Pure Water

[DrPizza]

Originally posted by: EmMayEx
You don't measure the "resistance" of water you measure the "resistivity". Resistivity takes into account how much distance the electrons travel and how much cross sectional area they have to travel in. The units are Ohms * meters^2(cross sectional area) / meters(length). If you divide out the meters the final unit is Ohm-meter (not to be confused with an ohmmeter, which is the instrument used to measure "resistance").

For water you generally immerse two carbon probes (because carbon is fairly inert in aqueous solutions unlike metals which corrode rather rapidly in ultra pure water) and measure the current transmitted when a precise voltage is applied then apply a calibration factor to get Ohm-meters. You get the calibration factor by measuring some conductivity standards.

The conductivity of these is determined from first principles, i.e. if you add X milligrams of salt that dissociates with a dissociation constant of Y you end up with XY positive ions of mass Mp and XY negative ions of mass Mn. You can then use a theoretical equation that calculates conductivity based on ion mobilities as a function of ion concentration and temperature while taking into account the self dissociation of water into OH- and H3O+ ions and calculate a theoretical resistivity for your carefully prepared solution at a given temperature. You make a few of these solutions (or buy NIST certified standard solutions from VWR of Fisher Scientific) then measure them with your probe. With some amount of dilligence you get a straight line that correlates measured current to conductivity in Ohm-meters or Megaohm-meters. 15 Megaohm water means a cubic meter of in a cubic meter box with 1 meter square electrodes on either side has a resistance of 15 Megaohms (ignoring the resistance of the electrodes, wiring and instrumentation).

Max L.

Picky picky picky! Actually, you're not measuring the resistivity... you're calculating the resistivity. Since you're measuring V and I, a quick R=V/I will give you the resistance. ALSO you said you're putting in two carbon probes... so getting back to the question

When you test the resistance of water, how far apart do you put the probes?

How far apart?

 

[rjain]

You put them as far apart as the water. ...

 

[Shalmanese]

Originally posted by: EmMayExIf you divide out the meters the final unit is Ohm-meter (not to be confused with an ohmmeter, which is the instrument used to measure "resistance").

Max L.

bwhahaha, that made my day

 

[wacki]

Originally posted by: NeoPTLD

Originally posted by: miguel
I've read that pure water will not boil due to the lack of impurities, but I've also read that pure water will boil, but a higher temperature. What's the deal with this?

Also, except in a laboratory setting, can you have such a thing as pure water? Can you actually buy pure water nowadays?

Depends on how you define pure. 100% water can't be made. We can get close with 99.9xxxxx%.
You can buy deionized water and distilled water at supermarket or buy some expensive HPLC grade water.

I think you're mixing it up with water in a very smooth container. When you microwave such thing, there is no core for water to start boiling and it becomes superheated. It will boil abruptly when it is disturbed as it is taken out.

I'm not trying to be an ass, but I do believe 100% pure water can be made by using vacuum scanning tunneling microscopy or similar technology. Granted you would have to pay about a dollar an atom tho, maybe more. Not the best solution for pure water, but it is a solution.

Then I guess you could make the arguement that dueterium based water wouldn't be filtered out... if heavy water isn't considered a part of pure water.

 

[EmMayEx]

>Picky picky picky! Actually, you're not measuring the resistivity... you're calculating the resistivity. Since you're measuring V and I, a quick R=V/I will give you the >resistance. ALSO you said you're putting in two carbon probes... so getting back to the question

Touche'. You don't "measure" resistivity you calculate it from measured current, voltage, and a calculated calibration factor.


> Quote
> When you test the resistance of water, how far apart do you put the probes?
>
>How far apart?

The probes I've used have epoxy potted graphite rods that are about 5 mm in diameter with about 10 mm protruding into the liquid. The rods are about 10 mm apart. There is a sleeve that covers half of the outer portions of the graphite rods so they can't be easily broken and possibly to focus most of the current between the probs so they are more linear (I didn't design them I just used them). If you screw the probe into a "T" fitting and use the other two legs as an elbow you can get water to flow through the slot and give a good reading.

It's hard to get good consistent readings on stagnant water. The biggest problem is that water sitting in open air will absorb some of the the 350 ppm of CO2 in the air raising the conductivity of the surface layer quite dramatically. This gradually diffuses down into the liquid making conductivity a function of depth and time if you don't stir it or a function of time if you do stir the water. That's why you measure ultra high purity water conductivity in a closed flowing system. The PTFE piping used in the semiconductor industry is permeable to CO2 so you have to recirculate the water through the purification system and let everything come to a steady state to get a good reading. Quartz piping can be very pure and is essentially impermeable to gas. Unfortunately hydrofluoric acid (HF) dissolves quartz and that is often what you are trying to rinse off in a semconductor wet process rinse tank (which is where most of the high purity water is used in a semiconductor fab).

When you talk about purity you can only approach the 100% absolute unless you look at very small clumps of atoms in a vacuum chamber with a spectrometer. It's not too difficult (these days) to produce a beam of 100% chemically and isotopically pure atoms but the mass generated is on the order of nanograms. Silicon for semiconductors is about the purest material made on anything resembling a large scale. The impurities are measured in parts per 10^15 (parts per quadrillion) meaning only one atom in a quadrillion is not silicon. Since it is purified by distillation of SiF6 (which is gas a room temperature) and the difference in mass of the silicon isotopes is considerable (unlike Uranium) it is generally isotopically pure as well as chemically pure. Still, as soon as you take it out of an HF bath or a vacuum chamber it starts to oxidize on the surface. This takes only a few seconds so even pure silicon wafers are clad in a very thinn (10 - 20 angstrom) coating of quartz (SiO2).

Water is even more difficult to purify as it corrodes nearly all metals to some extent and dissolves most gases to some extent. Even deoxygenating it is challenging since one of the components of water is oxygen and ambient radiation or contact with catalytic radicals is capable of splitting water into it's component parts. So about the only truely pure water you can have is a beam of ions in a mass spectrometer.

Max L.