Sciencemadness Discussion Board
Not logged in [Login - Register]
Go To Bottom

Printable Version  
 Pages:  1  ..  49    51    53  ..  55
Author: Subject: Preparation of elemental phosphorus
BluePlanet1
Harmless
*




Posts: 41
Registered: 30-1-2016
Member Is Offline


[*] posted on 27-4-2016 at 19:12


Quote: Originally posted by OldPhart  
You can now get micro drill bits at Harbor Freight! $3.99 for an assortment of 30. I have not yet tried them but the reviews seem to be more positive than negative.

http://www.harborfreight.com/high-speed-steel-micro-drill-bi...


Awesome find thanks for the link! I broke all my really small bits and had to settle for a #74 but I'm gonna pick up like 3 sets tomorrow.

As of now the burner & forge are basically finished. It took like a week just to dry. The orfice technically should be smaller at #78. And it's *slightly off center which leads to fluttering & backfire at low-med pressures. But when I turn it up all the way the fire stays 95% in the forge with minimal backfiring.

I've also done melt tests and it melts aluminum, copper & brass with ease.
More interestingly I swapped my propane for MAPP then stuck a big iron nail in. After a couple minutes the tip melted off so I bended it in half. Then stuck it back in and saw iron sparks blowing out. So I pulled it out, hammered it down and it fused like a legitimate forge weld. MAPP gas is expensive and only suppose to burn 100C hotter than propane... with small torches that is. But in the forge it seems to burn much hotter.

Anyway, all things are set to fire off. For now I'll have to do the larger 300gm one. Since my options for the forge are either "off" or "fullblast." That is until I size down and recenter my orfice.

Only thing holding me back now is courage. It all looks very neat, compact and sophisticated. But when I run my mind through the logic of what I'm actually doing... I'm setting off a controlled explosion within a controlled explosion.

I'm definitely not jumping into this full scale on my first run. Maybe I'll start with 50-100gms first. Or maybe I'll just try to drill a new orfice tomorrow. The worst I can see happening is the retort gets yellow-white in the forge....the reaction fires off driving it to a whiter yellow.... and I can't lower the heat (can only shut it off).. then I choose to keep it on cause I'm stupid and think the steel will hold up... meanwhile the retort melts launching molten white P4 and slag all over my lawn. Either that or the retort explodes right inside the forge blowing the burner off as it swings in circles like a flame throwing snake burning me & the whole town to ash.

F##k that. I'm doing baby steps first. Then I'll work my way up in increments of 50gms as I work on the orfice between runs. Nothing larger than 150gms till that orfice is perfectly centered.

To conclude, I just gotta grow some balls ideally by Friday. I'll set everything up, hopefully report back and have a video by Saturday. Just so the forum realizes though, this is all dependent on my mojo. If I disappear till next Monday I will come back with some lame excuses. But I DO PROMISE to get this whole process (trials 2, 3, etc) started by next week at the LATEST.

So Friday if mojo is good. Early next week if I somehow get smarter by Fri. It might be better to say I'm just waiting for a sign from the phosphorus gods. :D
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5405
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: pumped

[*] posted on 28-4-2016 at 09:50


Your apparatus should as far as possible be failsafe. Give the reaction products plenty of opportunity to vent safely when the reaction takes off.

Gruson used an oxy-acetylene torch to heat his retort. Take a look upthread.




The single most important condition for a successful synthesis is good mixing - Nicodem
View user's profile View All Posts By User
SilverMiner20ozTons
Harmless
*




Posts: 1
Registered: 1-5-2016
Member Is Offline


[*] posted on 1-5-2016 at 04:00
That's a lot of work


I remember just scrapping it off the back of matchboxes soaked in acetone? That was a joke only some members did.
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 11-6-2016 at 07:40
Furnace Concept


On my way to work (I have a long commute) I like to design things in my head, and got to thinking about how a phosphorus producing furnace might be built and operated. I assume that this reaction works better at larger scale, so the focus is on how a "max scale" system within the constraints of a residential set-up might be built at a reasonable cost.

