Wikipedia:Reference desk/Archives/Science/2017 March 6

Science desk
< March 5	<< Feb | March | Apr >>	March 7 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

March 6

Tin pest

If a piece of tin gets tin pest, is there a way to undo it?32ieww (talk) 00:16, 6 March 2017 (UTC)

Assuming that it has not been seriously damaged (e.g. crumbled to dust), annealing should do the trick. (You might not even need a furnace to do it, just sticking it in hot water for a day or two might do the job!) 2601:646:8E01:7E0B:F88D:DE34:7772:8E5B (talk) 01:32, 6 March 2017 (UTC)

You mostly need to re-melt it. It's possible that it was cooled, changed structure and wasn't handled enough to make it crumble, but usually (every piece I've seen) has at least craccked. Maybe it was grey tin before this, but I've only noticed because I could start to see physical damage. Andy Dingley (talk) 10:01, 6 March 2017 (UTC)

Tin pest, for anyone who doesn't know it. Fascinating stuff! SemanticMantis (talk) 15:20, 6 March 2017 (UTC)

What species of spider is this?

It has a red abdomen and a thick black body, and its front two arms seem thicker than the rest. Was found in San Diego, CA -- penubag  (talk) 00:35, 6 March 2017 (UTC)

Perhaps one of these. DuncanHill (talk) 01:02, 6 March 2017 (UTC)

Or perhaps Castianeira occidens, the Ant Mimic spider. DuncanHill (talk) 01:04, 6 March 2017 (UTC)

Hm, I think it's Phidippus johnsoni. Thanks!

Tin cans

Do tin cans use pure tin or tin alloy(I want to make gray tin)?32ieww (talk) 02:19, 6 March 2017 (UTC)

Tin cans are made of steel plated with tin, or of aluminium, or of tin-free steel. DuncanHill (talk) 02:24, 6 March 2017 (UTC)

• Try buying some modern lead-free solder instead. A plumber's merchant will sell this more cheaply than electronics suppliers, often (depending on country) on a green plastic reel, specifically for drinking water systems. Other countries simply mandate lead-free for everything now. Andy Dingley (talk) 09:59, 6 March 2017 (UTC)

That's because lead-free solder is mostly tin, right? Our article says it can a few other things mixed in. I suppose you'd have to separate them out to make gray tin? SemanticMantis (talk) 15:25, 6 March 2017 (UTC)

I think you would have to separate them out—my understanding is that they're there specifically to prevent formation of gray tin (tin pest). –PointyOintment (talk) 05:34, 7 March 2017 (UTC)

Then where can I get pure tin? I heard of an application of gray tin, that when you want to dissolve tin into something, you can take gray tin, crush it up into powder, and use it instead of a normal lump of tin. The trick here is that the surface area to volume ratio is MUCH higher, and so the gray tin dissolves into the solvent much faster. 32ieww (talk) 22:58, 7 March 2017 (UTC)

The additive for stopping grey tin forming is lead.

You can buy pure tin. You can most easily buy potable water plumbing solder which will (typically) be 5% antimony. This still transforms to grey tin, but it does it much more slowly than Napoleon's trouser buttons (it might take a year). It's enough to cause problems in freezers though.

If you just want high surface area tin then stick it through a jeweller's rolling mill and turn it to foil. Andy Dingley (talk) 00:27, 8 March 2017 (UTC)

• How much do you need? you can just order it from a chemical supply house, e.g., here: http://www.sigmaaldrich.com/catalog/product/aldrich/265640?lang=en&region=US -Arch dude (talk) 01:05, 10 March 2017 (UTC)

Lead

Can lead be forced, maybe under extreme cold, to take on a diamond crystal structure like tin or germanium or silicon or carbon?32ieww (talk) 02:19, 6 March 2017 (UTC)

Maybe. But no one's ever done it. Double sharp (talk) 07:36, 6 March 2017 (UTC)

Not a diamond crystal as such, but lead perovskite methylammonium halides have a nice regular crystalline structure

and are being looked at as cheap solar cell materials. loupgarous (talk) 23:43, 9 March 2017 (UTC)

Square graphene

Can carbon make sheets of graphene with squares instead of hexagons?32ieww (talk) 02:23, 6 March 2017 (UTC)

