Wikipedia:Reference desk/Archives/Science/2012 December 27

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.

December 27

measuring in a hypodermic syringe

How does one accurately measure small amounts of medicine in a hypodermic syringe? Is there medication left in the needle after the injection?Tmary (talk) 01:07, 27 December 2012 (UTC)

(1) The hypodermic syringe usually has markings on the side showing the volume of medicine to be delivered. (2) Yes, there is a small amount of medicine left in the needle after the injection, but it's unusable because it cannot be expelled from the needle, and also because it's mixed with the patient's (possibly infected) blood. 24.23.196.85 (talk) 01:16, 27 December 2012 (UTC)

2) I think what they are asking about is if the unused portion remaining in the syringe is accounted for in the markings. I'm not sure, but guess that this amount is insignificant, either way. StuRat (talk) 01:26, 27 December 2012 (UTC)

(2) All the syringes I've seen are calibrated "to deliver" -- which means that the markings show the volume actually injected. 24.23.196.85 (talk) 01:41, 27 December 2012 (UTC)

also, a standard amount of medicine is in solution by weight per volume in the liquid matrix. So one simply measures a volume of a usually pre-made solution to be injected, rather than measuring some miniscule weight. μηδείς (talk) 01:25, 27 December 2012 (UTC)

Seems like the graduations measure the amount in the barrel - so when you suck the liquid inside, you're getting a little extra inside the needle...then when you do the injection, that exact amount is left inside the needle at the end - so the reading on the barrel should be exactly correct no matter the volume inside the needle. I suspect that the limitations on precision are more to do with the skill and eyesight of the operator than anything else. SteveBaker (talk) 03:46, 27 December 2012 (UTC)

There is a small dead volume in many sorts of equipment. One quick way to avoid it being a problem is to keep its contents constant. That is, if you keep the syringe needle-down, pull the syringe barrel from the 0.0 mL marking to 0.2 mL and then push it out to 0.0 mL, you have picked up and squirted out 0.2 mL. The needle (and the inner part of the Luer lock and other areas) started empty and wind up empty. Or else you could pull to 0.3 mL, invert and push down to 0.2 mL (expelling the air that had been in the needle, etc.). Then when you push to 0.0 mL, you are still ejecting "0.2 mL to 0.0 mL", now indeed wasting the drops in the needle. Either way, you keep the same material in that space, whatever it is does not interfere when the barrel moves a certain number of volume-markings (standard laboratory practice when titrating by volume from a burette). DMacks (talk) 03:48, 27 December 2012 (UTC)

Note that having a constant 'dead volume' in a syringe depends on the contents being liquid (and specifically aqueous, as water is practically incompressible), i.e. the syringe is commonly tapped to bring any air bubble(s) to the top, and the uppermost contents, liquid and gas alike, are expelled down to the desired precise quantity. Wnt (talk) 04:40, 27 December 2012 (UTC)

Something else I should add is that the syringe should be chosen for the amount of med to be given. You wouldn't use a 100 ml syringe to administer 5 ml of meds, as it would be less accurate at that dosage than a 5 or 10 ml syringe. StuRat (talk) 04:43, 27 December 2012 (UTC)

The smallest-measuring syringes in common use are insulin syringes. The smaller ones are 0.3 ml or 0.5 ml, and graduated in units that are equivalent in volume to 0.01 ml. The imprecision of delivery has been studied. With careful technique the error of delivery is 0.003 ml or less, but can be higher with poor technique. alteripse (talk) 06:34, 30 December 2012 (UTC)

decline of space expansion and exploration

I came across an old (pre-70's) transcript of an Earl Nightingale radio broadcast in which he spoke of Seaborg's predictions for the year 2000. Seaborg predicted that moon trips would be commonplace by 1992, and people would be able to go around the world in 2 hours, and that we would be visiting nearby planets by the year 2000. Interesting that he also predicted the 'internet'... he said that people would be able to use radio technology to read books from remote libraries :) Now my question is - why is mankind not devoting resources into space expansion and exploration? Surely, with the large number of extinction events possible, that we would need a 'backup' of mankind in space bases on the moon and beyond? Yes I know the traditional response to this is that it's 'prohibitively' expensive, but surely there's got to be more to it than this? What's the use hoarding or misusing funds if a comet can wipe us out tomorrow? Sandman30s (talk) 09:51, 27 December 2012 (UTC)

Is the transcript of this Seaborg interview available online? Nimur (talk) 18:16, 27 December 2012 (UTC)