An arc furnace, like those used commercially, seems compelling to me since it easily gets the very high temperature required, and with internal heating is efficient and spares the reaction vessel from having to take the full heat.

The basic idea is this: use a graphite crucible for the reaction vessel, with a vertical rod (or rods) striking the arc at or close to the bottom (using a sacrificial graphite disk, not the crucible itself). The crucible would be in an ordinary steel drum, either 5 gallon of 10 gallon depending in the size of the crucible. The drum is used to maintain the gas seal, keeping oxygen out, and efficient containment and collection of the phosphorus vapor. Carbon dioxide flushes the can before the arc is struck, and a flow of CO2 is maintained during operation.

Some points about this:
Graphite crucibles are available to pretty large sizes at reasonable cost and are rated to 1510 C:
http://www.budgetcastingsupply.com/Crucible-Clay-Graphite-Bi...
The crucible would rest on a refractory insulating support so that the drum is not subjected to the intense heating.
The drum has an resistance heater inside of it for preheating to 300 C, and is well insulated. It may be that simply using the arc system in a "low power" mode to preheat the unit would all that would be needed.
The entire vapor capture system is insulated and heated (except where the product is collected) using heating tape, etc.
The entire apparatus - drum and collection system is preheated to 300 C before the arc is struck, so that there is no possibility of phosphorus condensation except in the collector.
Silicone gasket RTV is readily available rated to 343 C, and up to 400 C can be had. This would be used to seal the drum lid before operation.
I would put in some sort of odorant that volatilizes as this heating takes place to ensure the entire system is sealed, phosphorus vapor being quite toxic.

The crucible/arc could be set up and operated in one of two modes:
1. uninsulated crucible, with the arc designed to reach a temperature well above 1500 C inside the reaction mass, and using a localized reaction zone, with slag collecting at the bottom as the reaction proceeded.
2. insulated crucible with the objective of heating the entire crucible to 1500 C so that the entire loading would be evenly heated.*
In the first case you would use a max size crucible (like the one I linked to above), in the second a smaller crucible would probably be best.

Modern home circuits can support up to 12 KW (100 amps, 120 V) so that would be the max power for heating, and would inform the correct scaling of the crucible size used.

(Initial flushing of the system could be done conveniently be placing some dry ice in the bottom before sealing. The cold dense gas would displace and flush the air efficiently and cheaply. You have time to kill while the RTV sets.)

Any comments about this concept? Problems? Improvements?

*Insulation scheme, use a 1670 C rated castable for the crucible holder shell, with a light-weight insulating castable (rated to 1450 C) as an outer jacket. High temperature insulating wool or ceramic fiber paper fills the rest of the drum.

[Edited on 11-6-2016 by careysub]
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5405
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: pumped

[*] posted on 11-6-2016 at 09:13


I think your basic idea is sound. As you say the challenge is to develop a safe and economical device.

Page 47 of this thread shows an arc furnace design by Landis that would work for medium scale (back-yard) production. But it would be too expensive for occasional production with no sales to support the capital and operating costs.

I'm skeptical of using RTV for a sealant. I've used it before for P production with less than satisfactory results.

I'm a proponent of compression seals. Refractory braided gaskets and rope are cheaply available at barbeque stores. Large drums (33, 55 gal) have lever-operated ring closures available for sealing drums. These may be available for smaller drums also. I don't think they are intended to be liquid tight but with proper gaskets might be made gas-tight for low pressures like you anticipate.

The Landis design uses a gland seal for the consumable rod that is pushed into the arc. I would use a braided refractory rope for that also. Aluminum plant carbon electrodes employ a screw to maintain a seal. I think the commercial P plants do the same.

It is interesting that you intend to operate this device at a slightly positive pressure rather than a slight vacuum. A slight vacuum would help with containment issues.

What would be your batch production size?

