Sciencemadness Discussion Board

CBrCl3 -> CCl4 or CHCl3?

Sauron - 7-7-2008 at 23:44

As it happens, chloroform and carbon tetrachloride are restricted where I live. Fortunately, bromotrichloromethane is not. I just ordered a liter (2 Kg) of this stuff.

Apart from my immediate use for it, which is prep of sodium trichloromethanesulfinate, NaCl3CSO2, I am interested in other exploitation of this material.

I know I could turn it into rather expensive chloroform by treating it with HBr, an equilibrium reaction that also yields Br2.

Can I also turn it directly into CCl4?

The main technological use for CBrCl3 is as a chain initiator for polymerization of acrylics, and I am also given to understand it is a bromination reagent.

Anyone know anything about this compound?

ScienceSquirrel - 8-7-2008 at 03:56

I would be tempted to try mixing it with anhydrous FeCl3 and bubbling dry chlorine through it.
Then just do an aqueous work up and fractionate the organic phase to see what you get.

I am pretty sure that UV light or another suitable free radical initiator in the presence of chlorine would give you CCl4 as the first step in most of it's reactions is splitting the molecule to form CCl3. and Br. and you would be trapping this radical with chlorine.

Apart from that I am sure there is some horrible low temperature organometallic stuff you could do. It looks like a certain starting material for dichlorocarbene or an equivalent.

JayGatsby - 8-7-2008 at 05:02

I did a literature search on bromotrichloroethane. Here is the most interesting article I found (a matter of perspective):

A novel method for the synthesis of dichloroalkenes
Russian Chemical Bulletin, International Edition, Vol. 53, No. 11, pp. 2647—2649, November, 2004
Quote:

"...it was demonstrated that bromotrichloromethane is a more reactive alkenylating reagent than tetrachloromethane, even in the presence of a catalyst (0.5 mol. % CuCl). In the present work, with the aim of extending the range of the target products, we carried out reactions with a wide set of aliphatic aldehydes and ketones and obtained dichloroalkenes in satisfactory to good yields..."


Attachment: A novel method for the synthesis of dichloroalkenes.pdf (233kB)
This file has been downloaded 978 times

Sauron - 8-7-2008 at 05:35

Generating the Cl3C. radical is a good way to get hexachloroethane which isn't my target.

So at the moment I may have to settle for chloroform rather than CCl4.

I recall that wet Fe filings reduce CCl4 to CHCl3. Seems easier than HBr.

Thanks for the suggestions.

not_important - 8-7-2008 at 05:58

Does the HBr method require anhydrous conditions? If not the HBr method be modified to start with a small amount of aqueous HBr and use SO2 to reduce the Br2 back to HBr.

Iron does sound easier, although I remember that also taking the reduction further to for some CH2Cl2, although not much.

Turning CBrCl3 into CCl4 should be do-able through halogen exchange. But to force the equilibrium you'll need a huge amount of the chloride source, or maybe fractionally distill the CCl4 away from the reaction mix. It's easy with iodine, because of the differences of solubilities of the sodium halides in acetone, and with fluorine because of the stability of the C-F bond.

Sauron - 8-7-2008 at 07:14

Antimony trichloride or pentachloride, maybe.

I found some p-chem lit. on using these equilibria to measure accurately the heat for formation of CCl4.

Br2 + CHCl3 <=> CBrCl3 + HBr

CCl4 + Br2 <=> ClBr + CBrCl3

But I only have first page and will have to go beg for full paper. Maybe experimental details are there.

I suspect that most workers will be trying to make CBrCl3 from CCl4 or CHCl3 and not the other way around simply because the mixed halomethane is more valuable/less available. But in my circumstances things are different. CBrCl3 costs me $200 a Kg (500 ml). I would have to run a lot of rather large scale haloform reactions to end up with 500 ml CHCl3, or chlorinate a lot of expensive CS2 to end up with that much CCl4, and with a substantial amount of aggravation in between dealing with sulfur chlorides and purifying the tetrachloride.