Pretty sure no. Square planar molecular geometry requires d orbitals as far as I can tell, and the lowest energy d orbital is higher than any electron of ground-state carbon. That is, if you were to somehow make such a thing exist, it would not be stable. Someguy1221 (talk) 03:01, 6 March 2017 (UTC)

The carbon atoms in a square-tiling arrangement would be bonded to four other atoms and so they'd much rather have a tetrahedral molecular geometry around them than a square one. And they would also not be able to form the pi bond that is characteristic of graphene, because each carbon atom would already have four sigma bonds. So no. Double sharp (talk) 07:33, 6 March 2017 (UTC)

• No. It wouldn't be graphene if it was, I can't see how to form carbon into such a structure and although such a structure isn't implausible for other materials, it wouldn't have the high degree of adjacent bonding which is what makes graphene so interesting. Andy Dingley (talk) 10:02, 6 March 2017 (UTC)

  No, it would not be graphenes, but it would be fenestranes. Which do exist, but are highly unstable and tricky to make. --Jayron32 15:29, 6 March 2017 (UTC)

It would definitely be interesting if you could make the extended structure. I mean, if you had granules of that and liquid oxygen (or maybe red oxygen, as long as I'm daydreaming) I suppose it would make a fine rocket. Does the instability increase the larger the fenestrane, or can it conceivably be managed? Wnt (talk) 21:32, 7 March 2017 (UTC)

I think you're getting into cubane and octanitrocubane territory there. Andy Dingley (talk) 00:13, 8 March 2017 (UTC)

What is the average number of dates of the Gregorian year that a Full Moon falls on in a lifetime?

Without changing location. If astronomical cycles like the Metonic cycle and the tropical year were perfect and an integer number of days long then only about 19 dates per synodic month could get Full Moons under any single consistent definition of date (i.e. location X's noon to noon or midnight to midnight) but the Metonic cycle, Gregorian calendar, perigee etc. drifts over time so more could be fit in. Sagittarian Milky Way (talk) 16:09, 6 March 2017 (UTC)

Our full moon article says "The time interval between similar lunar phases—the synodic month—averages about 29.53 days". The Gregorian year says "For the Gregorian calendar the average length of the calendar year (the mean year) across the complete leap cycle of 400 years is 365.2425 days". Dividing the days in a year by days per full moon gives us 12.37 full moons per year. So, multiply that by your fave figure for average lifespan, in years, to get your answer. (But you've modified the Q, and I now get the impression you're asking a different Q, so you can ignore this if it doesn't apply.) StuRat (talk) 16:26, 6 March 2017 (UTC)

Right, I was wondering how many of the 366 possible dates get hit before you expire. Sagittarian Milky Way (talk) 16:34, 6 March 2017 (UTC)

If you know your own birth date and death date, then you can do your own calculation using the data here. Dbfirs 09:16, 7 March 2017 (UTC)

Taking that data at looking at every full moon from the 80 years starting Jan 1, 2000, there are only 3 days that don't get a full moon. It's actually much better coverage than you'd expect from random chance. Someguy1221 (talk) 09:28, 7 March 2017 (UTC)

Thank you for that fast analysis (I started doing it, then decided to leave it to Sagittarian). That's about what I would have expected given the slow drift of parameters. I don't think there will be any dates that are missed in all lifetimes in say, 200 years, but I'd be interested to be proved wrong. Anyone born in March 1972 might miss a February 29th full moon. Dbfirs 09:43, 7 March 2017 (UTC)

genetic article reconciliation

In reading the article on Y-DNA Haplogroup R1b, I noted a paragraph in which it describes the high percentage of R1b in the basque population in northern Spain. The paragraph also mentions that this concentration of R1b does not correlate with a similar mt-DNA lineage from the pastoral steppe of western Asia. Specifically:

"One of the highest level of R1b is found among the Basques, who speak a non- Indo-European language isolate. One hypothesis about the case of the Basques is that a male-dominated Indo-European-speaking people invaded and conquered the Basque region, and then, having brought no or few women with them, then married local women, possibly from a matrilineal society The women then raised the children that resulted to speak their language and cultural practices, rather than those of their fathers. This hypothesis is supported by the fact that while other high-R1b regions in Western Europe (such as the British Isles and southern Germany) also show disproportionately high incidences of MtDNA haplogroups that correspond to a Pontic Steppes origin (specifically MtDNA Haplogroups I, U2, U3, U4, and W), while the Basque region does not. In fact, the Basque region displays virtually no MtDNA for which Pontic Steppes origin could be claimed.[20]"