No, they were "Defy" (South African electrical brand) transcripts of the radio show that my grandfather collected. Email me and I will scan and send to you. I should probably make them all available online as there some real gems there and I don't believe there is any copyright. Sandman30s (talk) 20:57, 27 December 2012 (UTC)

There almost certainly won't be any comets wiping us out tomorrow because we're constantly monitoring for any nearby objects of sufficient size to do serious damage. Launching anything into space is ridiculously expensive, even for low-earth orbit. Creating a fully autonomous offworld base (in orbit or on another body) is an extremely difficult task by itself; there's an absolutely vast number of ways in which it could go wrong. The combined cost of developing the necessary equipment and launching it into space would be outrageous. The chances of an offworld colony being wiped out would be much greater than the chances of the entire human race being wiped out. If we really wanted to protect the human race from mass extinction events, I'd imagine building underground cities would presently be a much better choice than colonising space. As for space exploration, we are doing that, except we're using robots instead of humans, because it's just that much cheaper and easier. --Link ^(t•c•m) 11:01, 27 December 2012 (UTC)

Yes I know about monitoring near-earth objects, but we currently would not be able to prevent a large object (kilometres in diameter?) from crashing into earth. What's the use having underground cities (or even cities on the ocean bed) if a comet hits or supervolcano erupts? These events would wipe out all surface life as well as all crops, so how would the subterranean dwellers get food? I would also imagine that creating underground cities would be just as expensive as cities on the moon. Yes I do agree that a space base would be far more dangerous, but that's the whole point of creating backup(s)... if your primary source is gone then at least you do have a backup. Sandman30s (talk) 11:17, 27 December 2012 (UTC)

The ocean would be hard to disrupt, especially not quickly. Yes, if an event leads to the whole ocean boiling you're screwed in the long run, but no dinosaur killer or Snowball Earth event ever did that. (A full fledged runaway greenhouse, maybe...) There is actually extensive military-oriented regulation of trading in even a camera that can see things more than a mile deep, so there must be something down there, though the usual explanation involves crashed nuke sub(s). I don't think we can rule out that one or more such bases to shelter elites haven't already been constructed, though I know of no evidence. Wnt (talk) 15:33, 27 December 2012 (UTC)

Can you provide a link to where I may read more about the trading limitations on such cameras? Are you talking about pressure resistant cameras and housings? -- 41.8.83.148 (talk) 05:30, 28 December 2012 (UTC)

You could move underground during the most turbulent period, and then return to the surface once things start to stabilise again. Getting food would be problematic, but it's not quite easy to do in space either. You seem to be underestimating how incredibly inhospitable space actually is. Once you're out of the protective shell of Earth's atmosphere and magnetosphere, you're constantly pelted by radiation. No planet or moon in the solar system beside Earth has a breathable atmosphere. There's nowhere other than Earth where leaving the protection of a spaceship or space suit is survivable for even a minute or so. It would probably still be easier for humans to survive on Earth in the aftermath of a global disaster than it would be on Mars in the most favourable conditions. Sure, it would be useful to have an off-world colony, but the launch costs alone make it utterly infeasible (a space elevator would alleviate that problem, but we're not quite at the point where we can build one yet). Putting a permanently manned base on the moon is just about at the limit of what we could reasonably do right now. Creating a completely autonomous (i.e. capable of surviving indefinitely even if the Earth were to go kaboom) colony for a minimum viable population of humans is very far out of reach for the time being. Ensuring a few thousand humans could survive a global disaster on earth is easier, cheaper, and likely more effective than creating an offworld colony that can repopulate the Earth later. --Link ^(t•c•m) 16:49, 27 December 2012 (UTC)

You could get by the minimum viable population problem by keeping thousands of frozen embryos (or sperm and eggs), so you'd only need a "caretaker" population. StuRat (talk) 21:23, 27 December 2012 (UTC)

Thank you for your thoughts. Let's say terraforming became viable and feasible in the distant future. Would you still say that underground cities would be better than colonies on a terraformed Mars? Sandman30s (talk) 21:07, 27 December 2012 (UTC)

Well, except the runaway greenhouse. If the oceans actually boil and the water vapor greatly enhances the greenhouse effect, so that the whole planet gets a proper steam cleaning, then anywhere, even underground, will eventually be uninhabitable. Getting the energy to run enough air conditioning to live in an oven when you can't go anywhere seems like a long shot, though I suppose in such a thick atmosphere amazing things can be done with wind power (a little late, tho) Wnt (talk) 17:02, 27 December 2012 (UTC)