[Edited on 11-6-2016 by Magpie]




The single most important condition for a successful synthesis is good mixing - Nicodem
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 11-6-2016 at 11:29


Thanks, that's the sort of feedback I am looking for. I'll look at other sealing strategies. I view that as the most troublesome part of this scheme.

My thought about positive vs negative pressure is that I want the seals to be tight, so they should not be leaky (the barrel lid seal is the tough part), but I want to move the product efficiently out of the system for collection. Positive pressure (from a CO2 tank) is easy to supply.

The batch size for the "max" concept is whatever mass of reactant could be accommodated with 12 KW of heating, I have not tried to estimate that yet (anyone able to through out a plausible guesstimate?). There may be metric available about arc furnaces with rules of thumb about power densities such as KW/L or KW/kg.

Graphite crucibles are available in large sizes, the one I linked to could hold 20 kg of reactant (with a packing density of 1), which I am sure is more than can be used, so that is not a limiting factor.

This is (at this point) just a conceptual design exercise.

Once I get a good handle on a "max" system, I'll think about how it scales down. For example an inexpensive Harbor Freight arc welder which I could probably get discounted for $75 is a 70 amp unit, which would be ~8 KW, scaling down the power by a third. A suitably smaller crucible would fit in a 5 gallon pail I think, etc.

Since this is intended for batch, and not continuous operation (unlike the Landis design) many aspects can be simplified - I was not intending to have any carbon rod feed other than gravity perhaps (a holder at the top could provide mass to assist feed). It would be useful to know the dimensions and power level of the Landis system as a scaling guide.

[Edited on 11-6-2016 by careysub]
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 11-6-2016 at 11:43


10 gallon drum with a bolt ring enclosure looks promising. Even with a modest crucible, the greater volume of the 10 gallon might be valuable to provide insulating and working space.

"The Electric Arc Furnace" (1914):

https://ia800204.us.archive.org/19/items/electricfurnace00st...

looks like it has all the formulae and data (supplemented by actual refractory properties) to do design calculations. I'll get back to you all when I have done some.

[Edited on 11-6-2016 by careysub]
View user's profile View All Posts By User
DFliyerz
Hazard to Others
***




Posts: 236
Registered: 22-12-2014
Member Is Offline


[*] posted on 11-6-2016 at 12:05


While buying some steel pipe to build an apparatus to try this with trisodium phosphate, I found a large bag of inexpensive "triple superphosphate" (CaH4P2O8) which I want to try as well. However, I can't find any good sources as to the thermal decomposition products of this phosphate. Anyone have any ideas?
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5405
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: pumped

[*] posted on 11-6-2016 at 12:47


@careysub: Yes, I think the 5-gallon size would be big enough. Minimizing the internal volume would facilitate start-up and shutdown.

I don't know how fast the carbon electrode would be consumed. I suspect the arc length is critical so you may need a feeding mechanism. Manual seems OK if you monitor the amps and push in more rod if amperage drops.

When you finish a run and pop the lid the whole thing may catch on fire. I don't know how effective the CO2 purge would be. Like you say, if everything is kept hot it seems the CO2 purge would work.

I wouldn't want to push a 100a service to the limit. I would shoot for something less, say 75a or so.

I assume you are going to want to reuse the carbon crucible. Is the CaSiO3 plug going to be easily removable? It seems that a good tap should allow for it to drop out. Which brings up a design problem: will not the liquid CaSiO3 form an insulating layer on the bottom of the crucible? I think P furnaces form the arc between vertical carbon electrodes.

The more I think about the positive pressure aspect the better I like it. Any sucking in of air would result in burning P. N2 and Ar would also work as purge gases but may be more expensive than CO2.

[Edited on 11-6-2016 by Magpie]

[Edited on 11-6-2016 by Magpie]

[Edited on 11-6-2016 by Magpie]




The single most important condition for a successful synthesis is good mixing - Nicodem
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 11-6-2016 at 15:09


Quote: Originally posted by Magpie  
@careysub: Yes, I think the 5-gallon size would be big enough. Minimizing the internal volume would facilitate start-up and shutdown.