This information does not seem to be congruent or is missing from the main basque people's page which states:

"In 2015, a new scientific study of Basque DNA was published which seems to indicate that Basques are descendants of Neolithic farmers who mixed with local hunters before becoming genetically isolated from the rest of Europe for millennia.[53] Mattias Jakobsson from Uppsala University in Sweden analysed genetic material from eight Stone Age human skeletons found in El Portalón Cavern in Atapuerca, northern Spain. These individuals lived between 3,500 and 5,500 years ago, after the transition to farming in southwest Europe. The results show that these early Iberian farmers are the closest ancestors to present-day Basques.[54]

The official findings were published in Proceedings of the National Academy of Sciences of the United States America.[10] "Our results show that the Basques trace their ancestry to early farming groups from Iberia, which contradicts previous views of them being a remnant population that trace their ancestry to Mesolithic hunter-gatherer groups," says Prof. Jakobsson.[55]"

While this piece is mostly in reference to the claim that Basques are descendant from paleolithic peoples, there is no mention of the possibly older mtDNA lineage that predates the dispersion of R1b into western Europe, as per the haplogroup article. It is not clear which information is accurate, and the source listed on the haplogroup article only links to a eupedia page with generic genetic information. Can this claim be verified or not, and can both articles be adjusted to include more complete information? 97.78.160.165 (talk) 19:18, 6 March 2017 (UTC)

This is a good example why primary sources should not be used. Ruslik_Zero 20:53, 6 March 2017 (UTC)

Absolute nonsense. This is a good example why more than one primary source should be used. If you want to wait till the dust is settles on every little detail, you're not going to have any articles. Fgf10 (talk) 07:55, 7 March 2017 (UTC)

That does not compute. The guidance against using primary sources is not to make sure we only present information "when the dust settles", but to avoid overemphasizing individual, possibly insignificant articles or viewpoints. Someguy1221 (talk) 08:02, 7 March 2017 (UTC)

This is off-topic for the refdesk, but that is a nonsense reason as well. If someone posts an incorrect of irrelevant primary source, it'll get reverted. It's the vendetta against primary sources that makes many science articles on wiki so incredibly rubbish. I'm sure it's fine for humanities articles, as nothing is going to change, but it definitely doesn't work in science ones. Fgf10 (talk) 09:08, 7 March 2017 (UTC)

How do you, as a writer, decide which primary sources to use when adding text to an article? --Jayron32 11:52, 7 March 2017 (UTC)

You use ones in reliable journals. If other editors disagree, it'll get reverted. Surely I don't have to explain how editing here works?

In general, contradictions are good on Wikipedia. Summarizing any active field of research involves noting facts and interpretations that appear to contradict one another. Sometimes there is error, but sometimes the error is informative because the underlying technique is flawed. Sometimes it is just a matter of reconciliation, as demonstrated well below. See Hegel and dialectic - creativity works by smashing contradictory ideas at each other much like ions in a particle accelerator, and seeing what comes out. Wnt (talk) 13:27, 8 March 2017 (UTC)

Having had the time to read both Wikipedia articles and the scientific studies they are based on, the data itself has no conflict, but the interpretations do. Basically, the history of the Basque people has always been something of a mystery, partly relating to their unusual language, which lies outside the Indo-European language family. That family of languages was introduced to Europe by a wave of farmers arriving from the Caucuses or thereabouts. They brought agriculture, language, culture and their DNA, of course. The modern Basque people show almost no mtDNA associated with that migratory wave, despite showing YDNA so associated, leading to the conclusion in the article. The new PNAS article sequenced the genomes of eight pre-historic skeletons found in the Basque region, and these show mtDNA associated with both the original hunter-gatherer populations present in Iberia, as well as the agricultural group that migrated in later. Now, the simplest hypothesis would be that this group simply had nothing to do with the Basques, aside from being buried in the region, but their genomes show too much similarity to modern Basques. The authors of this paper instead conclude that the Basques have broadly the same ancestry as everyone else, but for whatever reason all mtDNA that came with the later migration got lost, perhaps due to some bottleneck event. More exotic hypotheses can also be made, but then we get into original research by yours truly. So anyway, tldr, the one Wikipedia article is based on a paper studying a very large number of living Basques, and the other is studying a very small number of long dead Basques. But this new paper is also not the first to sequence mtDNA from ancient Basque skeletons. There are so many papers, in fact, I'm sure there's a good review out there already. Someguy1221 (talk) 10:32, 7 March 2017 (UTC)

Precise fuel gauge?