I'll very likely do those anyway, but this is a fast expedient method. Sometimes CCl4 is really the best choice. N-halosuccinimide halogenations for example.

My other route to CCl4 is via chloral, or diichloroacetyl chloride, by photolyzing either and adding Cl2. This one also I want to do. But I may no longer have to rely on it on a larger scale.

Chlorination of methyl thiocyanate is yet another planned experiment. In general this follows same course as CS2 chlorination with the added fillip of a crop of cyanuric chloride at beginning. After that, overchlorination goes to CCl4 and sulfur chlorides, unless you want to stop and harvest thiophosgene or trichloromethyl sulfenyl chloride. But like CS2 the MeSCN isn't cheap.

Sauron - 8-7-2008 at 08:13

CBrCl3 boils at 105 C.

CCl4 at 76 C.

How about using NCS or TCCA as Cl2 source, refluxing the CBrCl3 and stripping out the CCl4 as it forms? Given a 30 degree delta, any decent column ought to seperate the two.

--------------

It appears that chlorination of CCl3Br produces BrCl interhalogen compound and the reaction is a chain and far from simple. I located an older JACS paper on gas phase photochemical chlorination of CCl3Br and hope to have full text shortly.

According to this nice JACS paper from 1934 illumination of CCl3Br and Cl2 in vapor phase by a Hg arc lamp with an appropriate filter, allows conversion to CCl4 with no formation of C2Cl6. Without the filter, hexachloroethane is a major contaminent. Paper is attached.

[Edited on 9-7-2008 by Sauron]

Attachment: ja01322a006.pdf (726kB)
This file has been downloaded 808 times

Sauron - 8-7-2008 at 19:34

According to that JACS paper, the authors prepared CCl3Br from CCl4 and AlBr3 by leaving the mixture standing for 3 days..

I wonder whether treating CCl3Br with AlCl3 will accomplish the reverse exchange?

They used 260 g AlBr3 and 350 cc CCl4. After washing the product and drying over CaCl2, they subjected the mixture to repeated fractional distillation, collecting a center cut of CCl3 boiling over a 0.15 C range. The cut was 45 cc. Obviously they were after a rigorously analytical product, the actual yield of a good grade would have been much higher.

260 g AlBr3 is just under 1 mole.

350 ml CCl4 is >3.5 moles.

So theoretical yield of BrCCl3 would be 1 mole 198 g. The middle fraction they took was 90 g (45 ml). Not bad, considering I'd estimate the actual yield was twice that.

AlBr3 is rather costly but can be prepared from the elements. Cf. Brauer.

[Edited on 9-7-2008 by Sauron]

F2Chemist - 11-7-2008 at 13:05

If you could get your hands on some Pd/C (~5-10%) and some H2 (from water), you could pretty easily make a heated column and reduce the CBrCl3 to chloroform (and HBr) by passing it with H2 over the Pd/C.

Sauron - 11-7-2008 at 19:50

Wet iron filings will do same thing (CCl4-CHCl3).

CBrCl3 is a selective brominating agent, I have to look up the lit. on this (cited in Aldrich) and see what it is selective for.

But getting to chloroform is not my priority. Getting to CCl4 is.

ShadowWarrior4444 - 11-7-2008 at 20:20

These articles may be of use, most likely to lead to other references though:
http://pubs.acs.org/cgi-bin/abstract.cgi/jpcbfk/2004/108/i30...

http://www3.interscience.wiley.com/journal/109570086/abstrac...

Photolysis seems to be mentioned quite a bit on the subject of chloro-carbon compounds; perhaps Cl2 gas bubbled through the CBrCl3 under UV light would work, or the same technique having them both in the gas phase. Though, to drive the equilibrium toward chlorine, one may wish to keep the CBrCl3 at a temperature above the BP of Br2 while below its own BP (140c?) as Cl2 is bubbled through; an apparatus to condense the Br2 and recycle the Cl2 would then be in order.

