Pages:
1
2 |
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
SO2Cl2 -> SOCl2 - Some Musings
Thionyl chloride (SOCl2) is a desirable but for many, unavailable reagent that is rather difficult to prepare.
Len1 has demonstrated the practicality of the classical method in an Illustrated Guide on the oxidation of sulfur chlorides using sulfur trioxide. The
downside is that for many SO3 is a problem.
I dug up a prep using SO2 and phthaloyl chloride with ZnCl2 catalyst, but Len1's attempt to replicate this prep by the method of Kyrides failed.
There is a patent on the reaction of PCl3 and SO2Cl2 (sulfuryl chloride) over granular activated carbon to produce thionyl chloride and POCl3. This
is great but again the unavailability for most of us of PCl3 renders it useless.
Too bad because SO2Cl2 is easy to make and would be an ideal feedstock for making SOCl2. SO2Cl2 is prepared from dry SO2 and dry Cl2 over GAC or
camphor as catalyst.
Both gases are easily generated in the lab and run through drying trains, mixed and run through a fixed bed (tube) of GAC and collected.
So what I am thinking about is: what alternative reagents might undergo a similar reaction to that of PCl3 and SO2Cl2?
One that comes to mind is PhPCl2 which is commercially available. Another is Ph3P. Also commercially available.
A third might be antimony trichloride, readily made from the elements. Antimony is in same periodic group as phosphorus.
I'm just running this up the flagpole to see if anyone salutes. I am not particularly hungry for SOCl2 myself, having invested a lot of time looking
for alternative chlorination reagents. Others like it though.
Comments?
[Edited on 18-8-2008 by Sauron]
Attachment: US5498400.pdf (441kB) This file has been downloaded 639 times
Sic gorgeamus a los subjectatus nunc.
|
|
DJF90
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 2266
Registered: 15-12-2007
Location: At the bench
Member Is Offline
Mood: No Mood
|
|
How about SCl2? The reaction would be like this SO2 + Cl2 + SCl2 => 2SOCl2 ? This is what wikipedia says, but I'm not sure under what conditions
this would be? I did suggest this as a project for len1 in his thread
|
|
The_Davster
A pnictogen
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 2861
Registered: 18-11-2003
Member Is Offline
Mood: .
|
|
Seeing as we are musing...
As sulfuryl chloride is prepped from SO2 and Cl2, and SbCl3 is prepared from the respective elements, and assuming your SbCl3 instead of PCl3 idea
works (I like it, it was one of the first ideas I thought of before I was finished reading your post), I would be very interested to see the results
of a one pot type reaction passing chlorine and SO2 over a mixture of antimony and camphor.
|
|
solo
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 3973
Registered: 9-12-2002
Location: Estados Unidos de La Republica Mexicana
Member Is Offline
Mood: ....getting old and drowning in a sea of knowledge
|
|
Interesting Reading.....
Synthesis of [18O]thionyl chloride
Timothy W. Hepburn, Gordon Lowe
Journal of Labelled Compounds and Radiopharmaceuticals Volume 28, Issue 5 , Pages617 - 620, 1990
Attachment: Synthesis of [18O]thionyl chloride.pdf (161kB) This file has been downloaded 1211 times
It's better to die on your feet, than live on your knees....Emiliano Zapata.
|
|
len1
National Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 595
Registered: 1-3-2007
Member Is Offline
Mood: NZ 1 (goal) - Italy 1 (dive)
|
|
Sauron, I did not try the Pthalyl chloride + SO2 synthesis of Kyrides, so I cant say it doesnt work. It most likely does, as he seems a reliable
source. What I described was something of my own thinking, to wit
1) Benzotrichloride + SO2 T~170C -> not even a trace of SOCl2
2) I got my pthalyl chloride by benzotrichloride + phthalic anhydride as Kyrides described. The phthalyl chloride has a much higher bp than the
benzotrichloride, and so the reaction with SO2 can be tried as Kyrides did. His temperature of near 200C is not accidental, I presume theres no
reaction at lower temperatures. What I noticed is that SOCl2 is unstable above 145C, and this explains why Kyrides got a low yield (~60%), something
he did not understand. My point is that this is a method of last resort due to the long winded procedure, long reaction time (24hrs) and low yields.