Having returned a rental car with a pre-purchase tank policy (in a touristic location in Southern Europe) I feel ripped off. I've paid in advance for a full fuel tank and the remaining fuel at return was estimated by the company, which decided I had almost no fuel in there (I strongly disagree about this, but can't prove them wrong). Why don't cars have a precise fuel gauge that would show us the exact amount in liters in there?--Llaanngg (talk) 19:38, 6 March 2017 (UTC)

What you said was “a pre-purchase full tank” and you got a full tank. Your getting confused with 'full to full'. In which case you would only have had to pay for the fuel you used. Caveat emptor.--Aspro (talk) 20:50, 6 March 2017 (UTC)

Most rental agencies have the expectation that you "top it off" and return the car full, they often charge exhorbitant rates for you not doing so, it is supposed to be punitive if you don't. Here is some advice on how to avoid being taken advantage of; and most of it you have past the point-of-no-return. Things like taking pictures of the fuel gauge, or keeping a receipt of the location, time, and date, and also noting the mileage of the last fill up, etc. Also, staying with the car and getting the agent to quote you the fuel price in front of you, and disputing the charge immediately. All of that is covered in that article. If you've been taken advantage of, but have no evidence of such, then there's not much you can do. --Jayron32 21:04, 6 March 2017 (UTC)

It was not a full-to-full. It was a full-to-empty, where I paid for the gasoline in the tank. They offer a refund of not used gasoline, but they estimate how much fuel is left. That is, they were not expecting to get a full fuel tank back. But the question is not about policies, but about the possibility of measuring exactly the amount of fuel (and not only estimate it with some shaky method). Why can't the car tell me I have 4.5 l left? Cars nowadays have a computer (or several actually). Couldn't it just calculate how much gasoline went in, and how much was spent by the motor? A modern fuel injection motor should know perfectly how much it has spent. Why not make the information available?Llaanngg (talk) 22:40, 6 March 2017 (UTC)

Many car computers do exactly that, but they usually display the result as miles left rather than litres left in the tank. Mine slightly under-estimates the fuel used, but only by a small percentage, and it's probably in starting that it gets the estimate wrong. You should have syphoned out the remaining fuel, or driven some extra miles if you were worried about the remaining fuel. Why didn't the rental company do an accurate estimate? Were they a reputable company? Dbfirs 22:53, 6 March 2017 (UTC)

No, it was not a reputable company. BTW, if I had done what you suggest, my loss would have been even higher. That is, time spent, absolutely no refund, not just an unfair refund. I see what's your point. This seems to be a variant of the Ultimatum game.

Are you saying the car had no fuel gauge or what? AFAIK nearly every single slightly modern (by which I mean maybe 30 years or so) petrol or I assume diesel car has some sort of fuel gauge even if it's just a simply analog 4 points one like that shown in our article or [1]. These gauges aren't always that accurate (as the linked source says, partially intentionally) and obviously lack precision but if the fuel gauge was basically at the E level or if the indicator light was on, I don't think it's that surprising they decided it was empty. What level was the gauge when you left the car? Incidentally, you're talking about cost but it's not clear to me what option you expect the company to use other than relying on the fuel gauge. Sure they could choose a car which has the best fuel gauge but there are generally more important considerations. They could perhaps come up with a more careful estimate from the fuel gauge, but we don't really know how bad their estimate was since you gave no indication what was actually showing on the fuel gauge. If you didn't actually check what was showing on the indicator before you left the car, it seems difficult to dispute their estimate. I guess if you later disagreed with their estimate you could have asked to see the gauge and if they refused perhaps you had a complaint, but you didn't mention any of this. Nil Einne (talk) 23:31, 6 March 2017 (UTC)

