Wikipedia:Reference desk/Archives/Science/2012 June 18

= June 18 =

Rhenium
Why is rhenium so rare? Double sharp (talk) 04:11, 18 June 2012 (UTC)
 * My guess would be: because it has an odd atomic number, and is not frequently produced by nucleosynthesis during supernovae. Plasmic Physics (talk) 04:19, 18 June 2012 (UTC)
 * It's very dense (21.02g/cm3) so it probably sank to the core of the Earth when it was still molten as did iridium. Also like iridium, it is very siderophilic, so it "likes" to dissolve in molten iron e.g. the Earth's core. According to it's only rare on the Earth's surface, not in the universe as a whole, so that rules out Plasmic Physics's hypothesis. 203.27.72.5 (talk) 04:36, 18 June 2012 (UTC)
 * Rarity is relative. Plasmic Physics (talk) 04:48, 18 June 2012 (UTC)
 * Touché. It could indeed be rare compared to other elements in the universe, but especially rare on the Earth's crust due to the processes I described. I can think of lots of elements with odd atomic numbers though, and many of them are not rare at all. 203.27.72.5 (talk) 05:10, 18 June 2012 (UTC)
 * According to Synthesis_of_precious_metals, 185W and 187W are formed from the irradiation of naturally occuring tungsten with neutrons (readily available in stellar conditions) and these decay into 185Re (stable)and 187Re (half-life 1010 years). So it may be generated in high metallicity stars. 203.27.72.5 (talk) 06:10, 18 June 2012 (UTC)


 * It is not that rarer than the other similar metals, but its chemistry is not leading to a strong enrichment in any minerals and therefore tit is hard to extract. Only Molybdenum ores contain significant amounts of rhenium. --Stone (talk) 10:01, 18 June 2012 (UTC)


 * "It is not that rarer than the other similar metals"
 * Don't forget that this graph is a logarithmic scale, so it really is a lot rarer than most elements; just estimating from the graph I'd say it's the 4th least abundant non-radioactive element. It's almost 10 times less abundant in the universe than gold.
 * Uranium and the late REE are less abundant but they are produced in considerable larger quantities. The amount the solar system and the amount we can extract are two very different things.--Stone (talk) 06:22, 19 June 2012 (UTC)
 * Also, to the IP above, note the trend in the graph with the odd-numbered elements: due to the Pauli exclusion principle, they are less abundant on average (save Hydrogen of course), and Rhenium is one of the heaviest stable elements, well beyond Iron-56, which is the last element which can be sustainably produced in stellar nucleosynthesis. I had never heard of the Goldschmidt classification though, clearly this plays a role in its rarity in Earth's crust. Thanks for the interesting read! - Running On Brains (talk) 12:37, 18 June 2012 (UTC)

If the subject of the origins and abundance of the elements interests you, I highly recommend P. A. Cox's The Elements: Their Origin, Abundance, and Distribution http://www.amazon.com/The-Elements-Origin-Abundance-Distribution/dp/019855298X which is perhaps the most fascinating and well-written science books I have ever read. I cannot recommend it strongly enough. μηδείς (talk) 17:04, 18 June 2012 (UTC)

Bird song identification
Just wondering if anyone out there can either identify a bird from a transliteration of its song, or point me at a site that can? Quite high pitched "didit, didit, didit" - UK Midlands. --TammyMoet (talk) 09:55, 18 June 2012 (UTC)


 * It's possibly a member of the Titmouse family - the name "tit" imitates the sound they make, as well as providing humerous material for many Carry On films. You can hear a Great Tit in this YouTube recording. Alansplodge (talk) 12:55, 18 June 2012 (UTC)


 * In the end I played my Bird Songs CD and they identified it as a wren. Personally I thought it may be a song thrush given the repetitive nature of the call, but I'll go with a wren. I hadn't heard it before, but I felt sure I'd seen it in a bird identification book. --TammyMoet (talk) 14:33, 18 June 2012 (UTC)

Xenon diiodide and krypton dibromide
Shouldn't xenon diiodide and krypton dibromide both exist, as they are isoelectronic with, respectively, the triiodide and tribromide ions? Whoop whoop pull up Bitching Betty 13:03, 18 June 2012 (UTC)
 * As you were told the last time you asked "why does/doesn't X given Y that is isoelectronic with it", you need to consider things like charge, electronegativity, etc. and also do a literature search before claiming something doesn't exist. That literature search would often reveal why not, as part of studies of analogous compounds. People really do study and publish "let's extend a known series into not-yet-known territory" articles on a regular basis--both of your proposed compounds have been studied (one even experimentally). DMacks (talk) 16:20, 18 June 2012 (UTC)


