Wikipedia:Reference desk/Archives/Science/2012 October 1

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.

October 1

Prevent superheating in microwave

How exactly does a spoon prevent superheating of liquids in a microwave. Does it matter whether it is a wooden/plastic/metal spoon? bamse (talk) 07:34, 1 October 2012 (UTC)

By providing nucleation sites. A rougher material is better at that, so a well-polished metal spoon might not work. Wooden spoons are always rough, I believe. Plastic I'm not so sure about. StuRat (talk) 07:56, 1 October 2012 (UTC)

I know there was a Mythbusters episode that debunked the myth that a microwave oven would explode if a metal spoon was put int it, but nevertheless I thought putting any metal object in a microwave oven was not a good idea because it could damage the magnetron and shorten the lifespan of the oven. Am I out of date ? Gandalf61 (talk) 10:33, 1 October 2012 (UTC)

Maybe the metal spoon is meant to be completely submerged in the water although that isn't as much fun. Sean.hoyland - talk 11:08, 1 October 2012 (UTC)

My understanding of the advice is to put the spoon in after the microwave has finished, so that, if the liquid is going to boil over when disturbed, it does so before you pick it up and you don't get scalded. --Tango (talk) 17:00, 1 October 2012 (UTC)

Per StuRat, the goal of the guidance is to provide nucleation sites. The "spoon" part of "wooden spoon" is irrelevant for this purpose; the real question is one of material. Wood is excellent because it is rough and it is an insulator (so that it won't in turn be burning hot when you take it out of the boiling water). Some plastics might work, but as you can superheat water in a plastic vessel, I wouldn't count on it to be rough enough. Metal, per Gandalf, should be considered bad due to the potential for unpleasant interactions with the microwave, plus it still might not be rough enough and would certainly become dangerously hot. Tango's understanding is incorrect; the object must be inserted before the potential superheating occurs; otherwise, all you'll do is trigger the explosive boilover when you insert the object or otherwise disturb the water, and that will burn your hand. In a laboratory environment, the role of the wooden spoon is filled by boiling chips. So, wrapping up: wood is the material of choice because it provides nucleation to prevent superheating, is insulative to protect your hand, and does not interact negatively with the microwave. Other common household kitchen options generally fail at least one of those three criteria. — Lomn 17:29, 1 October 2012 (UTC)

My understanding is that you insert the spoon carefully, at arm's length. I wouldn't say the boilover from superheating is "explosive" - when people get significant burns from it, it's usually from actually holding the cup at the time (you might get splashed a little otherwise, but not enough to cause serious harm). --Tango (talk) 11:13, 2 October 2012 (UTC)

While you could create a wooden object smooth enough to avoid nucleation sites, by applying a lacquer, I don't believe such lacquers are used on wooden cooking spoons, as they could come off when stirring boiling liquids (especially if they contain hard objects, like bone). StuRat (talk) 18:23, 1 October 2012 (UTC)

Thank you all for the explanations. On what length scale should be the roughness? (Strangely enough, all the microwave oven manuals I've seen only mention a "spoon", saying nothing about its material.) bamse (talk) 18:33, 1 October 2012 (UTC)

The manuals will have only mentioned "wooden spoon" specifically because it's the only sufficiently rough object that most people are likely to have in their kitchen that they don't mind putting in their food. Someguy1221 (talk) 23:18, 1 October 2012 (UTC)

Well, it's a fairly large scale, since you can feel the difference between wood and metal. Metal looks rough under a powerful microscope, but apparently that's not enough. StuRat (talk) 20:50, 1 October 2012 (UTC)

Thanks. I had hoped for an answer in terms of some fundamental constants. Do I understand correctly that this length would depend only on properties of the liquid and not on the microwave or the container? bamse (talk) 22:44, 1 October 2012 (UTC)

