Sciencemadness Discussion Board - Table salt heating bath?

Table salt heating bath?

Romain - 29-12-2015 at 06:54

Hi,

I want a way to heat a medium sized round bottom flask (100-500 ml) for a future dry distillation of acetic acid from a mix of NaHSO₄ and CH₃COONa. This reactions involves rather high temperatures (I don't know precisely how high, maybe around 200 or 300°C) I already own a hotplate stirrer and I don't want to spend money on a heating mantle so I considered using heating baths.

Apparently people have used sand, oils (cooking, silicone, mineral), paraffin wax, steel or copper shot, molten metals, molten salts, ... But I never heard of a table salt bath (in powder form, not a solution).

Table salt looks to me like a good substitute for a sand bath: It's cheap, doesn't fume like oil does and you don't need to buy a 50 kg bag of it... And as a bonus, the Mohs hardness of sodium chloride is around 2 or 3, that of sand is around 6 or 7 and that of borosilicate glass is also around 6 or 7, which means glass won't get scratched by the salt bath whereas it might get scratched by a sand bath (I haven't tested).

So has anyone any experience with table salt baths? Are they ever used?

P.S. I did run a search before posting but didn't find anything relating to table salt baths.

[Edited on 29-12-2015 by Romain]

Heavy Walter - 29-12-2015 at 09:35

Hi
My limited experience with salt is that it tends to solidify, making difficult to remove a flask.

Romain - 29-12-2015 at 10:34

Thanks for your input.

Did you heat enough to melt the sodium chloride (ca. 800°C IIRC)? Why do you suspect it solidified? Was it because of impurities like anti-caking agents, etc.? Was the salt slightly humid, causing it to dissolve in that small amount of water and crystallize around the flask?

I can't test for myself now because I don't have access to my lab unfortunately.

Heavy Walter - 29-12-2015 at 10:51

Hi

Just heating at ca. 400 °C.
And in very humid ambient. I used coarse salt and crystals tended to fuse between themselves.

careysub - 29-12-2015 at 11:31

How about calcium carbonate sand? This is widely available, many "play sands" are not silica sand. Much softer than silica.

ariep641514 - 29-12-2015 at 21:36

ordinary salt of chloride seems to be reactive at high temp. it promote rust, if you not clean after use, residue remain in your tools will absorb water , and corode your heating element, or ferous part of you equipment.

Pyro - 29-12-2015 at 21:51

I have used table salt baths often and found them to be very practical

Romain - 30-12-2015 at 09:53

Ok thanks for your replies! I'll be sure to test it when I can.

subsecret - 30-12-2015 at 19:52

But the salt can be easily washed off with water if it sticks. For a really good bath, I'd recommend copper powder, though I've never tried it myself. It'd be like a moldable heating block.

Romain - 31-12-2015 at 04:03

I've thought about copper powder bath too but I don't have any copper powder. I do have about 1.5kg worth of 400 mm^2 stranded copper cable though, so if I manage to cut it in small pieces, it could work for a heating bath.

I just ran a quick calculation: 1 kg of copper powder in the form of small spheres, when packed, takes up about 170 cm^3. (Density of copper: 9 g/cm^3; when packed (bulk density): 64%*9 = 5.75 g/cm^3). This means it's going to be quite costly to have a big enough copper powder bath. But I'll give it a try if I ever get around to chipping that copper cable.

careysub - 31-12-2015 at 09:59

Quote: Originally posted by Romain

I've thought about copper powder bath too but I don't have any copper powder. I do have about 1.5kg worth of 400 mm^2 stranded copper cable though, so if I manage to cut it in small pieces, it could work for a heating bath.

I just ran a quick calculation: 1 kg of copper powder in the form of small spheres, when packed, takes up about 170 cm^3. (Density of copper: 9 g/cm^3; when packed (bulk density): 64%*9 = 5.75 g/cm^3). This means it's going to be quite costly to have a big enough copper powder bath. But I'll give it a try if I ever get around to chipping that copper cable.

Try aluminum instead. Much lighter (1/3 the density) and much cheaper. It has half the thermal conductivity of copper, but that is still very high - the thermal conductivity of aluminum is 70 times higher than quartz.