The car had not precise fuel gauge, just one with 4 divisions, with 2 subdivisions each, that was rounded down for the refund. But again, this is not about the policies. Why doesn't it say 7.8 l or 12.5 and so on? Llaanngg (talk) 23:36, 6 March 2017 (UTC)

Yes that's the sort of fuel gauge that is common. Are you saying that the gauge was showing close to the first subdivision away from empty but it was still considered empty? If so I guess I can understand why this is annoying although I also wouldn't be surprised it the policy actually said somewhere this is what they'd do and if we're talking about a small car I'm guessing we're discussing under ~7L so maybe $20 at most. The article I linked to talks about how fuel gauges in cars normally work. It doesn't particularly discuss why they aren't digital but I'm not convinced what you're proposing is actually that useful. Something like that mentioned by Dbfirs, with distance remaining sure if it works properly, sure. Knowing you have 6L left instead of a quarter tank, not so much. Greater precision can be beneficial, and digital will make that easier although may or may not be easier to read/understand at a quick glance. Remember what matters is what works for a person driving the car wanting to know how much fuel they have left, whether they can get where they're going without needing to refuel or if they will need to refuel, roughly when. Most people will roughly learn from driving their car how this correlates with the level of the fuel gauge. They aren't going to be doing calculations based on km per litre. When they fill up, they'll often either do so completely or do a dollar amount and probably won't even look at how many litres that correlates to and will then go by the position of the gauge. Possibly people will also learn how to do this with a digital gauge with litres remaining but there remains the question of why. The manufacturer isn't going to be particularly concerned about disputes over fuel remaining with car rentals. Actually what they may worry about is the user complaining when the car only seems to suggest they gained 5L when they filed 7L etc. Nil Einne (talk) 00:07, 7 March 2017 (UTC) Edit: should clarify when I said the article, you need to look at both our article and the external source as these discuss different aspects. 01:23, 9 March 2017 (UTC)

(EC) I know your question wasn't directly about the dispute but ultimately the point is it doesn't actually matter if the car did have a gauge which said exactly how much fuel when the basis of the disagreement is unclear. (I.E. even if car said it had 4.5L left, this is irrelevant if you didn't actually check what it said, or are going to dispute it even if it said that because you're sure it's wrong.) Nil Einne (talk) 23:40, 6 March 2017 (UTC)

(ec here too) Simply b/c the fuel gauge is not meant to please penny chasers but to show one an estimate on how good you're on fuel to be safe not to run out of it before you hit the next gas station.--TMCk (talk) 23:44, 6 March 2017 (UTC)

It is not so much the cost of the fuel you used but the fact that in order for the next customer to get a full tank, an employe has to stop what he is doing, drive the car to the local filling/gas/petrol station. Fill up. Collect the receipt, submit it, process the paperwork, etc. Example: Go eat at a restaurant. The raw food cost price is only about 30% of the bill or less of which you willingly pay. The car rental company would go bust if they tried to absorb those service cost because they are competing with other car rental firms. Do the sums. If a cheap skate hire company only pays their employee 25K per year, their total cost of employment is more than twice that (when overheads etc., are taken into account), i.e., 50,000 per annum and with holidays that is over a 1,000 per week. 200 per day. Turn that into a hourly rate and it is about 25. How can they get the necessary 70% return on those extra hours needed to stay in business? Don't think you have any case at all.--Aspro (talk) 23:39, 6 March 2017 (UTC)

I don't see how this is relevant to the OP who has already specified that it was acceptable for the car to be empty because of the policy they took. In other words, AFAWK, at least based on what the OP has said they would have been charged or refunded the same (i.e. zero) if the car was really basically out of fuel. The dispute seems to be over why the company decided it was empty, which doesn't directly relate to the question anyway but as I said, I don't see how a better fuel gauge would help the OP if they either didn't check it or are going to disagree with it anyway so in some ways it does. Nil Einne (talk) 00:02, 7 March 2017 (UTC)

• If you were an engineer, how would you design a system that could precisely measure the amount of fuel in the tank? I cannot think of a simple system to do this that is inexpensive, safe, and will fit in a standard car. The only simple scheme would be to remove and measure all the gasoline, but this system would still be under control of the rental company and subject to dispute. -Arch dude (talk) 04:49, 7 March 2017 (UTC)