Yes, I imagine there's a bell-shaped curve showing the ideal scale of roughness, but I haven't found it. The properties of the liquid certainly matter, but the characteristics of the microwave and container also play a role. Obviously, a rough container will promote nucleation directly. Also, the microwave and container control how evenly the liquid is heated, which is an important consideration when super-heating a liquid. If heated unevenly, parts may get hot enough to boil, even without nucleation sites. To maximize the possibility of super-heating the liquid, you want it to be evenly heated and to avoid all sound/vibration. StuRat (talk) 23:09, 1 October 2012 (UTC)

for the claim a metal spoon isn't rough enough. Lomn has suggested it may not be, but I looked before anyone replied (I didn't reply because I couldn't find any good sources), and couldn't find anything suggesting that would be the case. I'm not even convinced the roughness matters much, simply the shape of the object and disruption of the surface may be enough. Many sources simply say a 'non metallic' object, but the reason for specifying non metallic is likely for the reasons highlighted above, not because it may not be rough enough. Otherwise they wold likely also caution to avoid something made of glass (in fact I even came across some sources which suggest a glass rod). Some of the examples I came across don't give a wooden spoon as an example at all, more commonly a skewer or stir stick. I can't speak for how things are in the US or elsewhere, but I would say it's rare to have a wooden spoon suitable for these purposes in most households in NZ, you may have wooden spoons for stirring pots but they will often be too long to fit in the microwave and even if not, often too large to fit in a cup so I would question the claim people are likely to have one in their kitchen. I.E. It's likely a bad idea to read too much in to the fact some sources suggest a wooden spoon. Plastic spoons are more likely, they've often not mentioned I presume because of concerns they will melt or at least leech rather then the believe they won't work. Nil Einne (talk) 08:55, 2 October 2012 (UTC)

I'm skeptical about a glass rod working. After all, if glass worked, then the interior of a glass cup would presumably also work. StuRat (talk) 08:59, 2 October 2012 (UTC)

The rod will have a sharp(ish) edge which will include nucleation sites even if most of the rod is too smooth. I don't think you need much area over which bubbles can form to prevent superheating. --Tango (talk) 11:13, 2 October 2012 (UTC)

I was thinking of glass rods with hemispherical ends. I think "more is better" when it comes to nucleation sites. StuRat (talk) 18:06, 2 October 2012 (UTC)

This fish prefers deep waters and it is illiophagus

Is illiophagus even a word? I can't find it in any online dictionary. Ebaychatter0 (talk) 15:53, 1 October 2012 (UTC)

It's spelt "illiophagous", and it means that the fish in question ingests mud. Dominus Vobisdu (talk) 16:07, 1 October 2012 (UTC)

• You beat me to it! I'm glad to see the OP is paying attention - I once found a vandalism "glucojasinogen" that was left in an article so long it was plagiarized to two different scientific journals... :) Wnt (talk) 16:12, 1 October 2012 (UTC)

I can't find illiophagous either. Google suggests oligophagous. Ebaychatter0 (talk) 16:31, 1 October 2012 (UTC)

On Google, under where it says "showing results for oligophagous", click on "search instead for illiophagous". That will show you the results for "illiophagous". Dominus Vobisdu (talk) 16:40, 1 October 2012 (UTC)

As to the possibility of "ingesting mud", it could be true, if it then filters out it's food from the mud. I believe there are some crabs that do this. StuRat (talk) 18:06, 1 October 2012 (UTC)

That would be the reason that catfish are not kosher. ←Baseball Bugs ^{What's up, Doc?} carrots→ 07:05, 2 October 2012 (UTC)

Is this in a WP article somewhere? If so, perhaps it needs to be restated in plain English, or a translation provided. Alansplodge (talk) 12:33, 2 October 2012 (UTC)

Waves going toward or away from shore

I'm on a Caribbean island with a westward-pointing pier on the western shore. In the morning as I walk out the pier, I see that the wind is blowing toward the sea from land (i.e. from east to west, so the land-sea breeze augments the prevailing trade winds); and the waves appear to be going in the same direction--away from shore. This fits with my understanding that waves are normally wind-driven. Yet waves are also lapping at the shore, coming west-to-east. What drives them against the wind? Duoduoduo (talk) 17:23, 1 October 2012 (UTC)

While wind can create waves, it's not very effective and cancelling waves. So, those waves may have formed out at sea, where the wind direction is different. StuRat (talk) 18:04, 1 October 2012 (UTC)

I don't think that's it. Farther out to sea, the wind direction certainly from east-to-west (the direction of the tropical trade winds), so that doesn't explain waves coming in west-to-east and hitting the west shore. In any event, let me ask a related question: why is it that at any given time waves are coming into the shore on all sides of the island--i.e., in every possible direction? Duoduoduo (talk) 19:14, 1 October 2012 (UTC)