Should that system work, you could have a very convenient source of Br2 at CCl4. The UV light might need to be quite strong; a TEA laser at 120Hz would certainly be an amusing choice for its source.

[Edited on 7-11-2008 by ShadowWarrior4444]

Sauron - 12-7-2008 at 00:25

UV catalyzes both the breaking of the C-Br bond and the formation of hexachloroethane, the trick is to filter the UV to do one and not the other. Fortunately I already have that from the literature. But thanks for your thoughtfulness.

PS the Wiley link does not work. What is the citation?



[Edited on 12-7-2008 by Sauron]

ShadowWarrior4444 - 12-7-2008 at 00:43

Quote:
Originally posted by Sauron
UV catalyzes both the breaking of the C-Br bond and the formation of hexachloroethane, the trick is to filter the UV to do one and not the other. Fortunately I already have that from the literature. But thanks for your thoughtfulness.

PS the Wiley link does not work. What is the citation?



[Edited on 12-7-2008 by Sauron]



The abstract and citation [perhaps it could provide the appropriate wavelength data]:

The reaction of trichloromethyl radicals with hydrogen chloride and a new estimation of the rate of combination of trichloromethyl radicals
I. A. Matheson, H. W. Sidebottom, J. M. Tedder
Department of Chemistry, University of St. Andrews, St. Andrews, Scotland


Abstract
The effect of the addition of hydrogen chloride on the photolysis of carbon tetrachloride in the presence of cyclohexane has been investigated in a companion paper. The data enable the rate constant ratio k8/(k5)1/2 to be determined. Since k-8 is well established, k5 can be estimated from known thermochemical data. The validity of the thermochemical derivation is checked by applying it to trifluoromethyl radicals. The photolysis of bromotrichloromethane and carbon tetrachloride in the presence of hydrogen chloride has been investigated over a range of temperatures. From these results and assuming reaction (5) has no activation energy, Arrhenius parameters for reaction (8) have been determined:




The activation energies for the reaction of methyl, trichloromethyl, and trifluoromethyl radicals with hydrogen chloride are compared, and at first sight surprising results are rationalized in terms of relative electronegativity.



--------------------------------------------------------------------------------
Received: 9 November 1973; Revised: 21 January 1974

[Edited on 7-12-2008 by ShadowWarrior4444]

Sauron - 12-7-2008 at 02:20

Here's the lit. I already obtained and the wavelengths are in there as is the composition of the green inorganic solution used as a filter. CCl4 was the sole organic reaction product (no hexachloroethane was formed.)

This was the gas phase reaction of CBrCl3 and Cl3.

These two papers demonstrated in 1934 that Cl2 liberates from CBrCl3 not Br2 but ClBr.

I'd be tempted to try to run this in liquid phase, using an immersion well to hold a Hg vapor arc lamp (the ones I have are 400 W) and use the green solution as both filter and coolant.

But if it must be vapor phase, that is just a slight modification to the glassware.

[Edited on 12-7-2008 by Sauron]

Attachment: ja01322a006.pdf (1.3MB)
This file has been downloaded 665 times

Sauron - 12-7-2008 at 02:35

This is a complete citation:

The reaction of trichloromethyl radicals with hydrogen chloride and a new estimation of the rate of combination of trichloromethyl radicals
I. A. Matheson, H. W. Sidebottom, J. M. Tedder
Department of Chemistry, University of St. Andrews, St. Andrews, Scotland

International Journal of Chemical Kinetics 6, pp 493-506

DOI 10.1002/kin.550060405

With this, one can expect to obtain the full text in References.

Without that DOI...rots of ruck.