3) The method solo posted is actually much closer to what I actually tried - they use 1,4-bis-thrichlormethylbenzene. I find it hard to believe that
having the -CCl3 group twice makes it so much more reactive that the reaction proceeds at low temperature.
4) Wiki does mention Cl2+SO2 + SCl2. No mention of conditions. I find it hard to believe for several reasons that it can take place at low
temperatures and pressures. High temperatures do not favour SOCl2
[Edited on 18-8-2008 by len1]
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
Len1, I agree that 24 hrs is a long rxn time, but 60% is not such a bad yield. Benzotrichloride is a pain to make but I just buy it. I also just buy
phthaloyl chloride (phthalyl chloride.) Have several Kg of each in slightly dusty bottle awaiting first use.
I guess I misremembered or misread your posts about this, thanks for clarifying.
I have not yet read solo's attachment. 1,4-bis-trichloromethylbenzene? Can't say I have encountered that before.
Hmmm, what Wiki is hinting at is SCl2 + SO2Cl2 somehow disproportionating to 2 SOCl2. Sounds like wishful thinking.
The Hill patent does work, the MFCs are not really required nor the UV-Vis monitoring. Nor the syringe pump for fine tuning the PCl3. Those are all
bells and whistles. Clearly intended as controls for scaleup of a pilot plant. There's another patent on an improved long-life catalyst replacing
GAC. And a paper from someone associated with the Geneva CWC watchdogs indicating they regard this technology askance. Poor babies.
The_Dayster, I doubt this is going to be a one-pot, any more than you could just throw in the rest of the mix with red P and expect that to go neatly
to PCl3. Preformed SbCl3, however might work.
Sic gorgeamus a los subjectatus nunc.
|
|
S.C. Wack
bibliomaster
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 2419
Registered: 7-5-2004
Location: Cornworld, Central USA
Member Is Offline
Mood: Enhanced
|
|
DE842041
DE852841
DE939571
US1861900
US3411886
http://dx.doi.org/10.1039/TF9474300667
It has also been made from SO2 with: BCl3, CCl4/AlCl3, NaCl/AlCl3, VCl4, and WCl6.
|
|
len1
National Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 595
Registered: 1-3-2007
Member Is Offline
Mood: NZ 1 (goal) - Italy 1 (dive)
|
|
Thank you for the patents.
The german ones deals with the reaction as I did it and I know it works.
One of the US ones deals with essentially the same process if you write H2SO4 = H2O + SO3.
I have illustrated an intermediate processes instead of the SO3 of Brauer and the above patent, or the conc H2SO4 of the US patent, I used about 50%
oleum - this worked better than pure SO3, and has the advantage that it leaves no S deposits in condenser and glassware - SO3 oxidizing the sulphur.
As for the other US patent - it similar to what I 'invented' and is illustrated in the practical guide, except I conducted it at 160C rather than the
cold. Frankly I dont believe it works in the cold, since there are patents where its done in a high-T bomb, I wonder why anyone would bother? Also
the processes is similar to the reaction with phthalyl chloride, which I believe works (because I believe Kyrides) and this required 200C
I suppose I could try to improve the method with conc H2SO4
[Edited on 18-8-2008 by len1]
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
As I recall one of the main US patents on SOCl2 from 65% oleum and SCl2 and/or S2Cl2, the reaction took 60 hours @ 70 C in a pressurized reflux system
of cast iron with glass liner and the pot has to be fed with dry Cl2 to suppress formation of pyrosulfuryl chloride. Otherwise the yield is way down.
That is the cost of using oleum instead of SO3.
Since the point of the SO3 is oxidation of the SxCly (whether or not the SO3 is neat or in H2SO4 as oleum) and H2SO4 per se is just about devoid of
ixidizing power, I am very skeptical of any patent using H2SO4 sans addition of SO3. But I guess I ought to pull the patent up and read it, anyway.
And I don't suppose it is necessary to point out that the economics of VCl4 and WCl6 make them absurd as stoichiometric participants in prep of SOCl2
- not catalysts.
The Faraday Transactions paper on direct catalytic oxidation of SCl2 by GAC charged with CaCl2 is attached. This is a vapor phase fixed bed reaction
with rather low efficiency amd is more of an outline for a process in search of a real catalyst, than a finished process.