I am currently looking at tens of thousands of perfect tomato plants sitting on a cement floor while a foot of snow is on the roof above, so getting food while living underground really would not be a problem so long as one has continuous energy (likely nuclear) and a one time supply of water, seed, fertilizer, and growing equipment. 50.101.153.9 (talk) 21:40, 27 December 2012 (UTC)

Right at this very moment people ARE going round the world in two hours (less actually), in the International Space Station. HiLo48 (talk) 22:14, 27 December 2012 (UTC)

Yea, if you don't count the take-off, landing, prep time, etc. Actually getting from the ground in Australia to the ground in the US takes a lot longer. StuRat (talk) 05:33, 28 December 2012 (UTC)

Link: Glenn T. Seaborg - 41.8.83.148 (talk) 05:44, 28 December 2012 (UTC)

quantum effects in processors

How, specifically, do modern processors use quantum effects? (i.e. "don't work without it", etc). This is not homework. 91.120.48.242 (talk) 11:49, 27 December 2012 (UTC)

If nothing else, transistors depend on quantum effects. It's hard to give a definitive answer, because really every property of matter is a quantum effect if you look deeply enough. Looie496 (talk) 15:47, 27 December 2012 (UTC)

But that's not what I mean. When they're designing a processor, they have modeling software (to lay out the circuits). Does this modeling software include quantum effects? 91.120.48.242 (talk) 15:54, 27 December 2012 (UTC)

No. Looie496 (talk) 16:33, 27 December 2012 (UTC)

Thank you, Looie. Is there a size where it would? I found a stackoverflow answer to the question "Is Quantum Physics used in manufacturing CPU's" (June '11). It says they mostly avoid it: "As component sizes have gotten smaller, tunneling has become an increasingly important limiting factor in the design and layout of chips." Is there some kind of size where you could no longer avoid these effects, but, for example, have CPU modeling software (for laying out new CPU's) that actively includes the quantum effects for you to include in the design? I am not talking about a quantum computer per se, just at what point traditional CPU design will directly work with (use) quantum effects, if indeed there is such a size. 91.120.48.242 (talk) 16:56, 27 December 2012 (UTC)

Modern off-the-shelf processors do not, but see http://en.wikipedia.org/wiki/Quantum_computer#Developments. The article states that, "In 2009, researchers at Yale University created the first rudimentary solid-state quantum processor." It goes on to summarize other state-of-the-art developments. --Modocc (talk) 17:06, 27 December 2012 (UTC)

This also depends on what you mean when you say "designing a processor." In modern engineering practice, a computer architect (or a whole team of them) designs a system architecture, including a low-level digital logic representation/implementation of the instruction set architecture. At this layer of abstraction, transistors are "guaranteed" to work. One could say that we build in a sanity-check any place we use an error correction code, because we handle any generic type of bit error without regard to the source of the bit-flip. That bit error could be due to thermal noise, or analog electromagnetic interference, or a timing error due to imperfect trace length, or a cosmic-ray alpha particle striking the transistor, or a quantum fluctuation... but it just doesn't matter. The computer architect doesn't care why the analog circuit might fail; at this layer of abstraction, we just have ones and zeros. Now, usually it is "someone else's job" to make sure that the bit-error rate is as low as possible. An entire team of engineers worries about each source of error: a layout crew makes sure the traces on the silicon get designed properly for minimum clock skew and best signal integrity. A packaging engineer (or a team of them) worries about using the right type of plastic encasement to minimize RFI and EMI and alpha particle strikes.

And finally, somebody - usually not just one engineer, but an entire company full of engineers - worries about what we calll very large scale integration. These guys don't care at all whether the system architect designed a billion transistors that comprise a processor, or a radio demodulator, or a giant billion-transistor no-op. It's their job to ensure that all billion silicon transistors get built right - as designed. This is sometimes called the "Process." Each manufacturer of silicon has their own process, and distribute large complex computer aided design tools that automatically translate a digital design into the correct number of photolithography masks - one mask for each layer of semiconductor, insulator, and conductor layer. It is the job of the fabrication process company to make sure that their masks properly convert a digital design into the analog world. They worry about such things as the optical resolution of their lithography masks; if they choose to use quantum mechanics to analyze the optical path, it may improve their yield. In fact, current semiconductor processes are so small that it's uncommon to use light during photolithography; the wavelength of light is larger than the features being etched. So, they may use x-rays, or ultraviolet light, or any other strange proprietary technique to get around the wavelength limitation. This method can be called a "quantum" effect: you cannot build a single transistor if the uncertainty principle dictates that your photon is larger than the transistor. Another team of engineers - chemists, electrical engineers, material scientists, and so on - worry about etching. They worry about how many seconds of exposure to various etching chemicals are required to remove photomask, and etch away the undesired parts of the semiconductor structure. Now that nano-layer fab processes have such a small number of atoms, the process of acid dissolving solid surface atoms can be modeled quantum-mechanically. This is a molecular physics problem, or a statistical problem. If a quantum-mechanical analysis of this chemistry helps the engineers etch better transistors, it will affect the yield.

