Wikipedia:Reference desk/Archives/Science/2016 January 4

= January 4 =

Medication Dose and Weight
In general, do larger adults need higher doses of medication? For example, does someone who weighs 300 lbs need to take more aspirin than someone who weighs 150 lbs in order to get the same therapeutic benefit? I'm not wanting advice about any specific medication: I've always believed larger adults need more (since the adult dose is always higher than the children's dose), but I've never read anywhere that this is really true, and now I'm curious. OldTimeNESter (talk) 12:04, 4 January 2016 (UTC)


 * There is no uniform answer. Some drugs, such as fat soluble compounds, do need to be scaled accordingly.  Other drugs, such as drugs that bind to blood proteins, often don't need be rescaled or would be adjusted by much less than expected based solely on excess weight.  Here is a document giving some examples of dosing guidelines for obese patients for a variety of drugs.    Dragons flight (talk) 12:40, 4 January 2016 (UTC)
 * (EC) Body weight and size can definitely be a factors in determining the optimal dose of medication due to differing pharmacokinetics and possibly other reasons. However the relationship is not simple, for example it will depend on the precise drug. Likewise an obese patient may not require the same increase as someone who's larger but not obese, although pharmacokinetics in obese patients is not always as well studied (but for obvious reasons there's increasing interest). Certain drugs have a relatively high therapeutic index and limited side effects so it may be a certain dose works acceptably well for most adults, particularly if only taken for a short time. With others like anti-coagulants and anaesthesia it's generally more important that the dosage is within a fairly narrow optimal range. Also, size and weight aren't the only factors that may need to be considered. See e.g.         for some discussion of these factors with various drugs. Nil Einne (talk) 12:55, 4 January 2016 (UTC)


 * (EC) Maybe in general, I'd say yes, but it's open to so many caveats. The dose of some medications is prescribed related to the size of a person, but even that is a guideline and the titration needs to be adjusted for the individual until the correct level of medication is reached. A 300lb person may need less of a medication than a 150lb person depending on the medication, the individual physiology, metabolism, general health, etc.

Regarding your point of adult vs child doses, there are many other factors other than body size that affect what is safe or effective, such as the ability for a child's relatively undeveloped internal organs to cope with medication (liver and kidneys spring to mind). Gastrointestinal differences between adults and children are significant when taking oral medication such as the aspirin you've suggested. Children take longer to break down oral medication in the stomach (as they have a lower level of gastric acid) and children younger than 6 months also have slow peristalsis, so the medication is absorbed at a relatively slow rate. I'm guessing that's why medication for children is more often in a liquid form, but I'm certainly not sure of that. Conversely children also have elevated metabolisms so for certain types of medication a child may actually need a higher dosage over time than an adult would. 95.146.213.181 (talk) 13:15, 4 January 2016 (UTC)


 * Thanks, all. It sounds like it's one of those generalities with so many exceptions that it's unwise to rely on it. OldTimeNESter (talk) 23:43, 4 January 2016 (UTC)


 * Generally, up until adolescence many medications are given to children as a function of patient weight (for example, 10 mg/kg body weight) because the size of the blood volume is changing all the way from birth to adolescence. After adolescence this change is much less pronounced, so adolescent children get adult dosages for most medications (at least the ones with high therapeutic indices, where overdosage of a relatively small or young adult is less likely).
 * At adulthood, the blood volume is statistically constant for most people, and that is what determines how much of most medications you get. Body weight is less consequential than blood volume, which is how most drugs get to where they do the job they are intended to.
 * There are exceptions - some drugs, such as the fentanyl derivatives, dissolve into fat readily, so that people with greater than usual fat volume metabolize a certain percentage of those drugs, and a certain percentage goes into the body fat - to come out again later. This accounts for what in the fentanyls is known as the 'renarcosis' syndrome, thought to have killed some of the hostages in the Moscow Theater Hostage Crisis when the very potent fentanyl derivatives used to subdue terrorists and hostages alike may have caused either prolonged coma or an acute renarcosis after doctors on site thought they'd counteracted the agent (misidentified by the Russian Minister of Health in a public statement). loupgarous (talk) 20:45, 7 January 2016 (UTC)


