My Idiot Science Question

[Loss Leader]

I have a dumb science question. I mean, it is Iamme level dumb. Let me say, first of all:

1. I checked the web and couldn't find the answer.
2. I'm a lawyer and, in other areas, am known for my sharp reasoning skills.
3. This question is really dumb.

Q. I know how a glass thermometer works. Liquid mercury (or alcohol or whatever) expands and contracts inside a thin glass tube. What I don't understand is where the air goes. A tube with mercury in the ball has an amount of mercury and an amount of air (or nitrogen or some gas). If the mercury expands, there is less room for the air. Where does it go? If it gets compressed, surely the mercury has to work harder to compress it as it rises (so the distance mercury could rise from 90 to 100 degrees would be more than the distance it could rise from 100 to 110 degrees). However, the pictures the thermometer I find show even distances between the degrees, not increasingly shorter distances.

Help.

[ClintonHammond]

I don't think there is any air IN a thermometer...

But what do I know...

[bjornart]

Since thermometers obviously work, and the wikipedia article on mercury thermometers says that although some have mercury and vacuum others have mercury and nitrogen inside, the force from the compressed nitrogen must be negligible against the pressure of the expanding mercury, within the operating temperatures of the thermometer.

But that's just an educated guess...

[hgc]

Dupe!

[hgc]

Loss Leader, you may call it a dumb question, but just thinking to ask that question identifies you as smarter about science than 99.9% of the population.

[Just thinking]

Many thermometers have a volume at the very upper end of the column in which the liquid rises. This may account for where the gas (if any) may reside. And if large enough compared to the column's volume, it may make pressure increases negligible -- after all, I believe the column is very very thin -- the glass just makes it appear larger due to optical magnification.

[Tanstaafl]

If there's gas inside the tube, they just take its pressure into account in the calibration. The effect would be consistent.

ETA: I'm guessing the high quality ones at least have a vacuum.

[Hawk one]

An idiot question is when you already think you know the "answer" and aren't willing to listen to any other explanation.

That being said, I would say that bjornart makes a lot of sense. It is possible that the pressure of any nitrogen present is non-negligible if you're aiming to make a -painfully precise- thermometer (I don't know if such precise mercury thermometers have ever been needed; perhaps meterologists needed them?), but your average thermometer sold to ordinary people most likely doesn't need to be -that- precise, and therefore any pressure difference could easily be neglected. Or so I imagine.

[Loss Leader]

Thanks for the quick and useful input.

[Metullus]

Thanks for the quick and useful input.
Said the barmaid to the...er, never mind...

[Arkan_Wolfshade]

http://home.howstuffworks.com/therm.htm

[jimbob]

Yes, a torelli vacuum: mercury vapour fills the rst of the volume.

It works because the volume of the bulb is a lot greater than the volume of the tube (otherwise the glass would expand, and introduce another variable into the system).

Remember that a mercury thermometer is only used over a narrow range of temperatures.

Absolute Zero was postulated from an "ideal gas", and indeed helium gas thermometers have been used, I seem to recall.

Platinum resistance thremometer are also used as reference thermometers (so must also be very linear) As the wire gets hotter, the resistance rises.

Jim

[Schneibster]

The correct answer (or more precisely, WHY it is the correct answer) is more subtle than it appears at first.

ETA: and by the way, Loss Leader, that subtlety makes it a much less dumb question than you thought it was. ;)

From a practical point of view, Hawk one has it right: the change of pressure of any gas in the thermometer tube is negligible compared to the force of the change of volume of the liquid. This is because there is far more interaction between the molecules of a liquid than there is between the molecules of a gas, and they are therefore closer together. Most liquids are essentially incompressible. That's why hydraulics works.

Most household thermometers use alcohol colored with red dye. Medical thermometers, which must be more accurate and more precise than household thermometers, use mercury, which is less compressible than alcohol. I don't know whether medical thermometers are evacuated, but I know that alcohol thermometers are not. I suspect that medical thermometers are, but cannot substantiate that guess.