 * I really doubt these exist, given the low electronegativities of iodine and bromine. Xenon has an electronegativity of 2.6, so iodine and bromine (2.4 and 2.8 respectively, I believe) don't have the muscle to get xenon in a bond setup.--Jasper Deng (talk) 16:29, 18 June 2012 (UTC)
 * One is a known excimer complex; it indeed does have a very short lifetime and readily decomposes to the elemental forms. DMacks (talk) 16:36, 18 June 2012 (UTC)
 * Which one? Whoop whoop pull up Bitching Betty 18:53, 18 June 2012 (UTC)
 * Prospective ligands around xenon has to be sufficiently electronegative to tease out an atomic obital to leach onto. Xenon has a pretty stable ground state arrangement of atomic orbitals, it's not going to sit iddle, while an unwelcome guest is trying to take advantage of it - it's going to fight it, and look for the first opportunity to get rid of it. Bromine and iodine are too weak to contend with a heavy weight like xenon. Plasmic Physics (talk) 23:40, 18 June 2012 (UTC)
 * If my explaination seems a bit abstract, it's because I anthropomorphise my molecules and atoms, I give them personalities and character. Their behaviour is easier to visualise that way. Plasmic Physics (talk) 00:18, 19 June 2012 (UTC)
 * Plasmic, I'm concerned about your explanation. A quick glance at a descriptive chemistry textbook (in my case, Greenwood and Earnshaw's Chemistry of the Elements, 1st ed.) shows that xenon displays extensive chemistry, with oxidation numbers of +2, +4, +6 and +8 and a full range of coordination numbers from 0 to 8, and that xenon dibromide and xenon dichloride had also been detected by 1986. The chemistry of krypton, however, was much less developed at the time of writing, with only KrF2 and a handful of complexes being known. From what I can recall of lectures from a couple of years ago, krypton chemistry still lags behind, argon is restricted to a couple of highly unstable compounds and neon is still effectively inert: using xenon as your example of non-reactivity might not be the best choice. Brammers (talk/c) 09:22, 19 June 2012 (UTC)
 * I didn't say that it was non-reactive, I said that it is one tough customer, that doesn't like sharing, not that it can't. Plasmic Physics (talk) 10:13, 19 June 2012 (UTC)
 * Interesting that Kr and Xe are less electronegative than several of the halogens (and even O and N!)--things with which halides form moderately stable compounds. DMacks (talk) 10:32, 19 June 2012 (UTC)
 * Electronegativity is only realy applicable when discribing the relationship between atoms in a bond, first you have to form the bond. Plasmic Physics (talk) 10:57, 19 June 2012 (UTC)

Clues to calculate sun's future
Average people would thik when sun dies nobody knows, but scientific forecast, figuring out what will happen to our solar system in billions of years future is quite easy, not like what non-educated peoples might think. I wonder how do scientist come up with these variables, maximum size of the giants, net of sun's mass loss, how much longer they have on main sequence. Do they send a spacecraft to another stars, or astronomers basically use special powered telescopes? I thought when they send a spacecraft to another stars, then they are able to get these informations quite easily. What makes the informations fuzzy? Spacecraft? The technologies on spacecrafts are usually pretty strong on collecting informations? or to look at foreign stars/solar system they use a special powered telescope, which the variables like maximum giants radius, net star's mass loss might be sketchy causing scientist to come up with countless of different stabs. Does the data memory depend on how far away they are from our solar system? Further away they are from our solar system, do the virtual data memory get weaker? --69.226.45.94 (talk) 21:17, 18 June 2012 (UTC)


 * We are a long way from being able to send spaceships to other stars. They are so far away that any crew would be dead, and the ship would be out of energy, long before it arrived at even the closest star (beyond the Sun). StuRat (talk) 21:20, 18 June 2012 (UTC)

