garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Drying Ammonia
I will soon be filling my 0,5L gas bottle with liquid anhydrous ammonia using my fridge compressor.
The ammonia will be generated by boiling 20% ammonia solution under reflux, the gaseous NH3 will come out at the top of the dimroth condenser.
It will be dried first and then stored without pressure in a 6L freezer bag from where it will be pressed into the bottle every time the bag is full.
The ammonia will liquefy at 9 bar/21°C inside the bottle.
1,4L of 20% ammonia solution will have to be degassed in order to get 0,45L liquid NH3 (280g, density of liquid NH3 at 21°C is 0,63 g/cm^3). I have
already acquired 3 L of 20% NH3. I also tried out if all the NH3 can be removed from the solution by boiling: I diluted the NH3 to 2% concentration
and boiled it, copious gaseous ammonia was evolved at such a high rate that some water in which I dissolved it got remarkably warm from the heat of
dissolution.
The compressor, the high- pressure line, manometer and adapter are ready and tested for tightness. The system has been tested with N2O and it was no
problem pressing 9 Liters of it into the bottle, producing 18 bar of pressure. The freezer- bag intermediate storage method is also very convenient.
Yet one problem remains: the question of the desiccant for the NH3. I am unable to acquire CaO which would normally be the drying agent of choice.
CaCl2 cannot be used since it forms an adduct with ammonia.
KOH was my initial choice, but it would be messy because the formed liquid would drip down and possibly clog the gas flow.
My drying agent of choice is now silica gel, I ordered 500g of it containing CoCl2 as an indicator (blue- pink).
I found no data about silica being a good desiccant for NH3- NH3 was neither listed under the incompatible gases nor under the compatible gases with
silica gel.
My question: can somebody reliably say if silica gel is a useable desiccant for gaseous ammonia? I do NOT want to waste my NH3 solution in order to
later find out that my liquid NH3 in the gas bottle contains a lot of moisture because the silica gel was unable to dry it.
It would also be unpleasant to find out that the ammonia is adsorbed by the silica gel.
Thanks for any help.
PS: my drying tube is made of PMMA. I hope that the ammonia won't attack it too much... the ester might be ammonolyzed forming
polymethacrylamide, releasing methanol vapor... I'll have to find out to what extent this reaction actually takes place.
[Edited on 23-10-2005 by garage chemist]
|
|
|
chemoleo
Biochemicus Energeticus
Posts: 3005
Registered: 23-7-2003
Location: England Germany
Member Is Offline
Mood: crystalline
|
|
When I played with NH3 gas to make NO2 over a Pt wire, I used a NaOH prilled column for drying. It worked very well. Silica - why shouldnt it work?
Comon, I bought CaO in G., it's in garden stores by the 20 kg sacks!
Alternatively, could you freeze out the H2O? I.e. by guiding the gasses through a -20 deg salt ice bath. Would need a large contact surface though,
and occasional emptying. Probably not worth the hassle.
Aside from this, don't you think it might be better to start of with cheap i.e. ammonium sulphate fertiliser, and dribble highly concentrated
NaOH into it?
Anyway, if/when (I hope  ) you succeed, make sure you post pictures! Including
one where Na or K is solvated!
Never Stop to Begin, and Never Begin to Stop...
Tolerance is good. But not with the intolerant! (Wilhelm Busch)
|
|
|
Dr. Beaker
Hazard to Others
Posts: 132
Registered: 9-9-2005
Location: between the med red and dead
Member Is Offline
Mood: Shaken, not stirred
|
|
Silica gel in considered an acidic medium.
it is used for chromatography, and such a basic ans polar molecule such as NH3 will create, in my opinion very strong chemical bond (H or even
covalent) with this medium (that's how salts, water and other highly polar comp.'s "stuck" at the top of the column) .
beside that - I have an idea for you. you can try seperate the ammonia by bonding it to various complexes.
for an example: Cu[NH3]4Cl2 (a buityfull purple color), that may be crystallized
without H2O and by heating decomp. to NH3 and CuCl2 for further use.
|
|
|
Magpie
lab constructor
Posts: 5939
Registered: 1-11-2003
Location: USA
Member Is Offline
Mood: Chemistry: the subtle science.
|
|
I also am going to make some liquid ammonia and place it in a steel cylinder - I'm just going to take a different approach by liquifying it
cryogenically. I have the equipment on order.
I will be generating ammonia gas as you are and also want it to be dry. So I read through the MS library looking for a procedure to do this. IIRC in
that huge book by Brauer it shows the use of a CaO column followed by a BaO column. But the BaO may only be needed for ultradry NH3 and might be
overkill for us.
I can get lime [Ca(OH)2)] cheaply at a garden store. I plan to fire this in my muffle furnace to reduce it to CaO. I don't know if you have
this luxury, however.
Powdered limestone (CaCO3) would also give you CaO if fired. But you know this.
The single most important condition for a successful synthesis is good mixing - Nicodem
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
In garden stores, Ca(OH)2 is available in 20kg sacks, but no CaO, I searched for it quite a bit.