That's the way to start thinking about it. You'd have to compensate for the variations in density of the fuel, with temperature, the varied size of the tank when you fill it (some cars have plastic fuel tanks), the amount of fuel returned (in some vehicles) to the tank from the engine. You'd have to account for condensation. You don't know how much fuel went in in the first place. You don't actually know the volume of the tank, or where whatever gauge system you are using is placed inside the tank. It is quite a tricky one to solve. Greglocock (talk) 06:13, 7 March 2017 (UTC)

... but manufacturers seem to have solved it to 95% accuracy in cars with computerised fuel gauges. Dbfirs 08:32, 7 March 2017 (UTC)

A fuel gauge is often a measurement of the height of fuel remaining in the tank. The technology employed varies, and often if it considered more important to accurately warn if fuel is low than to carefully track the progression at other times. (How angry would one be if you ran out of fuel before the meter got to empty?) Here is a description of one type of technology used. [2] As one can imagine from the description, it isn't a high precision technology, but it is relatively cheap. For cheap sensors, they don't even necessarily correct for variations in the shape of the fuel tank, so the gauge may actually change more rapidly if the tank narrows towards the bottom. Of course, some cars come with more accurate sensors, but those are by extension more expensive. Since most buyers don't care about fuel gauge accuracy very much, most manufacturers don't bother with better sensors that would increase their cost. There is no technical reason that one couldn't report a precise volume accurate to within say 5%, but situations where precise gas volume matters are rare for most drivers and so their is little demand for systems that would do that. Dragons flight (talk) 10:28, 7 March 2017 (UTC)

That makes sense. In every car I've owned, the fuel gauge varies somewhat depending on whether you're driving uphill, downhill, or flat. ←Baseball Bugs ^{What's up, Doc?} carrots→ 12:47, 7 March 2017 (UTC)

The approach above doesn't really make that much sense to me. All you really need to know is the pressure at the bottom of the tank. Conceivably, you could rig a small tube to go up from the bottom of the tank and use a camera to monitor the fuel level in it at all times; software that knows the shape of the tank and cross-references with accelerometer readings (like in one of those video game controls - I bet the car has one in it somewhere already) to figure out how it is affected by momentary motion. In practice that sounds leaky, and I'd expect there to be some tiny, cheap, idiot-proof pressure sensor available (actually several redundant sensors in a single module, in case there were malfunctions) to put at the bottom that works like an altimeter. The computing power and software should be extremely cheap now also though I should note a preference for a tiny dedicated module rather than putting every damn thing on the Internet for Kim Jung Un to fool with. Looking up cheap pressure sensor I get to an Alibaba list for them, which has $10 or less items for truck oil pressure measurement. I don't really understand why the manufacturers would use the float mechanism. Wnt (talk) 13:13, 8 March 2017 (UTC)

If you're going uphill or downhill, the part of the tank that's the "bottom" will change. ←Baseball Bugs ^{What's up, Doc?} carrots→ 13:42, 8 March 2017 (UTC)

True... I suppose I was thinking of a video game controller that detects not just acceleration but changes in orientation also. Wnt (talk) 00:17, 9 March 2017 (UTC)

Our article I linked to above mentions a pressure sensor but only for large tanks and I think it's mostly referring to immobile ones (and it also suggests they use some sort of float/resistive one). Besides the simple potentiometer example, it also mentions the similar magnetoresistance cases which are used in light aircraft and "highly accurate". I'd note that the external source I linked to does mention that modern gauges may be microprocessor controlled which allows compensation for things like tank the shape of the tank. (Perhaps it also uses some sort of memory or averaging during normal driving.) While it doesn't explicitly say it, I'm pretty sure it's still referring to some sort of float type. I strongly suspect this is what's used by Dbfirs example. Meanwhile large aircraft are said to use multiple low voltage tubular capacitor probes. While there is a great degree of inertia, particularly with stuff that simply works most of the time and is a known quantity, I'm also very hesistant to assume a suggestion by someone who doesn't really understand the engineering issues at all, will work better. Nil Einne (talk) 01:23, 9 March 2017 (UTC)

Oh, I doubt it also. There's almost certainly a reason why my brilliant idea is stupid, I just don't happen to know what it is ;) Wnt (talk) 23:50, 9 March 2017 (UTC)