The US patent using 100% H2SO4 employs catalytic HgCl2. I have seen that mentioned elsewhere, and it probably does work. I suppose 100% H2SO4 is an
improvement on oleum, though some oleum will have to be used to dry ordinary conc H2SO4 to "100%". The oxidation is really the job on the catalyst.
No catalyst no SOCl2.
[Edited on 18-8-2008 by Sauron]
Attachment: tf9474300667.pdf (548kB) This file has been downloaded 942 times
Sic gorgeamus a los subjectatus nunc.
|
|
len1
National Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 595
Registered: 1-3-2007
Member Is Offline
Mood: NZ 1 (goal) - Italy 1 (dive)
|
|
That oleum oxidizes SCl2 to SOCl2 under no more arduous conditions than reflux for 30 mins or so I can atest to personally. Moreover the second time
I did this using pure SO3 rather than oleum the yield was 10% lower.
I agree that the yield using conc H2SO4 should be lower - and the method seems less promissing.
The air oxidation seems not so bad - I think Ill try it
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
I will have to unearth the US patent I am talking about as it is one of the ones not included in SCW's list. I think it was referenced in Brauer but I
will have to go look.
The second US patent mentioned by S.C.Wack (first being catalytic HgCl2) is a 1965 patent to Sinclair Oil on the use of trichloromethyl-substituted
aromatics and FeCl2 catalyst to chlorinate SO2, forming the corresponding ary acy chlorides and SOCl2. The example compound they show is same as in
the paper solo posted, hexachloro-p-xylene. They claim this reaction proceeds at ambient temperatures in 48 hours in a pressure vessel. No mention of
Kyrides is made, not even a tipping of the hat. So much for searching prior art. The examiner ought to have been shot. Kyrides beat then by three
decades, published in the most prestigious American journal, and used benzotrichloride, well within their scope. He used ZnCl2 rather than ferric
chloride, and higher temperature, but, still.
I can try this eventually with benzotrichloride (on hand). The 1,4-bis(trichloromethyl0benzene is no longer available from Aldrich; from Across it is
$9 a gram, ouch! So sounds like a job for my photochemical reactor.
----------------------
The 1944 US patent to Hooker Electrochemical is the only patent cited by Merck 12th, along with the original Michaelis paper from Annalen, and the
Fisher monograph in Brauer's Handbuch. It is the one I have always principally relied upon, at least it is the reason why I bought 2 Kg of Merck 65%
oleum at horrible expense.
Attached below.
[Edited on 18-8-2008 by Sauron]
Attachment: US2362057.pdf (135kB) This file has been downloaded 650 times
Sic gorgeamus a los subjectatus nunc.
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
In looking over these US patents the two standouts, from point of view of milder conditions and ease of reagents, are US 1861900 and US 3422886.
The former teaches the use of 100& H2SO4 and Cl2 with a catalytic amount of HgCl2 to oxidize SCl2 to SOCl2. No oleum is required, apart from the
trivial amount needed to eliminate the few % water present in commercial conc acid. Such small amounts can be easily made by the persulfate and other
methods described on this forum.
Len1, please at your earliest convenence investigate this method.
The other patent employs trichloromethylbenzene compounds and catalytic ferric chloride to chlorinate SO2 in liquid phase at ambient temperatures. The
preferred reagent is 1,4-bis(trichloromethyl)benzene a.k.a. hexachloro-p-xylene. These are the mildest conditions, but commercial hexachloro-p-xylene
is expensive ($9/g) and so I have to look into preparing the reagent. Chlorinating p-xylene or preferably, if available, 4-trichloromethyltoluene or
1,4-bis(dichloromethyl)benzene seem likely.
I have lots of benzotrichloride, which is also claimed to work. And I think this can be done in a Parr shaker with a 60 psi pressure bottle which I
have. So all I need to do is condense some SO2 and try. Ought to get a yea or nay for benzotrichloride pretty fast. I think you can try this one too,
len1, if so inclined.
This is same reaction described in solo's posted paper. We were skeptical of that but now we have two references.
If either or both of these methods work then SOCl2 just got a lot easier.
SOMETIMES IT REALLY PAYS OFF TO SEARCH ACS JORNALS
especially when the goodies are in JOC and I have that on DVD thanks to Alter.