Ca(OH)2 could be heated to form CaO, only about 600°C are necessary for complete dehydration in contrast to the needed 1200°C for CaCO3
decomposition. But for the needed larger amounts this would be a huge effort.
I can get 20% ammonia solution for 0,75€ per Liter. So the 0,45L liquid ammonia inside the gas bottle will cost me 1,05€ (plus cost for drying
agent). I know of no cheaper source! NaOH would be much more expensive when used with an ammonium fertilizer to create the NH3.
Pics will be made, for pics of the filling apparatus look in the "Filling gas flasks at home using a fridge compressor" thread.
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
Do you have some limestone then? (Mind that it isn't dolomite, the MgO won't do you any good.) Break it into 1-2" chunks, then pile
it on top of charcoal inside a pipe, or at worst a hole in the ground, then add forced air blast for an hour. Remove calcined lime, seperate from
fused ash.
Tim
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Today I went to a glassblower and made an order for a costom made drying column. Inner diameter 3cm, length 30cm.
Lower end closed, upper end open (to add and remove the desiccant, in operation it will be closed by a rubber stopper). Two olives at the top and
bottom.
I think 30 cm are long enough. What do you think?
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
How about aluminum amalgam , it should do the job . You could get some of that diamond mesh gutter guard material and roll it into a cylinder form to
fit your dessicant chamber , amalgamate it and put it in the chamber where it would happily react with any moisture and or oxygen that touches it . I
don't think it would react with the ammonia at ordinary temperature , but it just might , and that could be one potential problem .
Another problem is hydrogen which would be the byproduct added to the gas stream . The hydrogen wouldn't be liquified by pressure and would
gradually pressurize your receiver unless you had a small amount of bleedoff to relieve the pressure of accumulating hydrogen over the liquid ammonia
. Simply " burping " the receiver from time to time would probably do the trick , and you could see on a pressure gauge when it needed to
be done since the pressure over liquified ammonia would hold steady , and only something not liquifying would cause a steady increase in pressure
above that .
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
H2O + NH3 + Al = NH4Al(OH)4 + H2? (unbal)
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Aluminium amalgam is an interesting desiccant, probably best made by first etching an aluminium wire mesh first with NaOH solution until the entire
oxide layer is gone and then dipping into dilute HgCl2 solution until all is amalgamated.
However, shouldn't it react rapidly with the water in the HgCl2 solution?
Is the speed of the reaction with water really high enough for use as a desiccant?
I doubt it very much.
Also the gas would be contaminated by Hg vapor. From a chemical point of view this would be negligible (perhaps a few mg Hg in the entire gas bottle),
but from a toxicological point of view things are very different.
I don't want to make my ammonia "dirty" and have to worry about Hg contamination of substances prepared with it.
I'll do experiments with the silica gel first and also search for ways to turn my Ca(OH2) into CaO.
EDIT: I found a table of affinity of various substances to silica gel.
http://www.sorbentsystems.com/desiccants_types.html
Ammonia is right behind water, binding to silica even more strongly than alcohols.
It also says that CaO is very slow in absorbing moisture, despite its high affinity for water.
OK, back to the originally planned KOH. This will remove the water rapidly and completely.
Does anyone know the water vapor capacity of gaseous ammonia at 15°C? That would be nice, since it would enable me to estimate how much water has to
be removed in total.
[Edited on 24-10-2005 by garage chemist]
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
| Quote: | Originally posted by garage chemist
Aluminium amalgam is an interesting desiccant, probably best made by first etching an aluminium wire mesh first with NaOH solution until the entire
oxide layer is gone and then dipping into dilute HgCl2 solution until all is amalgamated. |
Dilute HCl solution of HgCl2 is more rapid .
| Quote: |
However, shouldn't it react rapidly with the water in the HgCl2 solution? |
Even more rapidly with dilute HCl , since the soluble AlCl3 is rapidly swept away from a clean mirrored surface whited by tiny hydrogen bubbles coming
off like smoke . The use of HCl lets you see what you are doing and leaves a clean rinsed surface when you withdraw the prepared material , more
active and not covered with insoluble hydroxide crud . The amalgam does react rapidly but is rate limited by the solution of new aluminum into the
mercury film . On a heavy sheet metal substrate , it won't just suddenly disappear when it is reacting , it will take awhile , long enough for
preparation and handling of the material without it disappearing on you while the work is in progress .
| Quote: |
Is the speed of the reaction with water really high enough for use as a desiccant? |
You betcha it is 
| Quote: |
I doubt it very much. |
Life is for learning
| Quote: |
Also the gas would be contaminated by Hg vapor. |
At a level so small as to defy detection .
Amalgams are very attached to the metal substrate at any normal temperatures , about as likely to float off into space as is
a neodymium magnet about to fall off a block of iron .
| Quote: |
From a chemical point of view this would be negligible (perhaps a few mg Hg in the entire gas bottle), |
more like picograms than milligrams ,
( if that much ) .
| Quote: |
but from a toxicological point of view things are very different.