[Edited on 12-7-2008 by Sauron]

JohnWW - 12-7-2008 at 05:10

Quote:
Originally posted by Sauron
As it happens, chloroform and carbon tetrachloride are restricted where I live. Fortunately, bromotrichloromethane is not. I just ordered a liter (2 Kg) of this stuff. Apart from my immediate use for it, which is prep of sodium trichloromethanesulfinate, NaCl3CSO2, I am interested in other exploitation of this material.(cut)

Can you tell us WHY CHCl3 and CCl4 should be "restricted" in your country? They are useful non-inflammable polar aprotic organic solvents, while CCl4 can be used as a non-conducting non-aqueous sprayed fire extinguisher. I suppose CHCl3 could be used to render people unconscious, and both it and CCl4 are liver toxins and may have ozone-depletion implications, but that seems to be insufficient reasons to restrict them as laboratory reagents or solvents and for fire-fighting, as opposed to large-scale industrial use. Especially, as you say, CCl3Br, which can be used for many of the same purposes as CCl4, is not restricted in your country; and CF2ClBr and similar mixed halides are widely commercially available in fire extinguishers suitable for vehicles or boats, for electrical and fuel fires.

Sauron - 12-7-2008 at 06:55

Chloroform was targeted by the Thai narcotics authorities as a solvent used by mathamphetamine ("yaa baa") illicit labs.

Carbon tetrachloride and some other chlorinated hydrocarbons were targeted as being environmental problems.

As a consequence special licenses are needed to purchase these even in laboratory quantities.

Such licenses are rarely forthcoming.

Thus far DCM (methylene chloride) has escaped such stupidities.

Ethylene dichloride and ethylene dibromide were banned by some silly-ass UN convention because of their role as agricultural fumigants, apparently some undereducated African farmers were using them improperly and they were deemed to be a threat to wildlife.

Once again no exemptions were made for lab quantites.

It's all perfectly true and equally idiotic.

ShadowWarrior4444 - 12-7-2008 at 13:50

Quote:
Originally posted by Sauron
The filter removes wavelengths shorter than 0.485 u. I believe that Greek letter is a mu for micron, and the wavelength is more familiarly stated as 485 nanometers.

The composition of the filter solution is:

153 g CaCl2.4H20
16.7 g CuCl2
1 cc 2.5 N HCl,
178 cc water.

The path through the filter soln is 14 mm thick.

The solution is referenced from "Atlas of Absorption Spectra" by Uhler and Wood, Carnagie Institute Publ., Vol. 71 (1907)

I would utilize the filter solution as the coolant for the Hg vapor arc lamp and therefore the cooling jacket doubles as the filter. The immersion well needs to be redesigned so that the inside diameter of the cooling jacket is 14mm, and the other changes necessary to irradiate the vapor phase mixture incorporated at the same time. The final apparatus will be a triple walled cylinder.



I would be wary of using a salt solution as a coolant, especially CaCl2--don’t want the pumps to corrode or build up insoluble calcium deposits.

Though, a mercury vapor arc lamp should not require cooling, they are usually meant to operate at high temperature and pressure, lowering the temperature would cause it to emit almost exclusively in the UV range. [They traditionally have an outer bulb surrounding the inner fused quartz bulb to provide thermal insulation.] If you were to use a Xenon arc lamp, that would be a better candidate for cooling. (An Argon arc lamp may be even more useful, due to it's propensity to emit in the blue areas of the spectrum.)

You should also be aware of the glass that the reaction vessel and cooling jacket are made out of, to prevent any unnecessary losses. In addition, one should consider the depth to which the light can penetrate effectively, and design the reaction chamber accordingly. (Perhaps with one side mirrored to further increase efficiency.)

Incidentally, 485nm is blue light--will that have enough energy to break the CBr bond? (or are you relying on multiple photon absorption?)

[Edited on 7-12-2008 by ShadowWarrior4444]

Sauron - 12-7-2008 at 16:26

I commend to you the instructions for operating Ace-Hanovia medium pressure Hg vapor arc lamps. If you do not operate them in a properly cooled immersion well the quartz wall of the lamp will soften and bow. If the immersion well is borosilicate it will melt. if it is also quartz as the better ones are, it will also be in danger of deforming. The temperature of one of these lamps in operation is 600-900 C. That is why the coolant loops for photochemical systems include flow sensing alarm systems.