 * Some chemotherapy regimes require the patient to be weighed before each treatment (assuming the infusions are a few weeks apart, not daily). Pharmacists need to calculate the precise dose. Carbon Caryatid (talk) 20:11, 8 January 2016 (UTC)

Italian scientific journal
A citation in a paper is given as "G.Vicentini & Omodei, Cim. 27, p204, 1899". Presumably "Cim" is an abbreviation of the title of some Italian scietific journal. What is its full title? 120.145.144.103 (talk) 15:50, 4 January 2016 (UTC)


 * Perhaps Il Nuovo Cimento. Dragons flight (talk) 16:03, 4 January 2016 (UTC)


 * Yes, and I think the citation is incorrect. The paper at http://www.sciencedirect.com/science/article/pii/S0031891439907032 gives the citation as G. Vicentini, D. Omodei; Nuovo Cim., 27 (1890), p. 204. Looie496 (talk) 16:19, 4 January 2016 (UTC)


 * Thanks to you both. Googling Vicentini Omodei Cim revealed nothing relavent, but googling Vicentini Omodei Nuovo Cimento revealed the paper at the top of the list.  The poster formerly 120.145.144.103 120.145.26.149 (talk) 01:44, 5 January 2016 (UTC)

Are there dozens of models of computational linguistics?
According to xkcd (more exactly ), which is a half-way humorous cartoon but often with a realistic background, there are 12 dozens of contradictory models of computational linguistics. Is that just an exaggeration? If not what are the models? Can't you deal with any problem statistically, provided you get enough data and crunching power? --Llaanngg (talk) 16:09, 4 January 2016 (UTC)
 * The cartoon doesn't say 12, it says "dozens". The reality, I think, is that there are roughly as many models of computational linguistics as there are computational linguists. Looie496 (talk) 16:22, 4 January 2016 (UTC)


 * OK, dozens not 12. But still, some could be non-contradictory and be grouped together. There is no need to count two slightly different statistical models as two. Otherwise, we get as many models as practitioners in any field. --Llaanngg (talk) 17:39, 4 January 2016 (UTC)


 * I thought there might be some more information on explainxkcd.com, but there is very little there. AndrewWTaylor (talk) 17:18, 4 January 2016 (UTC)


 * In the spirit of providing a reliable reference, perhaps some external literary criticism is in order.
 * That specific comic strip was harshly criticized in one of the early posts on xkcd sucks. The author of xkcd is criticized thusly: "...[H]e's looked to as a bit of an authority in the areas of science; when he says something in the context of science many people will take him at face value. And why not? He has a degree in physics, he's worked in the scientific field; he has experience on his side and many people will trust that experience. As such, he has an intellectual duty to be as straightforward as possible in that regard so as to not accidentally mislead his audience. I'm afraid that many people are going to walk away from this with a misshapen view..."
 * As much as it may pain readers of xkcd, Randall Munroe is not actually a science and technology expert: he is a professional cartoonist who incorporates themes of science and technology into his comic strip. Most of his readers don't know or notice the numerous technical errors and category errors that appear in the comic.
 * Strip 114 is a perfect example - which is why it was so called out - because in that work, its author makes a random, non-sequitor claim; munges some technobabble that sounds "nerdy"; and takes an unexpected pot-shot at a fellow cartoonist; and this somehow delights his reader audience, who overwhelmingly perceive the geek-jargon and its non-sequitor conclusion as some form of humor. Meanwhile, educated readers who understand the topic or the references see the strip as "misleading" and "misshapen."  This theme occurs so frequently in xkcd that an entire website, xkcdsucks,  was set up specifically to call out the strips that partake in this form of non-humor and non-science.
 * If you're looking for an alternative to Mr. Munroe's very heavily-promoted new book, "Thing Explainer", I would strongly recommend The Way Things Work, by David Macaulay. Macaulay's books are significantly more technically accurate - even when they take artistic license through their characteristic cartoonish flair.  In addition to his great artwork, Macaulay's works include some incredible prose and narrative; it's a far shot better than Mr. Munroe's stilted minimalism.
 * Nimur (talk) 19:35, 4 January 2016 (UTC)