[ponderingturtle]

A high quality mercury thermometer could have a vacuum, but an alcohol wouldn't, in part because it is likely the alcohol would boil and ruin it anyway(this would happen to a lesser extent with the mercury)

either way it is not really a big deal, as it just requires the thermometer to be calibrated properly. and say varying the diameter of the glass could make it seem more linear if it gets too non linear.

I will have to check on my candy thermometer but I think it is mercury and likely a vacuum.

[Faustina]

Medical thermometers, which must be more accurate and more precise than household thermometers, use mercury

Kind of off topic, but as far as I know, hospitals and Doc's offices use digital thermometers now. I haven't seen a mercury one in a clinical setting.

[Schneibster]

Kind of off topic, but as far as I know, hospitals and Doc's offices use digital thermometers now. I haven't seen a mercury one in a clinical setting.My bad. You're correct; I saw an article once upon a time that said they are more accurate and precise than mercury, justifying their use in a clinical setting, particularly when the risk of accidental exposure to mercury is considered.

[Hawk one]

From a practical point of view, Hawk one has it right.

Well, to be fair, it was mostly bjornart who got the main idea right from a practial POV, I just added a couple of details I thought would be relevant. Anyway, this thread is one of those "A-ha!" things that are about seemingly trivial details of stuff that happens in the universe, but where posts like yours (and others here, to be sure) still make you feel like you learned something cool.

[Schneibster]

Who was it who said, "To see the universe in a grain of sand...?" I've always thought that was a penetrating observation.

ETA: William Blake. No surprise there.

To see a world in a grain of sand
and a heaven in a wild flower,
hold infinity in the palm of your hand
and eternity in an hour.
-William Blake

[anor277]

A high quality mercury thermometer could have a vacuum, but an alcohol wouldn't, in part because it is likely the alcohol would boil and ruin it anyway(this would happen to a lesser extent with the mercury)

either way it is not really a big deal, as it just requires the thermometer to be calibrated properly. and say varying the diameter of the glass could make it seem more linear if it gets too non linear.

I will have to check on my candy thermometer but I think it is mercury and likely a vacuum.

Just on topic, it is ridiculously easy to introduce an excellent vacuum in any glass tubing with whatever filling; i.e. suck on the tube (with a vacuum pump of any level of sophistication) and then seal the glass with a flame - the outside atmosphere pushes the molten glass inwards to form a gas tight seal. If there is volatile liquid in the tube (say alcohol, or even a gas), simply cool the base of the tube so that the liquid is not volatile, suck on the tube again, heat, and voila. The "plastic" properties of glass (its ability to make a gas tight seal with very crude instruments) are undoubtedly the reason why glass is used in so many scientific instruments.

[skeptigirl]

For Pete's sake people! No offense to anyone here but some of you need a review of basic science:

gases compress under pressure, liquids don't

both gases and liquids expand when heated and contract when cooled

expanding gases or liquids inside a sealed container (in other words a constant volume) results in increased pressure

You put the liquid in the tube and seal it. You vary the temperature which you are measuring with a second device (calibration).

You then mark where the liquid level is at each temperature gradient.

You don't need a vacuum! The air molecules just get closer together and the pressure inside the thermometer increases.


This doesn't mean anyone is dumb, it means you need to get out more.

[Loss Leader]

Thanks to all. It was one of those things that - although there were plenty of sources on thermometers on the web - never got answered. I'm glad I have you guys as a resource.

[Faustina]

My bad. You're correct; I saw an article once upon a time that said they are more accurate and precise than mercury, justifying their use in a clinical setting, particularly when the risk of accidental exposure to mercury is considered.