Cut aluminum wire pieces are available for cheap on eBay.

Romain - 2-1-2016 at 12:35

Aluminium is a good idea! I couldn't find the cut wire pieces you were talking about on eBay, do you have any specific keywords for a search?

I could also just melt some scrap aluminium and make shot by pouring the molten metal in a bucket of water, but its more troublesome than buying it, considering I don't have a furnace.

DraconicAcid - 2-1-2016 at 12:40

Quote: Originally posted by Romain

Aluminium is a good idea! I couldn't find the cut wire pieces you were talking about on eBay, do you have any specific keywords for a search?

I could also just melt some scrap aluminium and make shot by pouring the molten metal in a bucket of water, but its more troublesome than buying it, considering I don't have a furnace.

Molten aluminum in water??? Sounds like a very bad idea, at least based on the warnings I got about using water on an aluminum fire at the Norsk Hydro plant.

ETA: https://www.youtube.com/watch?v=29O-8LckmiU

Apparently not so much.

[Edited on 2-1-2016 by DraconicAcid]

blogfast25 - 2-1-2016 at 12:44

Quote: Originally posted by DraconicAcid

Nah. Not a problem at all. Been there, done that.

BromicAcid - 2-1-2016 at 13:17

Thermal conductivity of sodium chloride at 289K is 6.5 W/(mK). Silicon dioxide is 1.3-1.5 W/(mK). For comparison, copper is 385.0. But in each of these cases (table salt, sand, copper shot) you also have void space between the particles, this reduces the thermal conductivity even further. What I am getting at here is that even through you might see sand in a literature reference, a lot of hot plates that still have their safety information recommend against it because it is such a good insulator that you can overheat your coil. That being said, at least sodium chloride is a bit more conductive than sand but compared to copper the difference is negligible. Simply said, the same advice on a sand bath (be careful of using too high of a setting or heating too quickly) will apply to your salt 'bath'.

You could also compare specific heat to give yourself a better idea of possible performance differences.

Interestingly enough, all three of these are better strictly speaking than a mineral oil bath which only has a thermal conductivity of 0.162 W/(mK). Off the top of my head I would say however that because liquids can undergo convection movement and can stir you will end up with much better heat transfer than relying on thermal conductivity alone.

I didn't see it mentioned above so I thought I would mention it here.

JJay - 2-1-2016 at 13:59

I've used sand baths on a hot plate before, and almost every time I've tried it, the glass has broken due to thermal stress. I definitely don't recommend it.

careysub - 2-1-2016 at 15:19

Quote: Originally posted by Romain

Yes, the are calling them "aluminum pellets" and the vendor is "handi-ramp510".

They sell them with two different price+shipping options, a 10 lb box with free shipping for $100 (thus $10/lb); OR you can use a different listing they have that sells them for $5.00 lb, after $12.50 for the first lb. This second one may not come up on a search because this listing does not have the word "aluminum" in the title.

careysub - 2-1-2016 at 15:39

Quote: Originally posted by BromicAcid

Interestingly enough, all three of these are better strictly speaking than a mineral oil bath which only has a thermal conductivity of 0.162 W/(mK). Off the top of my head I would say however that because liquids can undergo convection movement and can stir you will end up with much better heat transfer than relying on thermal conductivity alone.

Convection will transfer heat orders of magnitude faster than conduction.

The thermal conductivity of the fluid is uninformative, the efficiency of convection depends on the "heat transfer coefficient" which is actually a combination of conductivity, viscosity, heat capacity, and the geometry of the situation.

A fundamental problem with any sort of powder bath is that conduction slows down dramatically with increasing thickness of the insulator (lets face it, it is better description than "conductor" for sand), and necessarily creates a large temperature gradient (did I hear your flask breaking?). I think sand baths are likely only really good for microscale work (thinner layers are better).

Convective baths always keep the bath temperature nearly uniform (which is why Thiele tubes work).

Romain - 14-1-2016 at 12:27

Ok, I tried the salt bath and it doesn't work very well.