Almost all of our knowledge of other stars comes from stellar spectography. Even helium was discovered by this method. μηδείς (talk) 21:51, 18 June 2012 (UTC)
 * You mean, helium outside our solar system was discovered by this method, right? 203.27.72.5 (talk) 22:09, 18 June 2012 (UTC)
 * Nope, you meant what you said. That's pretty amazing. 203.27.72.5 (talk) 22:11, 18 June 2012 (UTC)
 * The furtherest man made object from Earth is Voyager I which is not heading towards any particular star, but will pass within 1.6 lightyears of AC+79 3888 in about the year 42000. Voyager II will pass by Sirius in about the year 300000. Pioneer 10 would take until the year 2000000 to reach Aldeberan but it probably won't get there anyway. Pioneer 11 is headed vaguely in the direction of Scutum. Is hasn't arrived there yet, but when it does, I'll let you know. 203.27.72.5 (talk) 22:07, 18 June 2012 (UTC)


 * As the others point out, our predictions about the solar system are largely based on observations made from earth. The space probes and things like the Hubble telescope also contribute information. This tells us about the way things (probably) are in the solar system. To predict the future, scientists take this information and feed it into mathematical models which are based on physical laws. But all models are "wrong", in the sense that they make simplifying assumptions, and all of our observations of the universe have errors and "noise" in them. Me, I'm still amazed this stuff is accurate and precise enough to put a man on the moon :) SemanticMantis (talk) 22:27, 18 June 2012 (UTC)
 * I've read before that all of the Apollo missions were possible without any relativistic corrections to the Newtonian mechanics. In that regard I don't think putting a man on the moon needs such great accuracy in the underlying physical models. 203.27.72.5 (talk) 22:34, 18 June 2012 (UTC)


 * We had yellow stars, maybe tons of it which have came off the main sequence, and collapse into white dwarf. I thought scientist use the same model what happened to similar sun-like stars, and use that to see what will happen to our sun. i wonder if you ever thought about that.--69.226.45.94 (talk) 23:18, 18 June 2012 (UTC)

Stellar evolution is far too slow to observe, save for the dramatic nova and supernova events. We can draw similarities between the states of all different stars in the sky, then sort them into categories, and then apply theory from the known physical properties of matter and energy to generate hypotheses on which categories of star result from what conditions. The main variabilites in stellar condions are mass, metallicity, temperature and age. Theories that prove useful in explaining the different observed states of stars allow us to model stellar evolution and then, by constructing a model using the current age, metallicity and mass of our star, we can make some assumptions about how it's evolution may progress.

We don't watch main sequence stars become white dwarfs. We see white dwarfs and we see main sequence stars and we hypothesise that white dwarfs develop from main sequence stars through mechanisms that we propose. We then collect evidence to support these hypotheses in the form of spectral data (chemical composition), brighness measurements (temperature), etc. 203.27.72.5 (talk) 23:34, 18 June 2012 (UTC)

HV Electrical transformers - do they really need a hand to help them go bang?
Hi. I'm watching a video on Youtube in which a High Voltage transformer block on a power pole overheats, and promptly explodes in a complete mass of hell. My question to the desk is this: Why do the companies who make these things see fit to fill them with something explosive to keep them cool and happy?

If they overheat or flat out short, they (can, and often do,) explode - and mineral oil & the coolants used are highly explosive, which only seems to add "fuel to the fire", so to speak. Is this a design flaw, or are there really no alternatives to the inclusion of such highly combustible materials in the making of these products?

BarkingFish 23:03, 18 June 2012 (UTC)
 * We have some information in Transformer and Transformer oil. There's just not much that is economically viable and has the right thermal properties to be useful (shame PCBs are outlawed). It's also hard to appreciate how much energy there is in an arc flash. There's just not much at all that can withstand it. I should also point out that the materials are not explosive by nature and are not really highly flammable either. But they can be made to burn given a large enough added energy, or to boil just a small enough amount to get a BLEVE. DMacks (talk) 23:40, 18 June 2012 (UTC)


 * Don't forget the amount of energy flowing from the power line that is capable to power hundreds of households. Which when the electrical characteristics of the transformer fails can end up in a small enclosed container. And easily create something like combined steam explosion and fuel vapor bomb. Electron9 (talk) 04:01, 19 June 2012 (UTC)