If I can find one with the hoop closure. So far the smallest I have seen it in is 10 gallon.

Quote:
I don't know how fast the carbon electrode would be consumed. I suspect the arc length is critical so you may need a feeding mechanism. Manual seems OK if you monitor the amps and push in more rod if amperage drops.


If an arc is used then yes, this needs to be studied. This is an area I have not investigated in depth yet.

But it may not be an arc (I should have mentioned this in my original post). If the whole crucible heating method is chosen resistance heating with a continuous resistance element would be used which would not be consumed (at least, not in the same manner, some erosion would probably occur I expect).

Quote:

When you finish a run and pop the lid the whole thing may catch on fire. I don't know how effective the CO2 purge would be. Like you say, if everything is kept hot it seems the CO2 purge would work.


Right. This was part of my thinking. Perhaps gently increasing air admixture to oxidize remaining product before opening the container?

Quote:
]
I wouldn't want to push a 100a service to the limit. I would shoot for something less, say 75a or so.


I'd probably go with the 70A arc welder in reality.

Quote:

I assume you are going to want to reuse the carbon crucible. Is the CaSiO3 plug going to be easily removable? It seems that a good tap should allow for it to drop out.


Don't know. Perhaps pressing a layer of loose carbon material in the bottom to create prevent bonding? But yes, drilling a tap in the bottom to drain into a slag receptacle would be a good idea.


Quote:

Which brings up a design problem: will not the liquid CaSiO3 form an insulating layer on the bottom of the crucible? I think P furnaces form the arc between vertical carbon electrodes.


That is one thing I already thought about. Solution: that sacrificial graphite disk. It sits above the bottom with a drain and it makes electrical connection with the sides of the crucible. The slag drains to the bottom. This would be used even with the resistance heating option.

Quote:

The more I think about the positive pressure aspect the better I like it. Any sucking in of air would result in burning P. N2 and Ar would also work as purge gases but may be more expensive than CO2.


Yes, the idea of filling the system with an odorant during the pre-run prep allows you to detect leaks. Given the small pressure differential tight sealing should be possible. Welding, brazing components as needed. Tight pipe fittings. The drum hoop seal could still be backed with silicone compound if necessary.

My strategy is, if it leaks, it is not ready to run.

Since you are oxidizing carbon, and putting out CO exhaust, I figure CO2 would be fine as a cover gas. Cheap, convenient.
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5405
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: pumped

[*] posted on 11-6-2016 at 16:05


Quote: Originally posted by Magpie  

I assume you are going to want to reuse the carbon crucible. Is the CaSiO3 plug going to be easily removable? It seems that a good tap should allow for it to drop out.


I should have said good "rap." I think a drain tap might be overkill for this scale.

Making commercial P is a "submerged arc" process. The electric current is surrounded by the powder charge. Wiki explains this well here.. If you try to use the bottom of the crucible as one electrode it seems that it would get insulated by the slag. But carbon may well be the best crucible material due to its resistance to high temperature.

Seattle Pottery is also a good source of refractory materials as you may already know.

Here's a 5-gallon drum w/ring closure.


[Edited on 12-6-2016 by Magpie]

[Edited on 12-6-2016 by Magpie]




The single most important condition for a successful synthesis is good mixing - Nicodem
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 11-6-2016 at 20:34


The concept continues to evolve.

If the "whole crucible" heating approach is taken then one wants to insulate it extremely well, and minimize the thermal mass to be heated.

The Kastolite 26 insulating refractory is rated to 2600 F (reflected in its name, or 1440 C) so if one holds the reaction temperature to 1440 C (which seems fine from the literature I consulted) then this can be used to create a "super crucible" to hold the graphite one. It needs only be thick enough to bring the surface temperature down to 1100 C where vermiculite can take over.

So: I envision a 5 gal steel pail than is entirely filled with insulation - the Kastolite crucible holder, loose vermiculite fill on the sides, hard vermiculite board underneath, and also over the loose fill, where it is sealed air tight with refractory cement.