I tried distilling water from a 250 ml RBF and the hotplate temperature was set to 200°C. The bath consisted of a 115 mm glass crystallizing dish full of fine salt. The bottom of the flask was about 10 mm above the bottom of the dish and was surrounded with salt. After 1 hour, the bath hardly reached 100°C at the surface and was at more than 300°C at the bottom!
This means that the hotplate can't regulate it's temperature properly with a salt bath on top of it. It can get much hotter that it is supposed to and this could destroy the heating element.

The heating is also quite inefficient: it took around 1 hour of heating to start boiling 150 ml of water.

Basically, the salt bath is a bad idea: it is slow and could destroy your equipment.

However, I discovered that salt is pretty useful to clean flasks with gunk in them. I tried distilling nitrocellulose lacquer because it contained toluene after the distillation was complete, the nitrocellulose started precipitating inside the flask. after some manual scrubbing I was left with some nitrocellulose residue inside the 500 ml Erlenmeyer flask I used and could not reach inside with a brush. So I poured in around 100 ml of ethanol and some coarse salt, stoppered the flask and shook for a while. The salt scrubbed the residue and my flask was perfectly clean after that!

I hope these results will be useful to someone.

[Edited on 15-1-2016 by Romain]

stoichiometric_steve - 19-1-2016 at 14:42

Quote: Originally posted by Romain

The salt scrubbed the residue and my flask was perfectly clean after that!

That's actually quite a nice idea, thank you!

XeonTheMGPony - 20-1-2016 at 10:39

http://www.ebay.com/itm/10-LBS-COPPER-FINE-POWDER-99-purity-...

Was looking at copper powder as an idea and found it is pretty cheap.

Oil baths are messy but keep very uniform heat, especially if stirred, Other ones I have used is brine baths (Water super saturated with salt).

I have made super cheap mantles by molding plaster of parris with rope wrapped around the flask, then modified broken hair driers for the NiChrom coils with good success.

To expand the coil to meet the voltages you'll be using heat with a torch till red then stretch it out. I used mostly 12V as the heating current and a salt water rheostat.

careysub - 20-1-2016 at 11:25

Thanks, that is pretty informative. Good ideas.

In evaluating a copper dust bath (325 mesh is pretty fine) vs an aluminum shot (or cut wire) bath consider that the cheap copper dust here is $12 a pound with a density of around 6 (using the fact that fine sand is about 75% the density of quartz). This is $158/liter.

Aluminum shot can be had on eBay for ~$6/lb, and has a density of around 1.73 (2.71 * 0.64 sphere random packing efficiency). This translates to $23/liter. The packing efficiency for the cut wire pieces is probably a bit lower than spheres, so this may be a trifle high.

And the copper bath will be very heavy.

Can you post a picture of one of your plaster mantles?

Sulaiman - 20-1-2016 at 12:46

I had considered using plaster of paris to make a heating mantle (using nichrome wire)
but when I read the Wikipedia article https://en.wikipedia.org/wiki/Plaster
I thought that it would crumble when the anhydrite is formed above 130C

What maximum temperature have you run plaster of paris at ?

P.S. I happen to have copper powder from galvanic cell experiments and I have been keeping it in an airtight bottle
... would it not corrode in air, especially when heated, forming thermally insulating surfaces ?
... being fairly dense, I imagine VERY buoyant flasks

[Edited on 20-1-2016 by Sulaiman]

careysub - 20-1-2016 at 14:26

Plaster of paris is commonly used in metal casting molds, and moderate temperature furnaces - but you don't use pure plaster.

Typically 50% fine sand is added, you can also use perlite in whole or part for a lighter, more insulating product. Incorporating glass fiber (as loose wool, chopped fiber, or cloth) can make it even more durable. And the plaster is usually baked in an oven (450-500 F) for a few hours to dry after it is fully cured (3 days).

Perlite and furnace cement can be used for something that is really high temperature (far more than is needed in this application).

And I imagine getting a cheap metal bowl and casting the plaster or cement mix into it so that the bowl is a permanent part of heating mantle will ensure no crumbling or break-up problems ever occur.