The design of an electronic circuit in semiconductor requires accurate knowledge of the dynamic electrical properties of a material. In semiconductors, things like electrical resisitivity and free electron count and even thermal constants are dictated by atomic processes. In very small transistors, such as the ones in a modern process, these properties might require analysis of atomic physics, because there are so few atoms that we can't "average" everything out (as we do in conventional thermodynamics or large-scale statistical physics). An entire field of quantum mechanics analysis - semiconductor physics - is useful in analyzing the electrical and other properties of structures that are built from just a handful of atoms or molecules.

Finally, we can consider the statistical methods of quantum mechanics - independent of their applications to atomic physics. Quantum mechanics helps scientists speak definitively about uncertainty. It helps analyze quantized (countable) effects while also working with analog (continuously varying) phenomena. So, at very low levels of analysis, or at very abstract levels of analysis, we can use methods of QM to analyze things like defect count per wafer, and use that knowledge to guide and modify our designs.

Now, I often hear the pure physicists say things like "quantum mechanics makes the transistor possible." As a former physicist, turned silicon-engineer, I can say my piece: this is utter baloney. It's physicist-talk. Engineers have historically designed, and will continue to design in the future, without understanding the absolutist view of "fundamental processes" that phyicists continually strive for. Engineers are able to work with problems, analytically and quantitatively, while abstracting away the irrelevant details - any detail that doesn't directly affect the current problem. Most engineers in the semiconductor industry - even the ones down low in the process stack - never ever ever use QM or its techniques. Some engineers do. Unlike pure physics research, where scientists use the most sophisticated analysis possible, engineers only use the analytical tools that are needed to get the job done. When designing computer processor logic, QM is almost totally irrelevant. When working with the implementation of that processor as an analog circuit, QM is often irrelevant. But, there are cases where QM improves the process. And there are probably some areas where methods of QM are absolutely essential to the correct design and analysis. But on the whole, what you should recognize is that QM is a method of analysis, not a physical fact. Like all physics, quantum mechanics is a best-effort to accurately model the world as we observe it. It happens to be the best method we have for analyzing atomic physics, and many similar problems. When atomic processes affect analog circuits, we must use QM. But again, let me reemphasize: engineers work by abstracting problems and worrying only about the details that directly affect them. Only half-in-jest, I say that QM analysis is "always" somebody else's job; everyone knows that somewhere in the machine, QM rears its ugly head, but as long as each engineer can make her/his contribution work right, he/she doesn't need to do any extra mathematical juggling. Nimur (talk) 17:51, 27 December 2012 (UTC)

Thank you. I read all this (in a few sessions :). Basically, you are saying that the part that I am interested in - chip architecture design - does not and probably will not use QM, which is more like "noise" that has to be removed by error checks and carefully embodying the abstract design: it is literally at a lower layer and does not make it into chip design. Thus if I understand you correctly, my question is similar to the question "do Internet Protocols like TCP/IP or UDP make use of the fact that light is both a wave and a particle?" and the answer is "No, not at all", and double not for quantum entaglement. That fact has been abstracted away long before they see it. TCP/IP or UDP doesn't even know if it is going over fiber optics, over copper, or over radio. All it gets is the 1's and 0's. You are basically saying the same thing about QM effects: they are abstracted away long before the chip designers see them. Thank you for your detailed answer, Nimur! --91.120.48.242 (talk) 12:58, 28 December 2012 (UTC)

I too found Nimur's exposition an informative read. There are of course things at the lower layers that do not get entirely abstracted away: simulations of implementations can result in high-level logic design choices, e.g. differences at higher design levels may make a difference to the highest clock frequency at which a circuit implementation will function correctly. At times there are low-level aspects that are exploited at a higher level, e.g. in true random number generators, where noise (sometimes of a quantum nature) is the basis of the functionality required. — Quondum 13:28, 28 December 2012 (UTC)

hair/fur, not pelt

Please list the characteristics of squirrel fiber, squirrel tail fiber, and raccoon fiber.Curb Chain (talk) 12:28, 27 December 2012 (UTC)

Please do your own homework. Your quetion reads like it has been copied straight out of an assignment. If not, show us why not. Wickwack 58.169.234.153 (talk) 13:50, 27 December 2012 (UTC)

