. At Scandium (I have no data beyond) the correction is 1.2% so I expect the second order around 0.01% which indeed approaches the 0,46 vs 0,50 of 1.2%. Only one magnitude below, it would be the next necessary refinement. Is the mass (or mass + kinetic energy) correction applied once for all positions of the electron, or individually for each position? (Which would already be too subtle for me). Have you ever seen the inertia of the electrostatic energy injected in the hydrogen-like atom? I check only 1s orbitals here, that is, the last ionisation of every atom. Would there be any significant coupling with the spin?

A different way exists to compute induced voltages: it's the vector potential A http://en.wikipedia.org/wiki/Magnetic_potential less concrete than B, less often useful than d(flux)/dt, but sometimes invaluable, especially when the flux makes little sense. It's convenient when Biot-Savart can compute it as: http://de.wikipedia.org/wiki/Biot-Savart-Gesetz (the en fr es pt it articles have only B, not A) dA( r) = (µ/4pi)*(I*dL)/|r| (here A and dL are vectors, r if you want) and then the induced electric field is just -dA/dt. Nice and simple to evaluate parasitic couplings. It belongs to the computational toolbox of electromagnetic compatibility. Now, your wire passes by the coil's axis if I get it properly, and is perpendicular to the coil's axis.Just because the created E is parallel to the produced A which is parallel to I*dL, we can tell that the coil's symmetry prevents it from inducing an electric field in said wire. The parts of the turns symmetric above and below the plane conduct the same current at the same distance to the target wire and the same cosine but in opposite directions, so each complete turn has zero net effect. Again, we find that if some current flows through a closed circuit that contains the described target wire, it will result only from the rest of the circuit. This time, we can say more precisely that any voltage would be induced elsewhere in the circuit, not in the described wire, and that the return circuit would have to be above or under the plane of symmetry to get an induced voltage. The flux wouldn't give naturally an answer as detailed, the induction B not as simply. ---------- I realize only now that you've written "the field is along X", and this is an impossibility. The field and the induction make closed loops, so they have to change their direction. My answers hold for a magnet oriented along X. A field approximately along X would be possible within an electromagnet, not outside. For instance in Helmholtz' coils http://en.wikipedia.org/wiki/Helmholtz_coil or within a long solenoid. In this case as well, the return circuit would decide everything.

Waves have no definite size (nor shape). In the case of electron pairs and positronium (which has the size of an atom), the size of the wave would fit perfectly. Just like a photon of 1µm wavelength and square meters or square light-years spread interacts with a 100pm atom. A trace in a cloud chamber is wide, hugely wider than an atom. It doesn't need an electron smaller than an orbital. Nor does electron emission or capture by a nucleus need an electron smaller than the nucleus. It needs the ability for an electron to concentrate to this size; this is the probability that we compute with |Ψ|2, and call it "collapse" when the wave shrinks to fit the interacting particle. This ability to change the size wouldn't alone need the particle to be a point. Every interaction is computed over the common volume of both waves, that is, the ability of the bigger wave to shrink to the size of the smaller wave suffices. More: when a photon is absorbed by a semiconductor, it does not act as a point. It has to act on an electron that spans over >1000 atoms before and after the absorption. It's even worse for X-ray diffraction, where the photon must interact with a crystal and not with one atom. A different formulation would be whether we need something smaller than the wave to represent the particle, and my answer is yes for the particle's electric self-energy, and apparently no for the interactions.

Yes, that's it. Ferromagnetic materials can be "hard" or "soft" (historically observed at hard and soft steel, and you bearing steel is very hard), meaning that is takes some extra field, hence energy, to reverse their induction, and this is nearly independent of time. Because hysteresis losses (in joule per magnetization cycle) don't depend on the frequency, but eddy currents create power losses proportional to F2 (at mechanical and mains frequency domain), hysteresis is the more probable at 60rpm. Yes, you can understand it as torque in a rolling ball (in addition to understanding it as energy), where the ball's magnetization isn't aligned with the external field. You might perhaps feel with your hands a very small angle where the bearing has a spring behaviour before acting as a brake; this won't happen with eddy currents. And the torque will be felt even under 60rpm. The ball bearing has a remarkable shape where the contact between the balls and the races has a small area that concentrates the magnetic flux to a high induction, and extremely hard steel that is a bit difficult to demagnetize, though it's not a good permanent magnet neither. Hey, if you ilke such bizarre magnetic setups, I believe Branly's coherer still needs an explanation 130 years after it was introduced! http://en.wikipedia.org/wiki/Coherer