 * , even if xkcd 114 is inaccurate xkcdsucks thinks almost all xkcds suck (or are mediocre). This is clearly a very humanities person who doesn't share sciency peoples' interests or sense of humor (and sometimes doesn't even get the references). That'd be like me thinking a webcomic about interpretive dance sucks because I don't get it and blogging hundreds of curmudgeonly responses in a row to each one. Also, one thing that xkcdsucks doesn't get is that many xkcds exaggerate for comedic effect. The only people who laugh at jokes on obscure subjects probably would not be fooled in the slightest by the exaggeration and the exaggeration surely would make the in-joke funnier. (and if you didn't know computational linguistics do you really have a right to be pissed at a professor for making your exam answer wrong because you because you got it from a comic strip?) Someone who likes xkcd enough to read it much probably sees subjects they know exaggerated too so they should know not to use it as a textbook.
 * About the books, if you wanted to learn how things work I'm sure Macaulay's book would be a fine source. If you wanted that book to sound like this then read Thing Explainer. It's a surrealistic exercise in explaining complex tech with a lexicon that's too small (the ten hundred most common words). So you get really funny circumlocutions that no Anglophone with enough IQ to explain these would use. It's made for laughing. They're different niches. (Interpretive dance might be the most boring entertainment made by man, though. If I was in an empty cell 24/7 I think I'd barely even turn on a TV stuck on the interpretive dance channel. Unless I liked one of the dancers but they're mostly males anyway.) Sagittarian Milky Way (talk) 23:52, 8 January 2016 (UTC)
 * "Not a science and technology expert" is overstated, I'd say; according to the page Nimur linked, Munroe has a physics degree and has worked in robotics for NASA. That implies at least a certain level of expertise. --76.69.45.64 (talk) 04:28, 5 January 2016 (UTC)


 * As I understand Nimur's comment above, he ackknowledges Munroe has a science background. But that makes things worse. Munroe managed to be seen as an expert when talking about many fields, however, he's not quite careful when talking about science or technology, and he is not focused on his field. He is seeking more the impact factor than accuracy. According to the sources linked above, xkcd could need a little bit of fact-checking. Denidi (talk) 14:57, 5 January 2016 (UTC)


 * I'm going to avoid the other aspects, but the mention of Ryan North was obviously a bit of good-natured ribbing. xkcd and Dinosaur Comics each link to the other, and Munroe has guest-written for the latter. --71.119.131.184 (talk) 21:39, 4 January 2016 (UTC)


 * Macaulay's books appear like a good source for a general view of how stuff works. However, it does not seem to enter into computational linguistics models, although I think it deals with some IT topics. Which literature deals with how language can be computed/modelled/processed? That is, where is the bit about dozens incompatible models refuted. Llaanngg (talk) 00:32, 5 January 2016 (UTC)
 * Computational linguistics from Plato, the Stanford Encyclopedia of Philosophy, reviews the foundations of the discipline and provides a good overview, including a literature survey. Nimur (talk) 01:14, 5 January 2016 (UTC)

Reservoirs are pernicious
In 1837, the Ohio House of Representatives debated expanding a natural lake to form a canal reservoir, and a related document (published on page 390 of the House's journal for that year) concluded by saying:"It may be further observed, that while reservoirs may, to some extent, be pernicious to the health of the adjacent country, yet, in this instance as there would be but little, if any, additional ground covered by converting this lake into a reservoir, there could be no consistent objection made to this arrangement."Is this referring to the reservoir's potential to produce additional miasma, or are they talking about something else? The idea struck me as rather odd, and if it's not miasma-related, I don't know what to think. Nyttend (talk) 22:26, 4 January 2016 (UTC)
 * Were they perhaps concerned about Water stagnation? ←Baseball Bugs What's up, Doc? carrots→ 22:58, 4 January 2016 (UTC)