Speaking of exposure to mercury, I can remember when I was a little girl my brother and I broke a mercury thermometer. We had fun for hours rolling the mercury around, watching it run together and make a larger ball, splitting it up and watching it run together again. In our hands.
Explains a lot about my present day behavior.

[ail]

skeptigirl,

You're on the money here.

I am glad that some of us were awake during that science class.;)

[Zep]

Liquid thermometers also have errors due to the meniscus effect on the top of the thin column of liquid - the height of the column adjacent to the glass wall is different to the height at the centre of the column, either more or less depending on the nature of the liquid used, the diameter of the column, the cleanliness of the glass walls, etc, etc. Another reason why they use more linear and consistent measuring methods where high accuracy is required.

http://en.wikipedia.org/wiki/Meniscus

But for every day use, it makes little difference.

[Dilb]

For Pete's sake people! No offense to anyone here but some of you need a review of basic science:

gases compress under pressure, liquids don't

Incorrect. Liquids and solids are compressible, just not very much.

You then mark where the liquid level is at each temperature gradient.

First, you'd mark it at a temperature, a temperature gradient is the rate the temperature changes in any direction. Second, that would be really, really slow, so very few thermometers, if any, are made that way.

Loss Leader, you're right that as the mercury expands, it has to push air out of the way, which means the mercury does physical work, using up some of it's thermal energy. However, measurement is done at a constant temperature, so any energy the mercury needs to compress the air is given to it by the temperature. The extra work done by the mercury won't change the volume of the mercury, but it will slightly increase the amount of energy that needs to go into the mercury in order for it to reach the final temperature.

However, if the air is compressed, it's at a high pressure, and the air is then compressing the mercury. This will cause the distance between degrees to be slightly shorter. Fortunately for thermometers, gases at room temperature are about a million times more compressible than liquids, so the mercury can compress a lot of air without being noticeably compressed itself.

[Modified]

Speaking of exposure to mercury, I can remember when I was a little girl my brother and I broke a mercury thermometer. We had fun for hours rolling the mercury around, watching it run together and make a larger ball, splitting it up and watching it run together again. In our hands.
Explains a lot about my present day behavior.

My brother and I did the same thing, only we found the mercury from a broken thermometer rather than breaking one ourselves. I remember trying to take it home wrapped in a handkerchief (the surface tension is not quite high enough for that to work, it filters through).

I sometimes worry about all the crap I was exposed to as a child. Then again, kids these days are protected from every little thing, and the results are not encouraging.

[skeptigirl]

Incorrect. Liquids and solids are compressible, just not very much.I thought about getting totally perfectly fanatically precise but considering the way the thread had gone it just wasn't necessary for the situation.The same with all your details on calibration and I certainly didn't think someone sat around making individual thermometers. I was talking about the concept you wouldn't need to worry about the affect the increasing air pressure had because you would resolve that with how you calibrated your instrument.

So I'm sure some people reading the thread appreciate your details and while I find no fault with them, try to consider the fact we all don't have time to write or read dissertations.

Perhaps I might suggest instead of claiming someone is "incorrect", you might think more of adding additional details for those that would be interested. The amount a liquid compresses under pressure is irrelevant in an average thermometers and in brake lines.

.... Second, that would be really, really slow, so very few thermometers, if any, are made that way.Duh! I do hope that was obvious.

...However, if the air is compressed, it's at a high pressure, and the air is then compressing the mercury. This will cause the distance between degrees to be slightly shorter. Fortunately for thermometers, gases at room temperature are about a million times more compressible than liquids, so the mercury can compress a lot of air without being noticeably compressed itself.The meniscus is a much more critical issue than compression of the liquid is. In fact, if you want to get into the minutia I'd wager the width of the temperature mark is greater than the width of additional mercury compression resulting from the increasing air pressure in any thermometer you purchased outside of a scientific instrument supply store.

[Schneibster]

You know, I don't mind seeing five different explanations that all say the same thing; it's good for the newbs. What I mind is when someone posts the same explanation I did and then says mine is wrong. You might want to look into that, sg.