Hello you all! Models for the hydrogen atom and hydrogen-like often tell that the relativistic correction acts on the electron's mass as a consequence of the kinetic energy. Though, the kinetic energy equals half the (negative) electrostatic energy, so if the electron carried half of the electrostatic energy, the effect would cancel out the kinetic energy. What do you think? To check that, I took the energy of last ionisation for varied elements so the relativistic effect varies, there http://www.webelements.com/hydrogen/atoms.html hoping the data is measured and accurate enough, since the effect is at most 1%, and tinkered a spreadsheet RelativisticHydrogen.zip (expand, open with Gnumeric, Excel or equivalent). The first correction was to compensate by 1+m/M the effect of light nuclei to approach the Z2 ionisation energy law of immobile nuclei. Seems to work well. Then I already get a relative excess of the ionisation energy with Z that is nearly half the proportion of kinetic energy added to the electron's rest mass, which would be consistent with the kinetic energy alone acting on the electron's inertia. Though, I observe less than 50% of the relative kinetic energy acting on the ionisation energy, something like 46%. The fit is less perfect at hydrogen and helium (fewer digits in the data as well), so I observe the variation referred to lithium, where the very clean fit suggests that the 46% are meaningful despite they stretch the data's accuracy. So: what woud be the next correction after the kinetic energy added to the rest mass? Part of the electrostatic energy subtracting from the electron's mass? I feel normal to attribute far less than half of it to the electron, but haven't yet checked how much. Other causes? Thank you!

Hi Romix, welcome here! 5 or 10 because the crystal wants it that way. I can't imagine a way to predict that: it must depend on crystal packing, nothing related with valences or simple notions. I've seen variable amounts of crystallisation water, depending on the compound history; I can't tell for sure if, at small scale, there a few possible numbers which average out over many crystals to make a continuously variable composition, or if the number can vary in unit amount locally. For gem crystals, books indicate a uniform and integer number a water molecules per cell; this favours the first hypothesis.

I expect the hysteresis to create a worse drag than eddy currents do, as a gut feeling. Roll bearing steel is ferromagnetic and has an important remanence, making hysteresis losses strong. A simple test if still possible: hysteresis loss would create a torque at small speed also, while eddy currents create a torque proportional to the speed (it would stop increasing under conditions improbable here). Though, eddy currents are nasty in ferromagnetic materials, because the induction can be high and the permeability increases the AC resistance.

Hello everybody! You know that up to now, I considered the idea of particle useful to acount for properties like the charge, the spin... which don't split when a particle gets arbitrarily confined. As we have no means to confine a particle to a point, I felt a point particle unnecessary. This has evolved. I still believe that the interaction of a photon and an electron wouldn't need points and occurs over a significant volume: the volume of the orbitals if an atom or molecule absorbs a photon, thousand atoms volume or more if a semiconductor absorbs a photon... There, the photon's size adapts to the smaller (better localized) electron, the interaction is diffuse, and I feel that diffuse electrons and photons would be fine. An electron attracted by an atom's nucleus could be diffuse. We could even pretend that its charge spreads like q*|Ψ|2 provided its mass spreads like m*|Ψ|2 as well; this would keep Schrödinger's equation and its solutions. Better: the electron's electrostatic self-energy would be finite, saving the need for dressed particles, and with this energy varying like the kinetic energy does, I believe it could be tinkered into the electron's mass - so my ramblings... And anyway, when radiating light (wavelength >> orbital size) or not, an electron around a nucleus behaves like a charge spread as q*|Ψ|2. Two electrons (or electron pairs) interacting could still be diffuse. At an atomic force microscope, they interact as complete orbitals. What does need a point electron is the effect a spread charge would have on the orbital's shape. If the electron could repel itself, it would kind of screen itself from the nucleus, so that the outer part of the orbital would extend far more than it does. This would most probably give a spectrum that differs from the observed and accurately predicted ones, and for sure, van der Waals' forces completely different from the observations. So to avoid the auto-screening, it needs all the electron's charge at one point (or at least, much more concentrated than the size of the orbital) to write Schrödinger's equation. As a consequence, the mass must be concentrated as well. ---------- Or better, we need the charge and mass to be concentrated when considering the action of the electron on itself. For the interaction with other particles, we need an electron capable of concentrating as much as the other particle requires (... until I change my mind), but it acts over its full extension when the other particle spreads more. This holds for the electron because of its charge. Would you know more reasons that would apply to photons, which can usually superimpose without interaction? To neutral fermions? To all bosons? Your thoughts please?