There's an inherent asymmetry between motion toward and away from shore. Waves can be travelling in all directions out at sea. Indeed, I have never seen only coherent unidirectional waves without some very strong extrinsic reason. But as any wave which happens to be moving toward the shore approaches, it will slow down, rise up, and break. This also has the tendency of refracting waves moving obliquely along the coast so that they bend to move in paralleling the coast. Those are the waves we see as they rear up and come in face on to the coast. Depending on the steepness of the shore, you can also see waves reflected back out coherently, causing a zipper effect as incoming and outgoing waves intersect. μηδείς (talk) 19:24, 1 October 2012 (UTC)

You say that waves can be travelling in all directions out at sea. What happens when waves arrive at the same place from opposite directions at the same time? Do they both just keep right on going unless there's some relation between their frequencies or wavelengths? Duoduoduo (talk) 19:49, 1 October 2012 (UTC)

Yes, waves pass right through each other. This is true even if they have the same frequency/wavelength. However, you can get interesting effects in these cases, like a standing wave or a null point. StuRat (talk) 20:47, 1 October 2012 (UTC)

I recently learned about seiches, which are resonant standing waves in enclosed bodies of water, like lakes or even some seas. Pfly (talk) 22:44, 2 October 2012 (UTC)

Russian loudspeakers

I buy a couple of loudspeakers, brand “Radiotehnika” and the enclosoure say “4 Om”… the receiver I currently have is only 6 Ω… I just pluggued the loudspeakers this morning and and the system sound fine to me but…

what are the risks to the new loudspeakers blowout or burn to the hell or do any other electric bad thig to my receiver?? — Preceding unsigned comment added by Iskander HFC (talk • contribs) 17:42, 1 October 2012 (UTC)

It should work OK as long as you do not drive your receiver close to its limit. Keep the volume backed off 50% and the chances of transisters going up in smoke is much reduced. Graeme Bartlett (talk) 20:51, 1 October 2012 (UTC)

And they should put a label on the volume dial, or physically block it, if possible, to prevent anyone else from turning it to max. StuRat (talk) 23:57, 1 October 2012 (UTC)

Ok… thankyou very much! Iskánder Vigoa Pérez 00:55, 2 October 2012 (UTC) — Preceding unsigned comment added by Iskander HFC (talk • contribs)

There's no need to worry. In the early days (up to early 1960's)of transistor circuitry, transistor prices increased dramatically with power rating, so design engineers would cut things a bit fine. This issue went away with the introduction of low cost silicon parts in the late 1960's. Certainly in the last 30 years, no design engineer of any merit would design a reciever or amplifier circuit that critical - after all, the impedance of a loudspeaker is well understood to be only a nomimal value. In any case, it became the norm from the 1970's onward to include automatic protection circuits in recievers and amplifiers. With modern integrated circuits, this is always done.

In any case, marking the volume control to avoid turning it up is just plain silly. How hard the circuit works is primarily set by the amplitude of the bass, including the very deep bass that is barely audible, but loudness as percieved by the ear is a complex function of the bass, mid-range, and treble components. This means that the power level encountered by the circuitry varies considerably from musical item to item, even when the volume, as percieved, and as set, stays fairly constant. If you have a copy of the CD "Buddy Holly Story (London Cast)", play it on your stereo at a lowish volume setting while monitoring the loudspeaker voltage with an analog voltmeter. You will see a dramatic variation in voltage from track to track even though you aren't changing the volume control setting. (This particular CD is notorious for showing up amplifier weaknesses due to the deep bass in some tracks only).