I don't want to make my ammonia "dirty" and have to worry about Hg contamination of substances prepared with it.
|
Better just put up with a little mercury , it's everywhere just like gold , and lead and uranium and radium , you eat a little of it everyday
just a little ....not that much at all
And don't even think about the fillings in your teeth ! Or the fish you eat with them . Or the thimerosol preservative used in many medicenes ,
or antiseptics .
| Quote: |
I'll do experiments with the silica gel first and also search for ways to turn my Ca(OH2) into CaO. |
What I think you'll find is that ammonia has a chemical attraction for water that is so strong , it will be something like is the story for
hydrazine , and you will have trouble drying it with anything commonplace if there is any significant moisture in it like you will have distilling it
from aqueous solution . You will need an aggressive dessicant to dry it .
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
I once read that a concept for a hydrogen storage tank for a fuel cell car based on the reaction of Al amalgam with water had to be dismissed, only
because of Hg being present in dangerous amounts in the off- gas.
This is where I got the idea that Hg vapor will be given off in considerable amounts.
My aggressive desiccant will be KOH (mine is in the very convenient thin flake form, having a high surface area). It is able to dehydrate hydrazine
hydrate to the anhydrous substance, so it will be able to do the same with ammonia.
Again, do you have figures for the water vapor capacity of gaseous ammonia at 15°C?
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
What you read about the H2 generator for the car doesn't sound reasonable ,
and its easy to scrub mercury by forming the sulfide . All they would need to do is pipe the hydrogen through a pleated filter moistened with a
solution of a sulfide .
|
|
|
Magpie
lab constructor
Posts: 5939
Registered: 1-11-2003
Location: USA
Member Is Offline
Mood: Chemistry: the subtle science.
|
|
Perry's Chemical Engineers' handbook gives a lot of data on the partial pressures of H2O and NH3 over aqueous solutions of NH3.
Brauer says the column should be 0.5m long but he doesn't specify a diameter.
Today I made about 600g of CaO from my garden center hydrated lime. At 800C for 1 hr weight reduction was 28% against a theoretical 24%. I'm
assuming the extra 4% was due to moisture in the hydrated lime.
The CaO is a fine powder but does tend to form little BB sized balls if I roll the jar. I'm wondering how the ammonia is going to be forced up
through a column of powder if it is too fine. Perhaps it will have to be first applied to a substrate, like very small glass beads.
[Edited on 25-10-2005 by Magpie]
The single most important condition for a successful synthesis is good mixing - Nicodem
|
|
|
Magpie
lab constructor
Posts: 5939
Registered: 1-11-2003
Location: USA
Member Is Offline
Mood: Chemistry: the subtle science.
|
|
Garage Chemist your request for liquid vapor equilibrium data for the NH3-water system got me curious about my situation. So I swept out some cobwebs
in my head and took a crack at finding the answers. Here's what I came up with:
Assumptions:
P (total) = atmospheric = 739 mmHg
T (condenser) = 40F (4.4C)
Pot liquid mole fraction NH3 = 10%
Then from the tables in Perry's handbook:
At the pot:
T(boiling) = 158F (70C)
vapor mole fraction NH3 = 0.72
vapor mole fraction H2O = 0.28
At the condenser:
mole fraction NH3 in liquid = 42.2%
mole fraction H2O in liquid = 57.8%
mole fraction NH3 in vapor = 99.5%
mole fraction H2O in vapor = 0.5%
This means that for every 199 moles of ammonia leaving my condenser in the vapor stream there will be 1 mole of water.
If someone wants to check my calculations (or use of the tables) that would be good. 
[Edited on 26-10-2005 by Magpie]
The single most important condition for a successful synthesis is good mixing - Nicodem
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Thanks, Magpie! That's exactly what I needed.
I assume that the mole percentages of water and ammonia in the gas stream from the top of the condenser are the same no matter what percentage ammonia
solution is boiled inside the pot as long as the temperature of the cooling water is the same. Right?
If your calculations are correct it would be nice, since my 280g gaseous ammonia will contain only 1,55g water vapor according to your data. So no
excessive production of conc. KOH solution will occur in my drying column. Another worry is out of the way.
However, the temperature at top of condenser will have to be kept at 4,4°C in order to achieve this- lots of ice needed in the water reservoir for
the cooling circle of the dimroth.
BTW, how did you come up with 1 mol water per 191 mol ammonia when a mole percentage of 0,5 is given? Wouldn't it be more like 199 mol ammonia
and 1 mol water?
It seems like I'm missing something here...
|
|
|
Magpie
lab constructor
Posts: 5939
Registered: 1-11-2003
Location: USA
Member Is Offline
Mood: Chemistry: the subtle science.
|
|
Garage chemist: I think the anwer to both your questions is "yes."
I agree that the condenser vapor composition should be the same at a given temperature no matter the feed composition.
99.5/0.5 = 199/1, of course - this was a math error, poor lighting reading the calculator, & then not questioning the result as I knew it was
close. I will edit the above post - thanks for reviewing this.
The single most important condition for a successful synthesis is good mixing - Nicodem
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