The only empty volume in the pail is the headspace above the vermiculite board and the pail lid, which could be kept quite minimal. Escaping heat from the crucible would keep the headspace and ports above 300 C (I imagine that a good internal insulation lid over the crucible would be needed to keep the temperature down to something so modest).

The lid would have the CO2 inlet and outlet ports, which would get professionally welded with suitable attachment hardware. An excellent insulation layer is kept on top of the lid of course.

I'll have to work up a spreadsheet to calculate how large a crucible could be heated per kilowatt.

Didn't someone already try a system like this? I don't recall (from my various wander though the Great Phosphorus Thread), but it seems like someone would have.
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5405
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: pumped

[*] posted on 11-6-2016 at 21:18


Sounds good. I don't recall any previous use of an arc furnace. This will be a very interesting experiment. ;)



The single most important condition for a successful synthesis is good mixing - Nicodem
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 12-6-2016 at 22:20


I did a little investigation and some simple calculations to clarify this idea, and estimate its cost.

There are two power options, use a standard 20 A 120 V circuit, and a cheap arc welder for the power (48 V) for the supply - 2.4 kW max ($75 for the welder); or use a 50 A 240 V circuit and a more powerful welder ($300-$800 to install the circuit, and $220 for the welder) - 12 kW max. I'll look at the first most closely, since it is the most plausible one to actually try.

Steel pails run $50, a bag of Kastolite $60, a bag of vermiculite $40, graphite crucibles $57-136 depending on size.

I already have four 50 A, 240 V circuit breakers in my fusebox (two for my AC, two my electric oven), so adding one should be straightforward. If I were to develop some rationale for getting such a circuit for some other reason (probably not) then the 12 kW version would be a more likely project.

I examined the resistance heating option and discarded it - getting an appropriate resistance with a continuous carbon element seems infeasible in the small system, and marginal in the large one (requires a very slender rod). So arc heating it is.

The arc rod feed actually looks pretty easy to resolve. Weld a steel feed tube to the lid, and put a PTFE seal on the top, then advance the rod by hand, minding the ammeter. You aren't going to run this thing unattended anyway.

There are always complications (I build scientific hobby stuff all the time, so I know), but this doesn't look very difficult to execute. The main trick is getting the lid made, two gas hook-ups and a feed tube. I don't weld (but know guys who do) but would have this welded professionallly (maybe two of them, one as a back-up if you can get a discount on two).

Still have to work out the scale achievable with those two power levels.

One problem though is crucible conditioning - they must be heated to a red heat empty, with the direction that it be done slowly and uniformly, without an impinging flame. In other words, in a regular kiln. Using the arc furnace itself for this probably would not work.

A single purpose heat-soak box for the crucible could be built from insulating fire brick, along the lines of other DIY furnaces presented on SM which would cost a lot less than getting a regular kiln.

Yes, I know Seattle Pottery. I have samples of different refractory insulating materials from there.

But living in the Los Angeles area I have Aardvark Clay, Laguna Clay McMaster Carr, etc. locally so I do not need to order firebricks or heavy bags of Kastolite from Seattle.

[Edited on 13-6-2016 by careysub]
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5405
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: pumped

[*] posted on 13-6-2016 at 13:22


The Landis water seal for the carbon electrode might be a good option.

You could install a 50a breaker and wiring yourself if you feel confident of your electrical skills, thus saving a lot of labor cost. I ran a 240v 20a service to my lab for my tube furnace and space heater myself. It was no big deal.

The Landis design uses the crucible as one electrode. This is a good design for controlling the position of the central carbon rod electrode I feel. I understand that an arc must be "struck" before it will self-sustain. This design seems good for that.

I'm glad you have good suppliers locally. Shipping "dirt" is expensive. I love McMaster-Carr. ;)

Check your local college pottery shop or commercial pottery places to see if they will fire your crucible for you.