Keit60.230.201.133 (talk) 01:18, 2 October 2012 (UTC)

The 4 ohm/8 ohm categorization of most speakers is pretty much a fantasy; if you look at the impedance curves with frquency they go all over the place [1] and unfortunately, you can find speakers labeled 4 ohms that are at all frequencies higher in impedance than ones labeled 8 ohms, and so on.... What the 6 ohms on the back of the receiver means, in real world terms, is that the power supply doesn't have the muscle to handle a really difficult speaker system without distorting at higher volume; but that's true of most consumer grade/lower cost amplifiers even though they might not say 6 ohms on the back. Anyway, unless you have a really unusually difficult speaker you'll never hear anything wrong, and you certainly won't hurt anything. Gzuckier (talk) 06:20, 3 October 2012 (UTC)

Connecting the dots in a graph

If you obtain through direct measurement the points 2,2; 3,3; 4,4 in a x-y graph. Is it wrong to connect them with a line? Actually, you don't have the measure 2.5, 2,5, but your graph with connected dots would be showing them, and you could not rule out that y was 3 when x was 2.5. Isn't that a little bit like inventing stuff? OsmanRF34 (talk) 18:24, 1 October 2012 (UTC)

Creating a line or curve to connect graph points is fine. The connection between graph points is an interpolation of the data, and the line or curve beyond the end points is an extrapolation of the data. Interpolation is more reliable than extrapolation. While only 3 data points may very well be insufficient to draw a line or curve reliably, you will never have an infinite amount of data, so some extrapolation is required in order to determine any trend. StuRat (talk) 18:30, 1 October 2012 (UTC)

I understand that in certain scenarios that makes sense. But shouldn't such graphs have an indication of the density of the points and how much interpolation (or extrapolation) was done? OsmanRF34 (talk) 18:35, 1 October 2012 (UTC)

Yes. This is normally done by showing both the data points and the line/curve, on the same graph. Extrapolation is sometimes shown as a dotted or dashed line/curve, to show that it is less reliable. StuRat (talk) 18:38, 1 October 2012 (UTC)

I would expect to see error bars as a indicator of probability density. Also, there would be no justification for drawing a straight line unless there was a reasonable assumption that a linear (or regionally near-linear) model could be used to explain the trend. That is not always the case. Dominus Vobisdu (talk) 18:51, 1 October 2012 (UTC)

Indeed. "But...the line must be a good fit; I used Excel" is the bane of first-year laboratory lecturers. Do you have a model? Did you look at the data? Anscombe's quartet is probably my favorite cautionary illustration for linear regression. TenOfAllTrades(talk) 20:50, 1 October 2012 (UTC)

Indeed - just as first years (and even final years) taking a reading with an instrument and believing that's all there is to it because the intrument has a digital readout or a famous brand name on it. The reality is that you learn more, and detect mistakes, if before you take any readings, you know what to expect - that is you develop a theory that predicts what the readings will be. And if the readings fall where your theory says they will, your theory is confirmed (to a reliability that is dependent on circumstances and the number of readings), and you know how the dots should be joined. Soemtimes you cannot know the theory, but as far as possible, you should work out the theory before coming near any instrument. Ratbone23:45, 1 October 2012 (UTC) — Preceding unsigned comment added by 124.178.43.23 (talk)

If you want to connect your points with lines but need the individual data points to stand out, the usual approach is to mark them, using circles or squares or some other symbol, it doesn't really matter what as long as it is visible. Looie496 (talk) 03:54, 2 October 2012 (UTC)

It's not directly an answer to your question, but the mathematics for producing the best line to fit a given set of data is linear regression. If the experimental uncertainty in the data is known, there are formulas that can be used to help decide whether the graph is a good fit or not. With your three-item dataset, you couldn't rule out the possibility that the next point would be (2.5,3.0), however as the dataset gets larger the confidence one can have in the result goes up.--Srleffler (talk) 17:51, 2 October 2012 (UTC)

After all, there are an infinite number of points between any given pair of points, so even if you have measures at (2.00000,2.00000) and (2.00001,2.00001) you can never guarantee that the actual value at 2.000005 isn't 1,786.2 or some such. So it must be understood that any line between points, no matter how thick the points are, is always hypothetical.Gzuckier (talk) 06:26, 3 October 2012 (UTC)

• Just to add something no one has said explicitly here: the whole idea behind interpolation, extrapolation, and linear regression is playing around with likelihoods and probabilities. For any set of points, there are always an infinite number of polynomial functions which could include those points: In simpler terms you can draw a graph of any shape including any number of points, and that graph can always be described by an equation of arbitrary complexity. If you've got three points, you can draw a line through them, or any number of curves, and you can always reduce that to a function. That is, there is no way to know, with absolute certainty, that the equation you use to interpolate and extrapolate a set of discreet points into a curve or a line is correct with 100% certainty, just likely given what you may know about the data set you are working from, and the physical system which is being modeled. Or as a cynical professor one told me "For any set of data there exists some graph paper which can make the data look linear". --Jayron32 06:37, 3 October 2012 (UTC)