[skeptigirl]

You know, I don't mind seeing five different explanations that all say the same thing; it's good for the newbs. What I mind is when someone posts the same explanation I did and then says mine is wrong. You might want to look into that, sg.
If sg is me, I didn't say you were wrong so I'm sorry if it appeared that way.

I was just taken aback by how many people were posting about a "vacuum" in the thermometer. Your post and Bjorn's had the information about compressing the gas but it was a bit obscured and seemed to have been missed with the discussion of nitrogen and alcohol vs mercury. The link to "the way things work" didn't mention the gas as far as I could find.

What happened to the air was the main question in the OP.

Think about that question then look at the following answers which all discuss vacuums and pressure:

I don't think there is any air IN a thermometer...


the force from the compressed nitrogen must be negligible against the pressure of the expanding mercury

thermometers have a volume at the very upper end of the column in which the liquid rises. This may account for where the gas (if any) may reside. And if large enough compared to the column's volume, it may make pressure increases negligible -- after all, I believe the column is very very thin --

If there's gas inside the tube, they just take its pressure into account in the calibration.

It is possible that the pressure of any nitrogen present is non-negligible if you're aiming to make a -painfully precise- thermometer...any pressure difference could easily be neglected.

Yes, a torelli vacuum: mercury vapour fills the rst of the volume....It works because the volume of the bulb is a lot greater than the volume of the tube

Hawk one has it right: the change of pressure of any gas in the thermometer tube is negligible compared to the force of the change of volume of the liquid.

A high quality mercury thermometer could have a vacuum, but an alcohol wouldn't, in part because it is likely the alcohol would boil and ruin it anyway(this would happen to a lesser extent with the mercury)...either way it is not really a big deal, as it just requires the thermometer to be calibrated properly

it is ridiculously easy to introduce an excellent vacuum in any glass tubing with whatever filling; i.e. suck on the tube (with a vacuum pump of any level of sophistication) and then seal the glass with a flame -

I read all those and thought none of them answered the question that was being asked. I see now my comment about basic science was in error. I saw more about the vacuums and missed the few comments about pressure.

I should have said something like, sometimes people just need a simple answer, not a complex one.

[ponderingturtle]

Liquid thermometers also have errors due to the meniscus effect on the top of the thin column of liquid - the height of the column adjacent to the glass wall is different to the height at the centre of the column, either more or less depending on the nature of the liquid used, the diameter of the column, the cleanliness of the glass walls, etc, etc. Another reason why they use more linear and consistent measuring methods where high accuracy is required.

http://en.wikipedia.org/wiki/Meniscus

But for every day use, it makes little difference.

But with the hair thin glass tube how much meniscus is there?

[ponderingturtle]

The meniscus is a much more critical issue than compression of the liquid is. In fact, if you want to get into the minutia I'd wager the width of the temperature mark is greater than the width of additional mercury compression resulting from the increasing air pressure in any thermometer you purchased outside of a scientific instrument supply store.

So the width of the mark is much thicker then the meniscus at that radius.

[Blue Mountain]

Going a little bit off-topic here, I recall hearing more than once that mercury thermometers were not allowed on airplanes. For a long time I wondered why, until I read the Wikipedia article on mercury (http://en.wikipedia.org/wiki/Mercury_(element)#Mercury_and_aluminium). The bodies of most jet aircraft these days are made out of aluminum, and mecury reacts chemically with aluminum. A small amount of mercury can damage a large amount of aluminum.

[Loss Leader]

But with the hair thin glass tube how much meniscus is there?

I was told in science class to read the bottom line of the meniscus. All I remember was the meniscus was noticeable.

[ponderingturtle]

I was told in science class to read the bottom line of the meniscus. All I remember was the meniscus was noticeable.

In a thermometer? I have never seen one.

[Loss Leader]

In a thermometer? I have never seen one.