[Edited on 13-6-2016 by Magpie]

[Edited on 13-6-2016 by Magpie]




The single most important condition for a successful synthesis is good mixing - Nicodem
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 13-6-2016 at 17:45


Quote: Originally posted by Magpie  
The Landis water seal for the carbon electrode might be a good option.

....

The Landis design uses the crucible as one electrode. This is a good design for controlling the position of the central carbon rod electrode I feel. I understand that an arc must be "struck" before it will self-sustain. This design seems good for that.


I am pretty sure a Teflon seal, just a Teflon plug with a hole the exact size of the rod, will do fine. They did not have Teflon back in 1900. It will also allow running the seal at above 100 C (up to 280 C in fact).

Yes, using crucible as one electrode is exactly what I envisioned (though actually using an internal bridging disk, not the crucible body itself).
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5405
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: pumped

[*] posted on 13-6-2016 at 18:09


The Landis design does not show any kind of disk for the crucible electrode. I'm not sure one is needed. I wonder what the path of the current is from the rod tip to the crucible. Maybe once slag forms on the bottom of the crucible the current goes to the sidewall?

[Edited on 14-6-2016 by Magpie]

[Edited on 14-6-2016 by Magpie]




The single most important condition for a successful synthesis is good mixing - Nicodem
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 14-6-2016 at 07:35


Quote: Originally posted by Magpie  
The Landis design does not show any kind of disk for the crucible electrode. I'm not sure one is needed. I wonder what the path of the current is from the rod tip to the crucible. Maybe once slag forms on the bottom of the crucible the current goes to the sidewall?


Accompanying the Landis illustration is the following text:
"It consists of an iron box lined with vitrified bricks and having an inner lining of carbon blocks which form one electrode; the other electrode, E, being vertical... The carbon lining is composed of two layers of blocks, so that the inner set of blocks can be replaced, when worn away, without disturbing the outer layer. The inner tapping hole for the molten slag is plugged with a piece of wood, without any luting; an outer plug, which is luted, prevents access of air to the inner wooden plug."

His design has a sacrificial replaceable surface and a slag drainage system.

The disk in my design serves both purposes - the slag collects below the disk, the disk is the sacrificial electrode surface so that nice carefully conditioned crucible does not get eaten up by the arc. Burning a hole in your crucible is bad practice.
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5405
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: pumped

[*] posted on 16-6-2016 at 08:59


Here's a good thread to review ("Carbon Arc Furnace"):

http://www.sciencemadness.org/talk/viewthread.php?tid=2680

My friend (a ChE) bought a used Lincoln AC-225 "tombstone" arc welder for $100. He used it to make rubies and other gemstones. He had to hire an electrician to put in the electrical service, however, so that probably cost a few $100 or so.




The single most important condition for a successful synthesis is good mixing - Nicodem
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 16-6-2016 at 12:51


Thanks, I have now read that entire thread (and skimmed most of the available references included). I like a remark you made way back when (more than a decade ago):
"This is a very interesting problem. The more articles I read the less clear it is to me exactly how to proceed. But I'm going to keep reading."

That is pretty much how I feel.

Investigating the power supply issues has raised a lot of questions.

I note that the arc welding units I have been looking at have duty cycles of only 20% over 10 minutes. Two minutes of operation every 10 minutes is not going to cut it for an arc furnace.

This seems to be a straight-up thermal issue. The arc welder has a small air cooling fan as the only cooling mechanism (some have none).

Perhaps I could Rube Goldberg the arc welder - disassemble for the parts and immerse the transformer in an oil bath, with oil circulation and water cooling.

Alternatively there is the old, old (old!) salt water rheostat approach - which I have some experience with. A friend and I build an arc furnace in 7th grade using this, a flower pot, and electrodes from a battery (the water boiled in operation).

I thought this maybe a little too redneck for a high duty cycle furnace, but looking up liquid rheostats I see this is a very credible technology and might be a good approach with appropriate implementation choices. A key issue I have with this is not knowing how to actually design one properly, I like to be able to do at least rough performance calculations on anything I build. How much of the power is dissipated in the water bath? More research required.