Not a Wankel. This one has cylindrical pistons that move like usually. The cylinder arrangement is like on an axial hydraulic pump or engine. The "crankshaft" differs a bit from the usual pump construction. The rotating cylinders that move against the immobile head with air inlet and outlet is one possible design for an axial hydraulic pump. One such engine, but with the "crankshaft" resembling more today's hydraulic pumps, was built in the early days of the Diesel engine. I suppose, with no excellent reason, that the standard crankshaft is more efficient. Airtightness, yes... That's solved at hydraulic pumps, but air leaks worse than oil. A race car serviced after 1000km is easier than a consumer car. As well, valves are presently steered for variable inlet and outlet, which the head here may not enable as finely.

From what I read (not absolutely direct information) the metal, say aluminium, is first attacked by an acid to corrugate it deeply. Then the Ptfe is produced or deposited in situ and holds mechanically at the metals' cavities. I suppose that some alloys are better, for their heterogenous composition: grains and joints, eutectic precipitates... lets the acid dissolve them at uneven speed. I also believe that the coating isn't pure Ptfe, which would be far too weak against scratches. It's much loaded with a ceramic powder. ---------- The friction coefficient is complicated, poorly understood, and relates little with the surface tension nor with simple chemical properties. Proof of that: already at moderate heat (60°C?), Ptfe's coefficient jumps to values like 0.3, absolutely banal for a polymer. One of the many drawbacks that make it unusable pure as a bearing.

Hi Jayant, welcome here! In order for a current to flow, you must close the circuit, which isn't still the case with the conductor along y. Depending on if you close the circuit above or under xy, the current will flow in one direction or the other. This is consistent with the flux variation. Prior to closing a circuit, no flux can exist, since no surface is limited by the circuit. ---------- Once the circuit is closed and the current can pass, in which direction... You can try to make predictions using the corkscrew, Ampere's man or whatever you like. In such a simple case, you chances of getting it right are not negligible. In a more usual case like a squirrel cage motor, the chances are 50.001 to 49.999 so everyone foresees an inverter and experiments.

When electronic wristwatches nearly suppressed mechanical ones, manufacturers in Switzerland converted to other precision machining activities, one being small electric motors - meanwhile bought by US companies, whose motors presently run exploration robots on Mars for instance. I used some on rockets and a satellite, and: - They're made by machines, definitely. - On a 3mm shaft, the doc gave a tolerance of +0 -1µm. - Downstream a planetary gear, hence with a short shaft, you couldn't feel any play at all.

That cannot exist, so you don't need to shield it. It exists in liquids and solids. At any frequency including sensible ones, every solid makes a very efficient impedance mismatch with air. Matching air is difficult, mismatching very easy. Which is why displacements are not what you assume. 3GHz is commonly used in non-destructive testing of solids, with a displacement at the transmitter that is nowhere near cm nor mm. As 3GHz sound doesn't propagate through the atmosphere, no transmitter exists, no receiver neither, and materials can't be tested. Just for reference, the mean collision frequency of air molecules at ground level is 6.9e9/s, so a 3GHz sound can't exist in the atmosphere.

Graphite's anisotropic conductivity, both electric and thermal, is daily observation and common knowledge, not theory. It doesn't really need to be checked once more - but can be.