 * Yes, almost certainly a reference to miasma/malaria, see History of malaria. I know it wasn't eradicated from NYC until last century. μηδείς (talk) 03:44, 6 January 2016 (UTC)
 * Agreed, but it was probably larger than just that. It was only relatively recently that wetland was understood to be important in any context. Nearly two hundred years ago, a marsh would have just been seen as wasted land that could have been used for farming or forestry or something. But, yeah, malaria was probably the response they would have given. Matt Deres (talk) 17:30, 6 January 2016 (UTC)
 * Here's a rather fortuitously relevant source: http://mentalfloss.com/article/73002/fascinatingly-filthy-how-bad-science-saved-lives-victorian-london μηδείς (talk) 21:24, 6 January 2016 (UTC)

Ground glass joints with 2.5 bar pressure relief valve?
I have some pretty thick glassware that I am pretty sure will withstand 2.5 atm of internal pressure (one is a sturdy Erlenmeyer flask used for vacuum filtration). I am trying to use this instead of more expensive pressure reactors because I don't really want to pay for a reactor that can withstand 60 bar of pressure (I am not hydrogenating anything -- my sole purpose is to run reactions in dichloromethane at a higher temperature, like 2.5 atm / 70C. 1 atm / 40C is too low).

I have thought about purchasing made-in-China metal bombs with a pressure capacity of 60 bar, but I wouldn't be able to monitor the reaction, I wouldn't be able to use a magnetic stirbar (I am doing a DCM/water phase transfer catalysis reaction) and I really wouldn't want my reaction to run at 60 bar -- dichloromethane would boil at over 220-250 C (it's off the chart!) and it would basically destroy my sensitive intermediates and products. Or maybe I could use a metal bomb (which are still expensive) with a 2.5 bar pressure relief valve?

if I use a ground glass stopper with silicone grease -- how much pressure would build up inside my Erlenmeyer flask? I have tried looking for ground-glass-to-pressure-relief-valve attachments but I don't think I'm searching correctly. Any insights or possible hacks? Yanping Nora Soong (talk) 22:47, 4 January 2016 (UTC)

Would putting a metal pressure reactor in a water bath be a good moderating idea in place of a customized pressure relief valve? 100C is okay....220 C is way too much. Yanping Nora Soong (talk) 22:53, 4 January 2016 (UTC)


 * Once again,we do not do your hmework for you (see the guidelines at the top of the page), and Wikipedia is not a how-to WP:NOTHOWTO,so could you please explain why you are asking such questions? We can help with homework to a certain extent, but if you need professional advice you should be seeking elsewhere. μηδείς (talk) 04:04, 5 January 2016 (UTC)
 * Please seek expert advice, this can be very hazardous if you get it wrong. Don't trust strangers on the internet. I'm assuming there are people in your institute that can advise on this. (Also, obviously not homework, so just ignore Medeis). Fgf10 (talk) 09:46, 5 January 2016 (UTC)
 * Yeah, don't trust us - these desks are not rated for giving advice, only trying to share references to technical data. But if I were going to give advice I'd advise that if you're not trained in high pressure chemistry, don't mess with it.  There's a reason we never learned this in our undergrad chemistry classes!  It's inadvisable to try to build a pressure cooker bomb with the intent of setting it off at a distance, let alone with the hope and indeed urgent need that it not go off or else while you stand nearby monitoring the reaction!  Surely you should be able to replace the solvent, or add something to it.  This file I found on Google makes some suggestions like dibasic esters that have similar solvent properties and aren't so toxic, and might have higher BP also.  I expect one of the local experts can provide more ideas. Wnt (talk) 15:59, 5 January 2016 (UTC)
 * One important detail you seem to be just starting to recognize: you don't have to run a reaction at the bp of the solvent at your chosen pressure-limit, that's just the maximum temperature you can use. A double boiler isn't exactly essoteric equipment, it's trivial to thermostat an oil bath or laboratory water bath to pretty much any temperature you like. DMacks (talk) 16:06, 5 January 2016 (UTC)