Yeah, I don't know what I'm remembering. Maybe it was just in beakers and stuff. I took chemistry twenty-one years ago.

[hgc]

Yeah, I don't know what I'm remembering. Maybe it was just in beakers and stuff. I took chemistry twenty-one years ago.
You were probably just measuring some other liquid in a graduated cylinder.

[ponderingturtle]

Yeah, I don't know what I'm remembering. Maybe it was just in beakers and stuff. I took chemistry twenty-one years ago.

Maybe a mercury barometer.

[Schneibster]

If sg is me, I didn't say you were wrong so I'm sorry if it appeared that way. We're cool. Thanks.

[joe87]

So far in this thread no one has mentioned the need to "shake down" a mercury thermometer after it has been used to take someone's temperature. This is especially necessary if the last user had a fever. Why is it necessary? Because the mercury at the top of the column remains "frozen" in place after use, thus continuing to indicate the maximum temperature previously encountered.

What happens is the volume of the mercury indeed shrinks when the thermometer returns to room temperature, but one or more gaps appear in the mercury column, rather than all the mercury being displaced down toward the reservoir bulb. These gaps presumably contain only mercury vapor, otherwise they would not completely collapse when the thermometer is shaken down before the next use.

The reason the mercury is not forced down towards the bulb as it cools is because the pressure above the column is not enough to force it down. The pressure of whatever gas there is above the mercury is not enough to overcome the capillary action of the mercury in the narrow tube. This suggests that there is not much, if any, gas above the mercury.

This effect can be exaggerated if you "shake up" the thermometer, that is, you use the same whipping motion on the thermometer, but reverse its direction in your hand so that the force of the whip action drives the mercury away from the reservoir bulb.

Many years ago when I was an child (and not particularly concerned with damaging it) I tried "shaking up" a thermometer just to see what would happen. As I recall, part of the mercury was driven up all the way to the top, indicating that there was no gas above the mercury column, at least for that particular thermometer.

This did result in a large gap in the mercury column, but by subsequently shaking it down, I managed to restore the thermometer to its previous state, with my parents being none the wiser.

Anybody have a mercury thermometer they would like to use to try shaking it up to check out this recollection? Of course, if there is some air above the column and you drive the mercury past it, you may never get the air bubble back above the mercury, and this may render the thermometer forever inaccurate.

[Zep]

But with the hair thin glass tube how much meniscus is there?Depends on the accuracy you wanted.

For instruments like basic medical thermometers with narrow bores and a wider error-bar allowance, it made no never mind. A quarter or half degree error in a temp reading was just not significant.

But for precise scientific instruments requiring higher accuracy, they used larger diameter columns (to take into account the various issues mentioned above), but then meniscus did become a problem. Tenths of degrees were important, so care was needed. (Rule: Measure from the top of the meniscus, and avoid parallax errors!)

It was the same issue for any instruments involving pressure forcing a liquid up a narrow bore tube for measurement - thermometers, pressure guages, barometers, etc.

[ponderingturtle]

Depends on the accuracy you wanted.

For instruments like basic medical thermometers with narrow bores and a wider error-bar allowance, it made no never mind. A quarter or half degree error in a temp reading was just not significant.

But for precise scientific instruments requiring higher accuracy, they used larger diameter columns (to take into account the various issues mentioned above), but then meniscus did become a problem. Tenths of degrees were important, so care was needed. (Rule: Measure from the top of the meniscus, and avoid parallax errors!)

It was the same issue for any instruments involving pressure forcing a liquid up a narrow bore tube for measurement - thermometers, pressure guages, barometers, etc.

To get a larger bore you would have to have a every large volume of mercury at the base as well.

It sounds a bit large to be useful for many things.

[skeptigirl]

So the width of the mark is much thicker then the meniscus at that radius.The width of the mark would be greater than the variation in the increments due to the increased force of the gas pressure on the liquid as it expanded.

