Sciencemadness Discussion Board

Correction of thermometer readings

Boffis - 17-5-2019 at 14:17

Most thermometers have on them an "immersion depth" figure which is the length of the thermometer that should be immersed in the liquid in order to get an accurate value, most are 75mm, 100mm or "total". The problem is that I am often working with small quantities and even keeping the vessel small the liquid depth may be only 25-30 mm does anyone have any idea of the sort of error this causes and/or how to estimate it. Is the error positive or negative ie is the reading too high or too low?

Sulaiman - 17-5-2019 at 18:24

Boffis - 18-5-2019 at 02:43

Thanks Sulaiman but I couldn't see how to get what I want out of that formula. I can't see how you accomadate the immersion depth for instance.

wg48temp9 - 18-5-2019 at 10:12

Quote: Originally posted by Boffis  
Thanks Sulaiman but I couldn't see how to get what I want out of that formula. I can't see how you accomadate the immersion depth for instance.

Its a simple problem involving the apparent volume expansion of the part of the mercury/alcohol column that is at the wrong temperature. Either the part below the immersion mark/depth that is out side the chamber and therefore at room temperature or the part that is above the immersion mark/depth that is at the temperature to be measured and not at room temperature/calibration temperature..

Typical apparent volume expansion of mercury in glass is 0.00018/K. Meaning that the length of the mercury column changes by that ration for each K error in temperature. I convenient length unit is the temperature marks on the thermometer. So the error is simply the product of the length of the column in K marks time the error temperature + or - depending the sign of the temperature error.

For accuracy the mean temperature of the column that is at the wrong temperature should be used and the correct apparent volume expansion for the particular thermometer. However as the correction is small you can use it to estimate how significant the error is for you case.

[Edited on 18-5-2019 by wg48temp9]

SWIM - 19-5-2019 at 09:49

So this problem would apply as well to distillations, right?

The thermometer immersion depth must be right for that particular distillation head or corrections may be needed?

I can't believe I never considered this before.
May explain some of the mysterious anomalies I have gotten sometimes distilling things.

I assume the ground glass jointed thermometers used for some distillation heads are meant to be 'immersed' in the vapor up to the joint.
If this is so, then a short ground glass jointed thermometer like those used in shortpath distillations would have an immersion depth of only 25 mm.
Perhaps (assuming what I just said actually accords with reality) you could use one of those.
You'd have the joint in the flask just above the liquid level, but that shouldn't be a big problem.

CobaltChloride - 19-5-2019 at 13:35

I always assumed that for distillations you have to place the thermometer so that the bulb is just barely fully submerged in the vapor front as it's traveling towards the condenser. This is what I have done so far and generally, boiling points have been consistent with their theoretical value at my altitude. I find that using cheap low quality alcohol thermometers bring a much greater error and have since switched to using good quality mercury thermometers.

For example, the boiling point determined for diethyl ether during distillation with a cheap alcohol thermometer ranged from 30 to 38 C, while with a good quality mercury thermometer it stayed pretty much constant at 33 C (which is around the theoretical boiling point of diethyl ether at the atmospheric pressure here, ~700 mmHg) and then rose towards 35 C at the end of the distillation.