 * As an experienced laboratory chemist, I can't emphasize enough that you should not perform an experiment that will raise the pressure in a sealed flask that is not specifically designed to handle that pressure. Consider other options such as using a higher boiling solvent such as 1,2-DCE in place of dichloromethane.  I will also reiterate the concerns expressed above that you should not be asking for technical advice that has laboratory safety ramifications on a forum such as this.  -- Ed (Edgar181) 17:02, 5 January 2016 (UTC)


 * Why do people think I keep asking homework questions? Also, replacing the dichloromethane is not an option -- dichloromethane is a reagent in the reaction. I'm using a two-phase reaction involving tetrabutylammonium bromide as a phase transfer catalyst. I cannot afford to use methylene bromide or methylene iodide. Yanping Nora Soong (talk) 19:01, 5 January 2016 (UTC)


 * ACE glass flasks and glass reactors designed to withstand high vacuum should be able to tolerate 2.5 atm of internal pressure yes? (That's around 36 psi). I'll probably buy an ACE glass pressure tube, but the pressures they're designed to handle are 150 psi (10 atm) and have been pressure-tested to withstand 15 atm. I guess if I ever intend to run reactions at a higher pressure I'll have a backup. They're sooo long though (20 cm) -- do they have to be completely immersed in solvent? Yanping Nora Soong (talk) 19:08, 5 January 2016 (UTC)
 * You should know better! It's one thing to withstand pressure, another to withstand vacuum.  It's simply a different direction of force.  Compare compressive strength and tensile strength.  And vacuum never gets more than 1 atm away from 1 atm, while you're proposing a 1.5 atm pressure!  There's no way anybody can tell you to try that, except as an experiment to see if it blows up (at a safe distance).  Is what you're attempting something so proprietary and valuable that you can't just ask the chemists here or at university how to do it right? Wnt (talk) 19:23, 5 January 2016 (UTC)
 * At this point, I should point out some other obvious things I didn't think could be an issue, just so we're clear. To start with, I've been assuming that you were working in a fume hood with at least a sturdy raising sash, but now I don't know if you're taking even that precaution.  I'm assuming you would not be boiling dichloromethane in a residential space, even if you don't explode a flask full of it all at once.  I'm also assuming that this isn't some Alexei Shulgin recipe that would expose you to prosecution just for getting the reagents together for, as this is not exactly a discreet or private space for conversation and your purchasing records are very rarely confidential.  Heck, in the U.S. they've prosecuted people on serious felony charges just for having a three necked round bottom flask and a meth recipe!  I do not want to end up reading that an apartment building full of people was evacuated to hospital after toxic vapors poisoned them after someone asked how to make drugs on the Wikipedia reference desk.  Are we clear? Wnt (talk) 19:53, 5 January 2016 (UTC)
 * Minor factual correction - I think you mean Alexander Shulgin, not Alexi. Tevildo (talk) 20:18, 5 January 2016 (UTC)


 * the reaction scheme is entirely of my own design, assembled from literature reactions, using amino acids as feedstock. I'm not making anything illegal or formally scheduled: I just have really bad treatment-resistant complex PTSD, rape trauma syndrome, and intense depression (which I didn't want to mention because I didn't want medical advice) and am making a promising NMDA receptor partial agonist that has been through a few phase I clinical trials. My psychiatrist can only prescribe me on-market meds tho. Also, most of my reactions are pretty mild, except for the reaction involving DCM. No poisonous gases, nothing really toxic etc. Tho I have to figure out where to dispose my phase transfer catalyst (it's a quat). Yanping Nora Soong (talk) 22:22, 5 January 2016 (UTC)


 * ... this is now bordering on medical advice, and we don't give that. Even if you are asking only for scientific advice, our best recommendation, on safety grounds, has to be don't try this at home.    D b f i r s   22:25, 5 January 2016 (UTC)


 * I worked in chemistry research for three years. I dropped out of my program due to depression and PTSD. I can't see how this is more dangerous than deep frying or working at a nail salon. As a first year researcher I would get mildly inebriated simply from cleaning up acetone and diethyl ether spills in the lab. I got pressure tube rated at 150 psi so now I don't have to worry, I guess. Yanping Nora Soong (talk) 22:39, 5 January 2016 (UTC)