And then there are the microwave oven transformer option, which combines the transformers, after rewiring into step-down versions, from several ovens. Old ovens may be cheap but this is a heck of a lot of work, and I am less than thrilled with doing a lot transformer rewiring (electrics are not exactly my thing). Also, according to the sites I visited about this, not really that cheap since rectifiers and such must be purchased.
View user's profile View All Posts By User
wg48
National Hazard
****




Posts: 394
Registered: 21-11-2015
Member Is Offline


[*] posted on 16-6-2016 at 15:21


I have tried using a cheap welder and mains power via a ballast.

Cheap welders can be used continually but on lower current setting. The main problem I found was the open circuit voltage is only a round 41 volts ac so only a short arc is supported half an inch at max and easily extinguished when cold material fall into the arcs. That also means the electrodes need to be advanced frequently. I was using welding electrodes carbon ones may be better but a suspect not.

The mains ballasted version having an oc voltage of 240V was much more forgiving able to sustain several inches of arc or more in conduction mode. I used the electric cook as the ballast (oven grill and hot plates all on) perfect for quick experiments. Potentially very dangerous unless you know what your doing and your very careful.

You do not need to rectify the ac but if you do and use an inductive ballast it would be dc and support a longer arc.

The resistance of a water rheostat is adjusted by the concentration of the electrolyte. The required resistance is determined by your required maximum current under short circuit conditions. But its going to boil eventually unless the container is very large or a cooling arrangement is constructed.
View user's profile View All Posts By User
careysub
International Hazard
*****




Posts: 1339
Registered: 4-8-2014
Location: Coastal Sage Scrub Biome
Member Is Offline

Mood: Lowest quantum state

[*] posted on 16-6-2016 at 18:20


I am continuing this discussion on another thread:
http://www.sciencemadness.org/talk/viewthread.php?tid=12862&...
(this one seemed most applicable), since the discussion has moved on to discussions of arc furnace design not really specific to phosphorus.
View user's profile View All Posts By User
BluePlanet1
Harmless
*




Posts: 41
Registered: 30-1-2016
Member Is Offline


[*] posted on 7-2-2017 at 00:51


Hey all!

It's been 13 months since I posted (can't believe on the same page) but I think I've found my way to a great P4 distiller.

After my last video I thought I was close to figuring this P4 thing out. But after running many tests off camera, after months and months of research I started noticing some very basic, reoccurring themes & problems.

Not just problems I was facing but problems that were faced in commercial industry in the 60s, 70s and 80s. Many problems that were solved in the narrow scope of commercial industry... but are still very common.

Let me list the 2 biggest / most common issues:

1) Expansion. Also known as "intumescence" which leads to spalling.

For those who don't know what spalling is, it's when material expands and breaks away from the surface of refractory material. This was happening in every forge I made. And fire brick factories faced the same problem decades back.

Fire bricks use to be made with water. Usually some wet hydroxide like lime or magnesia lime. Which was mixed with alumina & silica then fired.

The bricks came out ok... until refractory liners began cracking & exploding.

So they switched the recipe to dry alumina and fumed silica. Which are "fused" together dry under 2 million PSI. Then fired dry to create a new type of non-porous ceramic fire brick that has no internal stress or water. This type of fire brick is what factories use today (under the liner). It can handle extreme stress. Does not crack and takes forever to age.

2) Water which creates internal stress. This was consistently my 2nd biggest problem. Refractory compounds are loaded with hydrates. And all hydrates behave differently.

Certain hydrates like sodium silicate expand 5-10xs their size when heated.

Other hydrates like plaster shrink to 40-50% their size. And worse, plaster is a flame retardant.... not an insulator! It has no role in a forge.

Many hydroxides also lose water. Like magnesium hydroxide. First it loses water from it's hydrate. Then drops it's hydroxl group at 380C to turn to into magnesium oxide which is used to line forges like aluminum oxide is.

Then you have silicates.

And this was the hardest lesson for me to learn so I'll write it in all caps...