Hi Nucleara, welcome here! I hoped someone would provide a better answer, but here are already my 2 cents... A graphite block consists of many crystals with random orientation, making it more or less isotropic. Only smaller objects like a fibre or a single crystal show a clear anisotropy. Then, graphite conducts electricity and heat far better along the planes, but does conduct across the sheets as well. Van der Waals forces appear anywhere. Being what's left when chemical bonds are fulfilled, they have far fewer constraints. They are able to make liquid of absolutely any gas for instance, including noble gasses that wouldn't bind chemically. So, among graphite sheets, sure - that's easier than between dinitrogen molecules or neon atoms for instance. What I've read up to now is indeed that pi electrons from graphite sheets make van der Waals forces among the sheets, and this shall explain that graphite sheets are strong (chemical bonds) but hold loosely together (van de Waals) and cleave easily. Though, this fact has been overstretched to too many explanations including the lubricating properties of graphite - pity, it doesn't lubricate in vacuum, wrong explanation. An other funny observation is that benzene crystallizes with molecule rims pointing towards neighbours' faces, and not as stacks, so that sheet stacking in graphite is not the arrangement with best van der Waals' forces - it's just the only available one with graphite. Transverse conductivity, from sheet to sheet: that's not very difficult anyway. Electrons can pass by tunnel effect. The distance between the sheets is bigger than would be comfortable, but still within reach. One atomic distance makes tunneling easy (this defines an atomic distance), two would only be difficult. So it doesn't formally need a chemical bond nor even a definite intermolecular bond: proximity would suffice.

Lucky you. If only I did.

Satellites are not up to the task. Their cumulated throughput is ridiculous as compared with cell phones needs, because they don't easily use the same frequency at varied locations. Communicating with them (36,000km if geosynchronous) is seriously more difficult than to the next pylon (2km). Developers try to improve these drawbacks but progress won't cancel out the huge handicap.

On boats like aboard planes, the indication for a coming collision is that you see the other vehicle in a constant direction.

Hi Alex87, welcome here! I'm surprised by stiff figures about a throughput and a range. The frequency alone doesn't decide that. Generally, frequencies over 40GHz have a limited range in the atmosphere because of absorption, but would propagate in space. Then, you have all the uses where a limited range is fine: an intrusion detector, a ranging radar (to park your car), a liquid height meter... There are so many! As components are already good at such frequencies, all traditional uses of GHz are possible excepted long-range transmissions. That's a lot! Do you mean a specific band around 70GHz? The whole spectrum to 300GHz and over is already sliced among possible uses, so each band has one intent: terrestrial communications, mobile communications, amateur, Ism, radioastronomy, and so on. There: http://www.itu.int/pub/R-REG-RR contents certainly available on the Web.

The whole story about Schrö's cat is nonsense. Why go on with that stupid story in 2014?

Esa's paper is not about interstellar laser beam because this does not works because we can't make the mirrors by any extrapolation of technology.

I'm not willing to reopen here a topic already discussed there... In a course, Pascal Picard explicitly rejects the Ricci tensor for the gravitation or moderate masses. Do what you want with that - I do nothing at all. I see that I asked the simplest case I could imagine (a reasonable double star) and got no answer, making me wary about more complicated arguments. That's why I looked for simpler reasons and found the atoms with few and many protons, and the kinetic energy of electrons there. This kinetic energy increases the atom's inertial mass. Observation already tells us that it doesn't let carbon atoms fall more quickly to Earth than lead atoms do - that's within our experimental accuracy, and consistent with the principle of relativity. But if you want to believe something else, that's your problem, not mine.

Obviously you've been fooled by the paper. That is why I dislike hype in science papers. One bit of a full adder takes ~70 transistors, an Alu some more. With 300 transistors they can have made a 4-bit Alu. A processor still demands a sequencer, instruction decoding, registers, memory addressing... which take very much more than the Alu. With 300 transistors they did not make a processor. They miss a factor of 15. Impossible. Sorry for that. The logic may be that this team achieved to integrate 300 working nanotube transistors, which is a very nice achievement. Instead of publishing "we integrated 300 nanotube transistors", they chose to make something like a 4-bit Alu with them, which is more concrete and makes a milestone. Possibly, they added a sequencer and so on in silicon around - or they didn't. All right. But then, someone (One author? The scientific journal? The general press?) described the 300 transistor chip (maybe a 4-bit Alu but nothing more) as a processor, and this is dishonest. This fools readers with the best scientific background who are not specialists of that area. This should not happen.