Beta carbon nitride

The article Beta carbon nitride says that the substance is expected to be harder than diamond, but the diagram implies a planar, rather than a solid structure. What am I not understanding here? μηδείς (talk) 19:12, 1 October 2012 (UTC)

The diagram doesn't do the model justice (although ironically this is just a colored variation of the diagram put forth by Cohen himself). Rather, bonding is tetrahedral. It's described in greater detail in [2], if you have access. Someguy1221 (talk) 20:43, 1 October 2012 (UTC)

No, instead of buying the parents of my niebles a one-year's subscription to Science, I bought them a lifetime subscription to National Geographic for their wedding present. Any chance you can link to an image available at Google Images? μηδείς (talk) 22:10, 1 October 2012 (UTC)

If I knew of one, I'd give it to you. What I can do is quote the caption that accompanied the original figure, which is identical to the one in the article but in black and white:

Structure of beta-C3N4 in the a-b plane. The c-axis is normal to the page. Half the atoms illustrated are located in the z = -c/4 plane, the other half are in the z = c/4 plane. The structure consists of these buckled planes stacked in AAA... sequence. The parallelogram shows the unit cell.

Someguy1221 (talk) 22:21, 1 October 2012 (UTC)

Assuming I take that right, it means that the atoms are in a crumpled plane, sort of like an egg crate. But still in a sheet. That still doesn't seem to get us to a solid 3D structure like diamond. Do some of the pictured carbons have bonds that are not shown? Arent all the bonds where the carbon and nitrogen atoms are shown as touching, rather than linked, double bonds? It seems to me every carbon has two single and one double bond in the crumpled plane, while every nitrogen has one double and one single. μηδείς (talk) 02:08, 2 October 2012 (UTC)

The paper describes the nitrogen atoms as being sp2 hybridized and the carbon atoms as being sp3 hybridized, with each carbon atom being linked to four nitrogens, so yes, there must be bonds in the model that aren't shown. Someguy1221 (talk) 02:18, 2 October 2012 (UTC)

Ah, and so the bonds are actually intermediate between those implied in the diagram? Each carbon shown is bonded to one not-shown nitrogen, and most of the nitrogens shown to a carbon not in the plane? Can you edit the image in the original article accurately to fit that, so future readers are not as flummoxed as I was? μηδείς (talk) 03:12, 2 October 2012 (UTC)

I would, but I still don't know what it actually looks like! With that image, the caption, and what I just told you, you now know as much as I do about the structure of beta carbon nitride. I might look into some of the references Cohen provides to see if any of them are enlightening. Someguy1221 (talk) 02:34, 2 October 2012 (UTC)

Well, at least you can see the source. perhaps you could at least add that certain bonds not in the plane are not indicated? μηδείς (talk) 03:12, 2 October 2012 (UTC)

I also have access to the source, but it really provides no further visual information except as the above. The black-and-white original provides slightly more depth (because it's more evident which bonds are coming up or going down relative to the viewer) but the missing C-N bonds (the third bonds on the smaller N atoms, the fourth bonds for the larger C atoms) are not drawn. Adding them is not simple - I'm not sure what angles they would take (probably perpendicular to the sheet). Absent a reliable source, accurately drawing this would require a 3-dimensional model with multiple layers (i.e., a fair amount of original research). -- Scray (talk) 04:17, 2 October 2012 (UTC)

No, I wouldn't expect you to draw the bonds. I just thought you could update the caption text to mention them in a way you believe is accurate from the source. μηδείς (talk) 17:26, 2 October 2012 (UTC)

Sorry I misunderstood. See what you think of this edit, which I hope is an improvement. -- Scray (talk) 18:32, 3 October 2012 (UTC)

My assumption was that only the nitrogens showing two bonds has a third out of the plane; some already show three. I have modified the blurb--please revert it if I am wrong. Also, should the sp2/sp3 bonds be mentioned? last I studied orbitals was 1986. μηδείς (talk) 20:44, 3 October 2012 (UTC)