 * See Transformer oil. It insulates and cools transformer windings. There are also air cooled transformers ("dry types"), but they must be securely protected from moisture, may need cooling fans, and do not seem to offer as high a power rating for a given volume and weight. They would fail if rain, snow etc got into the windings, and are not good candidates for poletop use. Transformer says that some dry types are now made in sealed tanks, insulated and cooled by sulfer hexafloride, which sounds like a good idea if the cost is not too high for typical distribution sizes and size and weight are comparable. It would probably be harder to keep a distribution transformer gas-tight for decades than to keep it liquid-tight. The transformer oil is sometimes mineral oil, similar to motor oil. It does not seem to be all that explosive at normal temperatures, but if it is heated above its flash point and sprayed out in little droplets, then you can get an impressive fireball. I have seen a relatively small oil-filled padmount transformer blow open and burn the face off the bricks of a brick wall. PCB was less flammable than mineral oil, but it could produce dioxin in an arcing fire and has generally been replaced. Silicone oil is a replacement which is not very flammable, but I've heard (no ref handy) it can be explosive if it gets contaminated with water. A transformer which steps down 12000 volts to 240/120 volts might well have 10000 amps of fault current flowing into a short or arc inside the transformer, should one occur. This could happen if there is an insulation failure in the windings, if the oil level drops low, if water leaks in when a gasket fails, if overheating due to overloads causes paper insulation to break down and release moisture, if the oil becomes carbonized from tap changer operation or overheating or arcing at loose connections, or even because lightning struck near the transformer. (I have seen 60 year old transformers with oil as dark as coffee), still in use, but utilities do regular tests of the fluid in larger transformers. It could also be caused by a switching error or resonant conditions in a cable-fed system ("ferroresonance"). It is then up to the primary high voltage fuse to interrupt the current before the metal container fails and flaming oil shoots out. High voltage power fuses are cleverly designed, but complicated gadgets. Some have an overcurrent element which melts, causing a strain element to release a spring which causes an arc inside a tube containing boric acid, the ionized particles of which interrupt the arc. This might ideally happen in less than the first half cycle of fault current. Other fuse types can't interrupt until the current hits a zero at the zero point of the sine wave.  The vast majority of the time, the primary fuse quickly does its job, and only customers served by the one faulted transformer lose their power, with others only seeing a momentary dimming of lights until the fuse operates. Then a utility lineman shows up in an hour or whatever and tries replacing the fuse and relivening (if there is no obvious damage) in case it was just a weak fuse or a lightning glitch. If the transformer won't reliven, then the lineman calls for a replacement transformer, and in a city, might be able to temporarily feed the outaged customers from the next transfomers, if the secondaries extend down the alley. When the transformer fuse fails to interrupt,  the thousands of amps of fault current continues to transfer many megawatts of power into the faulted transformer can until the transformer catches fire or an upstream backup fuse operates on a delayed basis, likely killing the power to other transformers as well. Edison (talk) 19:46, 19 June 2012 (UTC)


 * Mitsubishi makes sulfer-hexaflorude insulated transformers in distribution size and up: . No info on price, mean time between failures, and size/weight for a given KVA capability. Other manufacturers emphasize substation and high voltage gas insulated transformers. Edison (talk) 20:04, 19 June 2012 (UTC)


 * A common measure of the estimated energy released by an arc-fault is in equivalent amount of TNT, and common values appear to be are on the order of a few pounds per second of arc in the 15 kV primary transformer/switchgear level, down to "only" a few ounces-worth (about a hand grenade or so) for end-user secondaries. DMacks (talk) 20:19, 19 June 2012 (UTC)


 * If Edisons figures on volt and currents are correct the transformer is attached to a line that transfers the energy equalient of 12 000 V * 10 000 A / (4.2*10^3) = 28 571 gram = 28 kg of TNT per second! or the power that a Boeing 747 uses. Electron9 (talk) 01:24, 20 June 2012 (UTC)

Regarding fuses etc.. how come a CME is able to bypass fuse and over voltage protections of the electrical grid? Electron9 (talk) 01:24, 20 June 2012 (UTC)
 * Power system engineers calculate the fault current at some location, such as a transformer, by starting with the voltage at the source, then factoring in the source impedance as well as the impedance in the transformers and lines between the source and the fault, and any arc impedance at the fault. As a result, there will be some voltage drop. If a conductor is solidly grounded at the fault, then the voltage to ground would be very low. The phase to ground voltage from a 12 kv 3 phase circuit would be 7200 max, so it might be much smaller in a fault due the the voltage drop of the system conductors. Thousands of ampere at thousands of volts is quite possible on the distribution system. In some situations, you want pretty high fault current so that the fuses or breakers will "coordinate" in such a way that only the closest upstream protective device operates, to minimize the outaged area and decrease the extent of equipment burned up. You are still talking an arc at a temperature of thousands of degrees, unless it is a metal to metal fault. Lots of heat is produced, and lots of sprayed flaming oil and expanding gases. One hopes the fault is extinguished in a fraction of a second. Edison (talk) 19:58, 21 June 2012 (UTC)