They tend to grow outward from the skin, and are long and slender. Does that help? ←Baseball Bugs ^{What's up, Doc?} carrots→ 04:34, 28 December 2012 (UTC)

Thanks but can you give more distinguishing charactersitics?Curb Chain (talk) 06:24, 28 December 2012 (UTC)

Hydrogenated Polyisobutene in cosmetics

The entry on Polyisobutene makes it sound like something I wouldn't want to rub on my skin. But I've seen "hydrogenated Polyisobutene" as in ingredient in a number of hand lotions. What properties would make it appropriate for that? --71.189.190.62 (talk) 17:59, 27 December 2012 (UTC)

Let's provide a link to the article in question: polyisobutene. While our article doesn't talk about the hydrogenated version, it sounds like it can be used as a thickening agent to keep the oils from readily evaporating, and doesn't have much of a smell. So, it makes the hand lotion last a long time and not stink. The fact that it's used in chewing gum also makes it clear that it's not toxic. StuRat (talk) 18:07, 27 December 2012 (UTC)

As the article notes, PIB and many related compounds are common synthetic rubbers. I bet you've handled, used, and maybe even worn many things made out of it. We also have an article about hydrogenation. DMacks (talk) 18:39, 27 December 2012 (UTC)

If your cosmetic ingredients gross you out, don't think too much about the urea they frequently add. StuRat (talk) 18:58, 27 December 2012 (UTC)

Daily Comet Ison pic?

I'd like to rig up a picture on my computer's desktop to be a current image of Comet Ison, the one that could become a Great Comet in about a year. Is there a "live" image somewhere on the net I can link to? (By "Live", I mean "taken this week or so".)

Even better would be a time-lapse showing the motion of the comet from its discovery until now, but I'd be delighted just to find a weekly-updated image of what the comet looks like now. TIA - Tarcil (talk) 19:12, 27 December 2012 (UTC)

Mystery mineral

Any ideas?

I've had this item for a number of years, having purchased it from a rocks and minerals gift shop, but have unfortunately forgotten what it was called and have long since lost the handy little label describing it. This may have said something about it being an industrial by-product, perhaps from Eastern Europe or Russia, but I could be misremembering. It's certainly very pretty - the picture doesn't really do it justice. It's iridescent, and the edges are rather sharp. It wasn't too expensive either, around £20 I think, if that helps to narrow it down. Any ideas what it could be? the wub "?!" 21:41, 27 December 2012 (UTC)

Looks like slag. Graeme Bartlett (talk) 21:49, 27 December 2012 (UTC)

Could be schist. --Jayron32 22:19, 27 December 2012 (UTC)

Reminds me a bit of Bismuth crystals.--Gilderien Chat|List of good deeds 22:59, 27 December 2012 (UTC)

Might be silicon carbide. -Modocc (talk) 23:49, 27 December 2012 (UTC)

It seems flat. Is one side different from the other side? If so, how? Also, does it seem especially light or especially heavy for its size, or just of expected weight? I don't know what it is, but I think this information may help more knowledgeable editors to take a guess. Also—is it strong—or would it break very easily if you for instance applied manual pressure to it? Bus stop (talk) 00:04, 28 December 2012 (UTC)

The sides are all similar. I would say that it's quite light, and it's easy to break flakes off it.

Silicon carbide / carborundum definitely seems like the right answer. It looks very like this example, and the fact that it's found on the inside of blast furnaces sounds like what was on the card. Thanks everyone! the wub "?!" 01:04, 28 December 2012 (UTC)

Is it hard? Does it scratch glas? --Stone (talk) 12:11, 28 December 2012 (UTC)

It looks to me like quenched pyrite or galena, both being a sulfide mineral. Plasmic Physics (talk) 02:41, 28 December 2012 (UTC)

To tell some of the above options apart you can do a streak test. Use one of the broken off bits to scratch the rough back of a porcelain tile. bismuth galena and pyrite will give dark streaks, silicon carbide will just scratch the tile. Schist will probably leave a brown or grey mark and slag may leave a gray line of broken glass fragments. Graeme Bartlett (talk) 10:37, 28 December 2012 (UTC)

From what I remember of building up a rock collection when I was younger, the important part is that it's got a titanium coating on it - as someone else said, possibly just slag with the coating. It goes under a trade name because "worthless chunk of stuff with titanium coating" doesn't sell as well. 67.212.112.183 (talk) 08:33, 31 December 2012 (UTC)

Woops, wasn't logged in Lsfreak (talk) 08:36, 31 December 2012 (UTC)