Sorry I did not notice those, and thanks for correcting. I don't think we need to bring in specifics about those bonds - the caption is already long. -- Scray (talk) 22:58, 3 October 2012 (UTC)

Clouds and pressure

Why does a sudden reduction in pressure cause clouds (or vapor) to form? 74.15.136.9 (talk) 22:04, 1 October 2012 (UTC)

Because this implies a low pressure front has moved in, which generally means lower temperatures, too. If the temperature drops below the dew point, clouds and then precipitation occur. StuRat (talk) 22:08, 1 October 2012 (UTC)

See PV=nRT. μηδείς (talk) 22:10, 1 October 2012 (UTC)

Here's a more detailed answer. Generally speaking, air closer to the suface has a larger amount of water vapor in it (in terms of absolute humidity, that is the grams of water per cubic meter of air) because it is warmer (and warmer air can hold more water vapor) and it is of a higher pressure, which means there's more of everything in a given unit of volume. What a low pressure area indicates is a trend of rising air (see Vertical draft for a short explanation or Low-pressure area for a fuller explanation), which means there is a general flow of ground air up to the higher areas of the atmosphere. If you lift a lot of warm, moist, higher pressure air into the upper atmosphere, it cools, spreads out, and releases all of its water vapor as little droplets. Hense, clouds. --Jayron32 22:45, 1 October 2012 (UTC)

StuRat basically has part of the right answer, and Jayron's explanation is also part of the answer, but the full story involves some additional complexities. A sudden reduction in pressure usually means that a low pressure system has moved in. Low pressure systems spin cyclonically (clockwise in the northern hemisphere), which means that they bring in air that is often warm, moist, and unstable. Also, as the low pressure systems move eastward (the usual direction), they tend to create frontal boundaries, which are particularly unstable. Looie496 (talk) 01:38, 2 October 2012 (UTC)

Thanks for the answers everyone. I have two more related questions: when a nuclear weapon goes off, there will often be a condensation cloud that accompanies it. As the article explains, the clouds form because of a reduction in temperature, which in turn is caused by the low pressure that follows a shock wave. My questions are 1) why does this sudden rarefaction of air produce a reduction in temperature? (The answers given above don't seem to work, because the low pressure evidently isn't caused by a influx of cold air. 2) why is there a low pressure zone after the shock wave? Thanks again. 74.15.136.9 (talk) 21:37, 2 October 2012 (UTC)

For #1, see adiabatic cooling which is how a refrigerator works. Energy is a zero sum game, which means that rapid changes in pressure can cause equally rapid changes in temperature if there is minimal energy exchange with the surroundings. That means that a rapid pressure drop, which happens too fast to allow energy to properly equilibrate with the surroundings, will result in an equally rapid temperature drop. This is how both a refrigerator and a diesel engine work, in opposite directions. The refrigerator works by taking a compressed gas and allowing it to expand rapidly, the resulting expansion cools the refrigerator. The compressor in the refrigerator re-compresses the gas to it can be allowed to expand again. In a diesel engine, the pistons generate enormously high pressures, and the massive increase in pressure causes the fuel-air mixture to spontaneously combust. With a nuclear bomb, you've essentially got the refrigerator effect happening: as the air rushes away from the blast site, you create an area of rapidly forming low pressure, which results in a near instantaneous temperature drop. The shock wave situation is the same thing: As a shock wave moves away from its origin, the origin is left with less stuff there, and that rapid pressure drop causes a rapid cooling. --Jayron32 21:52, 2 October 2012 (UTC)

To help visualize why a blast creates both high and low pressure, consider these particles at rest:

XXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXX

Now a blast wave comes along and moves some of those particles into the crests, leaving troughs where they were removed:

X    XX    XX    X
XX  XXXX  XXXX  XX
XXXXXXXXXXXXXXXXXX

The number of particles remains the same (air molecules in this case, or water molecules during a tsunami). The peaks represent high pressure, in the case of an air blast, and the troughs represent low pressure. So, just like you get a receding waterline between tsunami waves, you also get low pressure during a nuclear explosion. StuRat (talk) 02:08, 3 October 2012 (UTC)