Just because some thermometers are thin doesn't mean they all are.

[ponderingturtle]

The width of the mark would be greater than the variation in the increments due to the increased force of the gas pressure on the liquid as it expanded.

Just because some thermometers are thin doesn't mean they all are.

Where can I see this giant thermometers?

Thinking more, is it that the tube can be a more consistent size? Thinner is better because you will always be able to make more accurate measurements of changes in total volume with a smaller diameter thermometer.

[EternalSceptic]

So far in this thread no one has mentioned the need to "shake down" a mercury thermometer after it has been used to take someone's temperature. This is especially necessary if the last user had a fever. Why is it necessary? Because the mercury at the top of the column remains "frozen" in place after use, thus continuing to indicate the maximum temperature previously encountered.

What happens is the volume of the mercury indeed shrinks when the thermometer returns to room temperature, but one or more gaps appear in the mercury column, rather than all the mercury being displaced down toward the reservoir bulb. These gaps presumably contain only mercury vapor, otherwise they would not completely collapse when the thermometer is shaken down before the next use.

The reason the mercury is not forced down towards the bulb as it cools is because the pressure above the column is not enough to force it down. The pressure of whatever gas there is above the mercury is not enough to overcome the capillary action of the mercury in the narrow tube. This suggests that there is not much, if any, gas above the mercury.

This effect can be exaggerated if you "shake up" the thermometer, that is, you use the same whipping motion on the thermometer, but reverse its direction in your hand so that the force of the whip action drives the mercury away from the reservoir bulb.

Many years ago when I was an child (and not particularly concerned with damaging it) I tried "shaking up" a thermometer just to see what would happen. As I recall, part of the mercury was driven up all the way to the top, indicating that there was no gas above the mercury column, at least for that particular thermometer.

This did result in a large gap in the mercury column, but by subsequently shaking it down, I managed to restore the thermometer to its previous state, with my parents being none the wiser.

Anybody have a mercury thermometer they would like to use to try shaking it up to check out this recollection? Of course, if there is some air above the column and you drive the mercury past it, you may never get the air bubble back above the mercury, and this may render the thermometer forever inaccurate.

What you mention here is a maximum thermometer, and you are correct in that the mercury must be shaked back, but I do hope you are not worried if I correct your explanation:

There is no need to shake back a normal mercury thermometer, only maximum thermometers require that. The reason is, that in normal thermometers the diameter of the glass tube is more or less constant from the reservoir up to the end of the tube while in maximum thermometers there is a short, very narrow section just above the reservoir which causes the mecury to separate there when it cools down thus keeping the reading near maximum. In normal thermometers you will see separated strings of mercury only in irregular situations (shock, maybe forced cooling or heavy unequal cooling over the length of the tube)


Btw. mercury fever-thermometers are not sold here in Austria anymore. Instead of mercury they use a liquid with similar properties which is not poisonous (don't ask me which one). But the effect is the same.

[jimbob]

Btw. mercury fever-thermometers are not sold here in Austria anymore. Instead of mercury they use a liquid with similar properties which is not poisonous (don't ask me which one). But the effect is the same.

Alcohol, I'd guess.

If you're too hot for an alcohol thermometer your'e undergoing UHT treatment, so fine for medicine. No need to worry about Hg posioning, if the thermometer breaks, just jagged bits of glass

Jim

[EternalSceptic]

Alcohol, I'd guess.

If you're too hot for an alcohol thermometer your'e undergoing UHT treatment, so fine for medicine. No need to worry about Hg posioning, if the thermometer breaks, just jagged bits of glass

Jim



Maybe, but I doubt it. It should be a heavier liquid. It's obviously lighter than mercury (one needs to shake harder), but it should be still much harder if it was the very light alcohol. IIRC I heard something about an alloy of several metals, but I may be mistaken. (The price lets me assume there must be a lot of gold in it :)