SODIUM SILICATE WILL DESTROY YOUR FORGE.

I understand lots of people use it. I still use it. But there's a reason why so many sodium silicate compounds have been recalled from the market.

It's highly unstable. It ages fast. At the same time it heats & expands creating high internal stress... it also becomes soft like talc and crumbles to pieces.

Not that you shouldn't use any waterglass, but if you do you need to know what you're doing.

Which gets me to my next major finding. Something magpie vaguely mentioned a year ago...

CLAY IS YOUR BEST FRIEND.

If you're gonna build a forge or system to make P4... learn about clay.

Clay is unbelievably cheap AND stable. It's strong. It's easy to mold and work with. Firebrick is made of clay. And it can be mixed with SO MANY different materials to make anything you can think of.

I'll give 2 examples:

1) If you fire pure wet clay it will shrink and crack a bit. But post fire it's still very strong & stable. And if you paint a *dilute solution& of water glass to the OUTSIDE of clay, it works its way deep into the pores. Then if you fire it again you get a material that has the hard, durable strength of clay with the softer, reflective and highly insulative properties of sodium silicate.

The outside can spall and crack if you use too much sodium silicate. But it can't spall and crack the core since the core is clay and clay is stable.

2) Clay can be mixed with perlite. So you can make lighter, more insulative fire bricks.

For people who don't realize, commercial furnaces do the same thing. They use high clay fire brick for the foundation. So the foundations are hard like steel. Then soft clay liners for the insulation.

It's really a balancing act between a hard foundation that holds too much heat (like metal) vs a soft liner that reflects heat.

You don't want either or for white phosphorus, you want BOTH.

Mold your foundation with high clay composition. So you can shape your furnace into anything you need. You can build in a feedstock if you want.

Then line it with a sodium, magnesium or calcium silicate. All 3 are easy to make. A 50/50 mix of sodium silicate with magnesium silicate (talc) or magnesium oxide (cooked out from milk of magnesia) will use the sodium to bind (water soluble) and the magnesium as a stabilizing agent to slow down aging, expansion and spalling.

This post is long and more about refractory materials than white phosphorus.

But as you'll learn refractory is much more complicated than a straight forward reduction reaction.

And once you learn this stuff you can build ANYTHING YOU WANT.

The sky truly is the limit.

Right now I'm working on a sexy clay "furnace" that looks more like a mini-white phosphorus factory.

I can add forced air if I need. Can add a feedstock to push in more crude phosphate. I can power it with oxyhydrogen from my water torch to do carbon / phosphate. Or fuel it with waste oil for higher temperatures.

But again, the biggest thing that's helped is learning to work with CLAY.

Clay turns to ceramic around 1600C. And I know I'm reaching that temp with regular air + propane since I can see the clay fuse into ceramic as it cures (the outside melts smooth). It makes a super hard ceramic that you can quickfire with no spalling.

One last thing. I promised over a year ago to get more videos up and I will. I've been busy testing different compositions of firebricks and will show exactly what I used, how easy it is to make everything, then I'll start experimenting on video with different fuels. I really think oil will work with carbon-phosphate reductions since oil gets hot enough in air in to melt steel.

I'll keep the forum updated! :)
View user's profile View All Posts By User
Chemetix
Hazard to Others
***




Posts: 203
Registered: 23-9-2016
Location: Oztrayleeyah
Member Is Offline

Mood: Wavering between lucidity and madness

[*] posted on 7-2-2017 at 13:24


Great post! Refractory is critical to so many processes, your rediscovery is certainly going to help me design some high temp work.
View user's profile View All Posts By User
TheNerdyFarmer
Hazard to Self
**




Posts: 92
Registered: 30-9-2016
Member Is Offline


[*] posted on 7-2-2017 at 14:48


What sort of clay do you use?

Edit: Also what is your youtube channel? I'd like to check it out.

[Edited on 7-2-2017 by TheNerdyFarmer]
View user's profile View All Posts By User
 Pages:  1  ..  49    51    53  ..  55

  Go To Top