XeonTheMGPony - 28-1-2016 at 06:51

Quote: Originally posted by Sulaiman

I had considered using plaster of paris to make a heating mantle (using nichrome wire)
but when I read the Wikipedia article https://en.wikipedia.org/wiki/Plaster
I thought that it would crumble when the anhydrite is formed above 130C

What maximum temperature have you run plaster of paris at ?

P.S. I happen to have copper powder from galvanic cell experiments and I have been keeping it in an airtight bottle
... would it not corrode in air, especially when heated, forming thermally insulating surfaces ?
... being fairly dense, I imagine VERY buoyant flasks

[Edited on 20-1-2016 by Sulaiman]

Yes they disintegrate, basically only used for the very rare occasion I need high temps, the NiChrom is stored in hard card board tubes. just a very fast cheap methode when you don't need it enough to pay high dollar. they do hold up for more then one flask though.

Andabove post states how one could make it more perminant

[Edited on 28-1-2016 by XeonTheMGPony]

chemrox - 28-1-2016 at 15:17

Wouldn't Al powder be rather toxic? Especially heated up? I recall that early Alzheimer syndromes were attributed to people smoking drugs with Al pipes.

careysub - 28-1-2016 at 17:11

Quote: Originally posted by chemrox

Wouldn't Al powder be rather toxic? Especially heated up? I recall that early Alzheimer syndromes were attributed to people smoking drugs with Al pipes.

The old proposed aluminum-Alzheimer's link has no support in current scientific literature.

Aluminum is not a volatile metal, unlike zinc and cadmium.*

If aluminum were really toxic, the widespread use of aluminum cookware, and use of foil for food packaging would be extremely problematic. The use of aluminum hydroxide in medications ditto.

*Cadmium reaches 0.1% Atm Pressure at 380C, it reaches 1% at 470 C.
For zinc this is 477 C and 579 C.
For aluminum this is 1544 C and 1781 C.

JJay - 1-2-2017 at 14:51

I tried a table salt bath yesterday and found that they work pretty well. I didn't manage to reach a temperature above 285 C, and it took a while to get there, but the heat source was only 600 watts, and I made no special effort to insulate. It's definitely less messy than an oil bath.

Bezaleel - 9-2-2017 at 14:02

What comes to my mind is casting a cylinder of aluminium and cutting out a spherical hole on a lathe. Hole size close to the size of your RB flask. I've done such casting with zinc molten in an old steel pan over a Teclu (Bunsen) burner. Aluminium would just require a lot more heat to get it molten, but has a much lower heat capacity (specific heat, to use the correct physical term), so would heat up much quicker in use.

I have an IKA magnetic stirrer/hotplate, and using a thin steel cup filled with lamp oil, it took about 45 minutes before 100C oil temperature was finally reached (the stirrer temperature sensor was put in the oil). I guess putting it at full power and switching to sensor control when the water in the RBF starts to boil, will be a whole lot quicker.

JJay - 9-2-2017 at 14:21

A few companies sell those under the designation "heating blocks." They are expensive.

Lamp oil, huh? That seems like it would be dangerous... when I use oil, I usually use soybean oil, which, while flammable, has a flash point above 300 C.

PirateDocBrown - 9-2-2017 at 19:35

This fellow made plaster heating mantles:

https://www.youtube.com/watch?v=QKFC0ke_DOU&t=315s

I have considered this route, especially for a large one.

JJay - 9-2-2017 at 21:02

I have tried making heating mantles and found it to be more trouble than it's worth, but I never got around to making one with clay....

I have one but have only used it once, and I've never liked them very much, mainly because you need one for each size flask you plan to use. They are more efficient than a hotplate in terms of watts required to achieve a given temperature for a given period of time, though. Temperature control with one is actually not bad, but I'm sure an oil bath gives you more precise control.

Bezaleel - 10-2-2017 at 04:40

Quote: Originally posted by JJay

A few companies sell those under the designation "heating blocks." They are expensive.

Good to know, thanks!

Quote: Originally posted by JJay

Lamp oil, huh? That seems like it would be dangerous... when I use oil, I usually use soybean oil, which, while flammable, has a flash point above 300 C/

When heated to around 100C ignition won't be a big problem. The composition is alkanes in the range C7-C20, according to the label. I should distill of the low boiling components, but didn't get to it yet.