There are pumps, and there are pumps, and there are pumps. You seem to be assuming that I would use some particular sort that might be affected as you say. I would be inclined to use a peristaltic pump with an appropriate selection of tubing. Such a system precludes all contact between solution being pumped and anything but the inner wall of the tubing.

Frankly I have not yet looked at the spectral characteristics of the glasses concerned. Quartz passes just about the whole spectrum, borosilicate has a definite cutoff, which I ought to go compare to 485, and Vucor is intermediate between these. It is a common practice in photochemistry with these medium pressure Hg vapor arc lamps to use a quartz immersion well and then insert a sleeve of thin wall vycor or borosilicate tubing (made to fit the ID of the immersion well closely) between the lamp and the well to act as a filter. My comments upthread were based on the unverified assumption that the technique of the authors of the cited papers wouild be required. But if the same job can be done by the more common method, as exemplified by 15-20 examples of photochemical reactions with Ace Hanovia equipment in Org.Syn., then even better, and water can be used as coolant with a conventional recirculating chiller.

One final comment. Lamps must not be overcooled. Doing so can destroy them and damage the power supply/lamp ballast. Once again, read the instructions in the photochemical section of Ace's catalog online. These are costly systems not often employed by amateurs. I have a couple of them. Oh, and I ought to mention that they are very hazardous to eyesight The apparatus needs to be enclosed in a containment. The old method of wrapping reaction vessels in aluminum foil is in my opinion, very unsafe.

But thanks for your criticisms and concern.

[Edited on 13-7-2008 by Sauron]

Attachment: PhotochemPacket.pdf (752kB)
This file has been downloaded 815 times

Sauron - 12-7-2008 at 19:38

PS:

I did not answer your question about whether or not the lamp so filtered would have sufficient power.

I guess you ought to read the JACS papers, because unless they are a total fabrication, the answer can only be YES. The reaction did work, and selectively. CCl4 was only organic product; hexachloroethane was absent. As this is the result I am after, replicating their procedure makes sense.

ShadowWarrior4444 - 12-7-2008 at 20:38

Apologies--I seem to have missed the posting of the JACS paper previously. According to that article, the reaction was performed at 366nm and higher, still in the UV range. No reaction will occur at 485nm and higher, as this is well out of the UV range. They indicate that CBrCl3 decomposes to CCl6 and Br2 at 335nm and lower, so you would need to filter those but not anything of higher wavelength. 350nm-480nm should still be present.

They seem to indicate that BrCl forms and has a retarding effect on the rate of the reaction, but no effect on the end product. You would simply need a clever way to remove the BrCl as the reaction proceeds, or a way to slow its formation; though, these are not required, as the reaction will still proceed to completion, simply slower.

A note on the mercury vapor lamps: It indicates in your PDF that they can be operated outside of an immersion well in open air, the lamp surface stabilizing at 800C as normal. The quartz will not deform in the absence of cooling. I believe the only lamp that will cause its own quartz envelope to deform is a sulfur-plasma lamp.

It seems that the immersion wells are used to keep the rest of the glass apparatus cooled--so the borosilicate doesn’t soften when in close proximity to the lamp.


Ancillary: Several interesting articles on the selective breaking of the C-Br bond using laser light. References should the link not work:

Mode-Selective Control of Surface Reactions
John C. Tully
Science 19 May 2006 312: 1004-1005 [DOI: 10.1126/science.1126341] (in Perspectives)

Laser Control of Chemical Reactions
Richard N. Zare
Science 20 March 1998 279: 1875-1879 [DOI: 10.1126/science.279.5358.1875] (in Articles)

Sauron - 12-7-2008 at 21:01

You are quite right, the filter is designed to pass 366 nm and higher, the desire is to exclude 335 nm region. The nature of the filter so far has eluded my notice,

I will delete the post about the 485 nm filter as it is irrelevant.

