Sciencemadness Discussion Board

Controlled Electrical heating for Catalysis of S2Cl2 to SOCl2.

Natures Natrium - 10-1-2005 at 12:30

Alright, I have recently been looking into constructing a catalyst tube for the conversion of S2Cl2 into SOCl2. My first thoughts, since the temperature needed is a mere 193C, was to use a test tube in an oil bath setup. However, looking at the contact times needed to achieve a reasonable yield (to me, >50%), this method would not be very viable.

So, I happen to have a 100mL, 75cm long borosilicate titration column with glass stopcock. If only using roughly the 50cm mid portion, that still gives much greater contact time than the 13cm testube. The problem then, of course, is heating.

I am fairly confident I can obtain the Ni-Cr wire, the problem is powering it. I will admit up front to a severe lack of knowledge in the area of electronics, so I am unsure if this power supply/dimmer will suffice without meltdown or other problems.

The device is a "model hobby transformer", for powering electric trains. It is rated for a DC output of 17V, and a total output of 7VA. It has a round dimmer switch to control the train speed. Would this work to power a ni-cr wire wrapped around 50cm of glass tube? (Hmm, not sure how much wire this would take...)

Well, my thoughts are still "half-baked" so to speak, a problem compounded by my lack of knowledge in this area of expertise. Any help, adivce, or reference material that would set me on the right track would be most appreciated.

Thanks,
Nature's Natrium


Edit by chemoleo: Changed title to something more appropriate

[Edited on 11-1-2005 by chemoleo]

axehandle - 10-1-2005 at 13:49

Quote:

The device is a "model hobby transformer", for powering electric trains. It is rated for a DC output of 17V, and a total output of 7VA. It has a round dimmer switch to control the train speed. Would this work to power a ni-cr wire wrapped around 50cm of glass tube? (Hmm, not sure how much wire this would take...)

No way. That's only 7W of power. I doubt the coil would even feel warm to the touch if powered by that transformer!

What I humbly suggest is that you use wall outlet AC together with a 300W light dimmer (most of the cheap ones are dimensioned for exactly that, 300W tops), and dimension your coil for a maximum power consumption of 300W.

My male intuition tells me that the best wire thickness for your project probably lies between 0.3mm and 0.5mm.

Let's do a calculation for a 0.3mm Nicrothal-80 wire (an 80/20 NiCr alloy):

Resistivity: 1.09 Ωmm<SUP>2</SUP>m<SUP>-1</SUP>
Resistance/m: 1.09 / (area) = 1.09 / (0.15<SUP>2</SUP>*pi) = 15.42Ω/m
Power P=300W
Voltage U=230V
Current I=P/U = 300W/230V=1.30A
Resistance R=U/I = 230V/1.30A=177Ω
length of wire L=177Ω/15.42Ωm<SUP>-1</SUP> = 11.5m

Say your glass pipe is 10mm in diameter.
Then the circumference = 10mm * pi = 31.416mm
Number of turns N then becomes 11500mm/31.416mm = 366

So, the coil should have 366 turns in it. Unless I made a mistake in the calculations, which would be very typical for me. :)

EDIT: corrected typo


[Edited on 2005-1-11 by axehandle]

Fantastic!

Natures Natrium - 10-1-2005 at 14:29

Excellent information. Although your tabulated answer is most helpful, the equations represented are even better! :cool: Now I have an idea of how these things are calculated, I can use this information to figure out what will work from what I have available to me, provided I go and purchase a light dimmer as you suggested. Thanks again!:D

Sincerely,
Nature's Natrium

EDIT: Ok, yet another question. If I am understanding this right, wouldn't 11.5m be the maximum length of the wire? Or is that the minimum length of wire able to withstand the up-to-177Ohms generated by the power supply? Either way, would it not be possible to use a shorter length of wire and simply not turn the power up that high, or is there some fundamental concept I am missing here?

[Edited on 10-1-2005 by Natures Natrium]

axehandle - 10-1-2005 at 16:22

Quote:

EDIT: Ok, yet another question. If I am understanding this right, wouldn't 11.5m be the maximum length of the wire? Or is that the minimum length of wire able to withstand the up-to-177Ohms generated by the power supply? Either way, would it not be possible to use a shorter length of wire and simply not turn the power up that high, or is there some fundamental concept I am missing here?

This is a bit hard to explain... I'll try:

11.5m is the length of 0.3mm wire that will draw 300W of power without any dimmer connected, given 230V and the resistivity of the alloy. If a shorter wire, say half the length, 5.75m, were used, it would draw 600W, blowing the dimmer's fuse. I'm not really privvy to how dimmers work, but they are fundamentally different from solid state power regulators. A power regulator, like the one I use with my furnace, would work like you described (at least I think so). A dimmer won't, if a load with higher power consumption than it's rated for is connected through it, it will blow out its fuse (or worse). It's just the same as if you were to hook up a 500W bulb to a 300W max dimmer -- the dimmer would be destroyed.