 * We are glad to hear that you are not taking unnecessary risks. Exploding glass in your kitchen would be more dangerous than acetone.    D b f i r s   23:36, 5 January 2016 (UTC)


 * Well, since you've told us that much you might as well give us the formula - at least we should scan the literature to see how the authors of that study made it, and see if we can sniff out any ongoing research you might have missed. I think it is instructive for you to work out how to do this synthesis, but not necessarily to do it, definitely not to use it.  I cannot overemphasize how much we should not recommend you taking your experimental substance - even though you're not trying to make drugs, you're making something to take by a new experimental route, which is just as dangerous.  Ever hear of MPTP?  The inventors deserved some kind of medal for the advance of biological science, but I don't think they had a chance to collect!  Do you have a way to characterize this substance, check all impurities?  I doubt one acid-base extraction and a crystallization is going to cut it. Wnt (talk) 01:30, 6 January 2016 (UTC)
 * Also, while I don't know your situation ... it seems like a more straightforward solution is to stick with more reliable treatments while you see if you can get reinstatement or get transfer credit to continue your research training. If you want to try something unconventional, well, you surely know medical marijuana has some partial agonist activity on glutamate receptors and is pretty well known for its use for PTSD. You don't have to use yourself for a science experiment. Wnt (talk) 01:54, 6 January 2016 (UTC)
 * The more I think about this, the more I don't like it. A 150 psi pressure tube doesn't mean you "don't have to worry" - the higher the psi rating, the bigger the boom when your system explodes.  Do you have a gauge to watch what the pressure is?  With an experimental synthesis, you can't know for sure that it won't do something unexpected and produce a gas, so if you're just using a thermometer and expecting the pressure will be X based on a curve for methylene chloride, you could get a big surprise.  Between pressure, fumes, fire, toxic impurities, and the very dubious safety of an experimental drug itself, there are a lot of ways to die here.  This kind of drug development isn't generally done by people who can't afford methylene bromide - people ask for millions of dollars in taxpayer money, or spend exorbitant sums on venture capital, for a reason. Wnt (talk) 13:24, 6 January 2016 (UTC)
 * This is easy to say for someone without mental health issues or PTSD. Do you feel like dying at 1 am every night? Do you get panic attacks and flashbacks so bad you dissociate and lose sense of place and time? Yeah, it's risky, but much less risky than on-market treatments that don't work. Do you have a mental health issue so debilitating that your friends don't understand and keep asking you, "Why do you keep doing sex work when you're so smart?" Successfully treating my PTSD is my gateway out of prostitution and back into scientific research, so yeah. I would buy the experimental product on molport if I could, just that vendors are charging hundreds of dollars to make 500 mg. looool
 * Also according to the vapor pressure curve of DCM, you need a temperature of 120C to achieve an equilibrium vapor pressure of (150 psi / 8000 mm Hg). I'm pretty sure that it would be hard to exceed that even with a boiling water bath. Yanping Nora Soong (talk) 19:47, 6 January 2016 (UTC)
 * I feel terrible to hear that, but there are a lot of different kinds of psychological treatment. I cannot believe that a trained professional doesn't have as much a chance of coming up with some treatment that is more likely to be effective than a chemistry experiment.  There are lots of things already known that affect glutamate receptors - why do you think this one is the one and only option?  It isn't always healthy to mix scientific curiosity and other motivations - what are you going to do if you get a product, and it's 72% pure, and you don't know what the rest is?  And you missed my point above:  you can't go by the curve for DCM, because you have no way of knowing for sure that when you use the synthesis route you devised yourself, that you haven't even had us comment on, that it's not going to liberate ammonia or carbon dioxide or something in sufficient quantity to raise the pressure?  Also, if you won't give the experts a chance, well, this is kind of a wacky idea, but I'm thinking as you're in New York you're close to the last remnants of the Yippies - the narks and the bankers may have moved in on Dana Beal and sacked Bleecker Street, but he still circulates around.   I bet that guy could recommend you twenty people in bad graces with the APA, who can recommend things none of us have ever heard of. Wnt (talk) 20:25, 6 January 2016 (UTC)