I doubt it is just a glass filter as even pyrex passes too much below 366 nm to be effective. Vycor and quartz are out of the question.

Yes the cooling of immersion wells is to minimize heat transfer to borosilicate wells, or in the case of quartz wells, heat transfer to the reactants. Lab type lamps are rarely very long, as opposed to those used in technological applications such as printing, textiles, etc. and are far less likely to soften and bow. The use of a Hg medium pressure lamp in a housing external to the reaction vessel is far from unknown in the lit. but immersion wells are preferred for obvious reasons. There is an enormous cost discrepency between borosilicate and quartz apparatus as anyone who has ever purchased the latter can attest. This amounts to about a factor of 3.

Sauron - 12-7-2008 at 22:03

Aha!

I finally spotted the details.

They used a Corning #586 9mm filter which passes only the 365 nm lines of the Hg vapor arc lamp.

A preliminary googling suggests tat Corning spun off this specialty business to Corning/Kopp. It is now a matter of figuring out of #586 filter glass is still made or what the current equivalent might be. If I can't determine this on my own, I can always ask Kopp.

http://www.koppglass.com/filtersAndScientific.php


Anyone have any knowledge of or experience with such UV filters?

I am also interested in whether or not this vapor phase reaction also works in liquid phase. My guess is yes but probably at a lower rate since it is somewhat temperature dependent. I'd expect Cl2 to be quite soluble in CBrCl3 just as it is in CCl4.

[Edited on 13-7-2008 by Sauron]

Sauron - 13-7-2008 at 04:57

Kopp acquired Corning's filter glass specialty business in the early 1950s.

Here is url of specs page for Corning/Kopp 5860 filter which has peak at 365 nm.

This appears to be the filter of choice.

http://www.koppglass.com/filterdata/UVTrans/5860.htm

That is a UV transmitting, VIS absorbing filter. It has a very narrow bandpass entirely between 300 and 400 nm. Its peak is at 365 nm, but the percentage of transmission at peak is only about 43%.

The 7380 filter looks quite promising. This is a UV absorbing VIS transmitting filter with a sharp cutoff below 350 nm, but it passes a lot more % at 365 than does the 5860 filter and a lot less (effectively zero) at 335 nm.

Now if we examine the spectral output of our 450 W Hg vapor arc lamp we will see that the watts at 365 nm are 25 while at 335, 2.5 only. (Spectral data is from Ace fie entitled photochemical packet.pdf, posted upthread.)

So let's ask ourselves whether it is really necessary to remove the visible and IR components? Because if not then the 7380 filter does a much better job of discriminating between the two Hg A lines (365 nm that we want and 335 nm that we don't).

The 5860 passes about 10 W at 365 nm and about 0.5 W at 335 nm.

The 7380 passes 15 W at 365 nm and zero at 335 nm (well, something like 6 x 10to-3 W.)

I'd be inclined to try them both.

http://www.koppglass.com/filterdata/UVAbs/7380.htm


[Edited on 14-7-2008 by Sauron]

ShadowWarrior4444 - 13-7-2008 at 13:31

The 7380 is the clearly superior choice. The reaction is solely dependant on how many 366nm and lower (to 335nm) photons it is exposed to, therefore a filter that transmits 60% of those photons is superior in efficiency to one that transmits 40%. In addition, the 5860 allows 20% of the 335 photons to pass, leading to decomposition of the reactants and therefore lower yields.

Visible light transmission is not an issue as those photons do not possess enough energy to affect the reaction, and the reactants do not seem to like undergoing multiple photon absorption. (At least not without an IR laser.) If you are concerned about the visible light, layer a 5970 over the 7380.

As for whether it will work in the liquid phase, the answer is probably yes, though liquids tend to slow and refract light, so that would be a concern.

[Edited on 7-13-2008 by ShadowWarrior4444]

Sauron - 13-7-2008 at 17:49

I agree Perhaps at the time of the original research the 730 filter was not available.