Just think of the coil as a 300W lightbulb.

(This explanation sucks, but I'll post it anyway...)

EDIT: Arrghh! Anyone with deeper knowledge of electronics than I please help explain!


[Edited on 2005-1-11 by axehandle]

Natures Natrium - 10-1-2005 at 18:10

Well, I am still struggling with the concepts. Half of my brain has accepted, but it just seems so counter intuitive somehow.

Anyways, I went out and bought a dimmer already (as my obsessive compulsive self is liable to do sometimes), and did the calculations based on the equations represented here earlier.

Assuming .3mm Ni-Cr wire (80/20), the following values were derived.

P=300w
U=120V
I=2.5A
R=48Ohms
L=48/15.42=3.11m

So, using this as a power source, I should be able to get away with 3.11 meters of wire.

To sum this up with questions, if I wanted to, I could easily use a 4m length of .3mm nichrome wire for this project without causing harm to my power supply? And by using a longer wire, wouldn't the temperature range available by turning the dimmer broaden? (ie 'more wire'='less change' in temperature per revolution of dimmer dial?)

axehandle - 10-1-2005 at 18:47

Yup, your calculations look correct.

3.11m is the minimum length that won't blow your dimmer. A longer wire (--> higher R) wouldn't gain you anything, it would only lower the maximum temperature of your setup. Or perhaps lower the maximum power of the setup, I'm not sure, depends on how the dimmer works. Either way, I'd aim at _exactly_ 300W, perhaps 280..290 to keep within safety margins.

EDIT: Come to think of it.. a longer wire would lower the maximum power of the setup from 300W to that of the wire itself (e.g. 200W), plus it would narrow the scale of the dimmer (0..200W, say, instead of 0..300W). Not good anyway.

I really should think longer before I speak...


[Edited on 2005-1-11 by axehandle]

chemoleo - 10-1-2005 at 19:41

Just went through the calculations, they also look all good to me :)
Natures Natrium, if you heard of these concepts for the first time, then you are a pretty damn fast learner!

Anyway - just to add - if the wire is longer, the amps it draws is less due to its higher impedence. This means the max temp. is lower, and indeed, if you have a dimmer (etc) attached to regulate the power, then of course your (i.e.. 200 W) power spectrum will be spread by 300W/200W = 1.5 fold; so yes, your resolution on the dimmer will increase 1.5 fold. Conversely, your maximum temperature will decrease by 1/3 (if you do not consider temperature diffusion coefficients etc).
This is for you to determine, i.e. first buy the longer wire, determine the max. temp., and then decide whether a shorter wire would do to facilitate efficient conversion of S2Cl2 to SOCl2.

Anyway - if I was you, I'd probably worry more about the glass being able to withstand the steep temperature gradients.
I hope it's borosilicate or something!

Further - way back in my glorious past (:D) I used, just like you intended, a glass tube for EXACTLY this - but for a different reaction. I didnt use fancy nichrome wires, just normal flower wire (term? - you know, the 0.5 mm thickness iron wire for tying flowers etc together), wrapped about 5 times around the 1.5 cm tube, and attaching this to my 24 V transformer (or so I remember). The wire turned red hot, and the heat was sufficient to catalyse the reaction inside - which was NH3/air going across hammered Pt wire - to produce NO2.

What I am trying to say - dont be too fixated on getting decent high temp wire, about any wire will do, particularly at the temp. you require.
If your dimmer is not accurate enough for your liking - you can always attach an Amp-meter to the input/output side and thus work out what amperage is needed to get this or that constant temperature (how you'll measure that temp. is yet another question).

Natures Natrium - 10-1-2005 at 20:54

Ok, well, I played around with the numbers some more, I am starting to get a firm grasp on the relationships between these variables.

For instance, if I double the length of wire from 3.11m to 6.22m, the amps are cut in half to 1.25A. (Which makes sense, since ampres are a measure of current, and doubling the length of the wire increases resistance Ohm by 2x, thereby allowing less current to pass through.) Also, by doing so, it would decrease the maximum temperature achievable, however it would also decrease the sensitivity of the dimmer dial. i.e., my 270 degree dial would be sensitive to .92w/degree on a 3.11m wire, and .46w/degree on the 6.22m wire, allowing for somewhat more precise control of temperature. (This is based on the dimmer having a minum power of 50w and scaling from there to 300w.) I feel this is important for this project because I only need a catalyst temperature of 193C, although I expect some leeway (+/- 10C), this is a fairly low temperature for Ni-Cr wire.

(Hum, I just noticed chemoleo's post. First, thanks for the compliment! :D Second, be sure to thank Dr. axehandle for his excellent lecture on the subject by providing both the data and the means from which it was derived. I learned more about electricity on one evening here at sciencemadness than I did working for over a year helping to install dimming equipment! (Although at the time I mostly did grunt work, to be sure. ;)) Third, this isn't the first time I have heard of these concepts, just the first time I have ever been able to understand them. I did also consider using a different wire, but I suddenly very much want to build an annealing oven for glass, which is an art I have been dying to get into for some time. Commericial annealing ovens run like $2k+, which is way out of my league. Now that I have this knowledge however, it could be a possibility, assuming I have some Ni-Cr wire. I should note, however, that I tend to dream big and do much less, an unfortunate effect of having to work for a living and liking video games just a little bit too much. ))

Using data on .32mm Ni-Cr wire I found at Omega, I derived that running 141.6w, or 1.18A, ought to create a temperature of around 200C, regardless of wire length. Next, I think I am going to sit down and calculate what using some other more common wires would do for this project.

I have to say it one more time, axehandle, thank you so much.:)

Sincerely,
Nature's Natrium

EDIT: Oh yea, I forgot to mention that I picked up some "Furnace Cement", claiming to be good to 1400C+. I was thinking of wrapping the wire around the borosilicate tube, and then coating it with a 5mm layer of the cement for insulation, protection, and support. Is there anything wrong with this idea? Any problems which might arise from doing this?

[Edited on 11-1-2005 by Natures Natrium]

This might work.

FloridaAlchemist - 10-1-2005 at 21:07

Try one of those toaster ovens . Just drill holes threw both sides insert the glass tube . Should give temp around 190 C. :o

axehandle - 10-1-2005 at 21:47

Quote:

EDIT: Oh yea, I forgot to mention that I picked up some "Furnace Cement", claiming to be good to 1400C+. I was thinking of wrapping the wire around the borosilicate tube, and then coating it with a 5mm layer of the cement for insulation, protection, and support. Is there anything wrong with this idea? Any problems which might arise from doing this?

The wire will expand slightly when hot, which could crack the cement. Also, if the thermal expansion coefficient of the glass pipe is higher than that of the cement, the glass pipe will crack the cement when the temperature rises.

There is a well-know workaround for the first: Dip the wire coil in molten paraffin/wax/similar prior to placing it over the pipe and covering it with cement. Then "burn out" the paraffin afterwards using a kitchen oven set at max. As for the second: I don't know, perhaps the pipe's
outside could be covered with paraffin so that once burnt off there will be a small airgap between the cement and the pipe wall.

Me, I'd simply put the coil around the pipe and wind rockwool around the whole package, secured with steel wire. Rockwool will be fine up to 900 degrees C or so.

bio2 - 16-1-2005 at 22:47

Having built a few of these tubes and lots of "dimmers" which are triac phase controlled circuits, I can offer a few tips.

Don't get a cheapo dimmer because the hysteresis causes them to drop out at low settings and the repeatibility is no good. Choose a ceiling fan speed control that has the trim pot or
build a symmetrical firing unit (only takes 4 resistors and 4 diodes more than a typical "dimmer";) PM me for a schematic if you need it. These are great, as once calibrated, the dial can be marked in true power and the circuit is dead on repeatible given the same input voltage.

I have never seen a cheap lite dimmer with a fuse or any other type of overcurrent protection. Yes, you can use a higher wattage than rated, with lower settings, but the inrush is likely to fry the triac and if you accidently turn it up too high it will melt down quite nicely (they always short). I have a 600watt one I modified for a heater and routinely run it at 1KW continous because I put it on a real heat sink.

The tube you describe wouldn't take more than 50-75 watts or so for the temp you need, assuming decent insulation and correct wire sizing. Stay with the NiCr and choose a size that gives you about a hundred degrees C more wire temp than you need at the expected operating current. Make sure the wire temp at the maximum current you will use is well below the softening point of the glass type or use stainless steel if it's compatible with the reaction.

Look in Vogel for details on wrapping these furnace tubes. Thick asbestos paper under and over the wire is really the best and easy too. Try to find the tan sheets as the black fumes off some tarry shit before it sets.

Good luck and have fun.

neutrino - 17-1-2005 at 07:24

Asbestos? You mean fiberglass, right? Asbestos isn’t that available these days.:o

bio2 - 18-1-2005 at 16:16

Asbestos paper and board is available at the hardware store around here. Despite all the hype it's quite safe as long as you don't powder it up and inhale it. Fiberglass won't take the temp for most furnace tube resistance wire temperatures . You could use ceramic tape underneath and tape or blanket over the wire. There is this tape called Zetex that may work but I have never used it.

bio2 - 18-1-2005 at 18:20

Check this out, NiCr helix calculator applet

http://www.wcrl.ars.usda.gov/cec/java2/heater2.htm