The reaction of trichloromethylbenzenes including benzotrichloride, but especially HCMX (hexachloro-m-xylene) have been extensively studies by duPont
and reported in a 4 part series in J.Org.Chem vol 41. Parts 1-4 attached below
This reaction works at temperatures consistent with what Kyrides reported, but without catalyst. It also works at much reduced temperatures with Lewis
acid catalysts, consistent with what was reported in the Sinclair Oil patent and the paper solo posted.
Hexachloro-m-xylene is readily prepared by the photo-catalyzed chlorination of m-xylene.
We have a whole new ball game here.
These papers also report reactions with P2O5, SO3 and various metal oxides.
This looks very sweet.
For example Part 2 describes how HCMX and P2O5 react above 190 C to form isophthaloyloyl chloride and POCl3. No catalyst. No pressure vesssel.
Just a fractionating column. POCl3 is collected and then the isophthaloyl chloride is distilled in vacuo.
More interestingly, benzotrichloride itself, readily available, reacts with P2O5 in absence of catalyst at >190 C. And Tricalcium phosphate reacts
with HCMX, to give 36% of the theoretical yield of POCl3, then the reaction mass turns dark and reaction stops. I would still regard this as a
practicable route.
Downsides:
The preferred catalyst SbCl5
The SOCl2 forming processes are done by necessity in metal pressure vessels; hot SOCl2 is very corrosive. Hastalloy C lines vessels are best. These
vessels are not cheap!
Good news:
While these reports were in press the US patent to Sinclair Oil cited upthread was issued, describing room temperature reactions similar to the above,
with FeCl2 catalyst. That patent uses HCPX rather than HCMX but there is no real difference in reactivity. This opens up preps in glass vessels or
glass lines reactors. Much cheaper than Hastalloy!
[Edited on 19-8-2008 by Sauron]
Attachment: jo00884a017.pdf (457kB) This file has been downloaded 4250 times
Sic gorgeamus a los subjectatus nunc.
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
I apologize for double post. For those who unlike myself derive more pleasure from patents than from journal articles, here are the seven count ' em
seven US patents Dupont obtained in connection with the work described in the 4-part series of papers posted above.
Let's summarize:
1. The paper solo cited and the Sinclair Oil patent teach that hexachloro-p-xylene reacts with liquid SO2 at ambient temperature and autogenous
pressure slowly to form SOCl2 and terephthaloyl chloride. FeCl3 (anhydrous) is catalyst.
2. Dupont researchers in a 4-part series in JOC and in 7 US patents tech that the same HCPX, its meta isomer HCMX, and other trichloromethylarenes
react with SO2 in same manner either at elevated temperatures and pressures, or catalyzed by a variety of Lewis acids especially SbCl5, FeCl3 and
AlCl3; some radical initiators are also effective catalysts.
3. Substituting SO3 for SO2 gets pyrosulfuryl chloride.
4. Benzotrichloride reacts with P2O5 at 1 atm above 190 C and no catalyst to form POCl3 and benzoyl chloride. The rate of reaction is half that of
HCMx (4-6 hrs vs 2-3 hrs.)
5. Various P compounds such as tricalcium phosphate and syrupy orthophosphoric acid react with HCMX to give lower yields of POCl3 (c.36%.)
[Edited on 19-8-2008 by Sauron]
Attachment: 7duPontpatents.pdf (2MB) This file has been downloaded 704 times
Sic gorgeamus a los subjectatus nunc.
|
|
len1
National Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 595
Registered: 1-3-2007
Member Is Offline
Mood: NZ 1 (goal) - Italy 1 (dive)
|
|
Thank you for that. I did rather think that the 6 Cl substituted xylene could be replaced by my benzotrichloride. They subsequently admit that hex
chlorinating xylene is very difficult.
I had some hope that the discovery that oleum reacts better with SCl2 than the SO3 was mine. This as the patent you posted shows was to good to be
true.
Id need a source of pressurised SO2 and an acid resistant pressure vessel to get the benzotrichloride conversion going. To be honest oleum is not so
frightening, and I think beats the hassle of the former. The POCl3 from P2O5 is good though - and as the author points out completely unexpected -
think ill try it.
As for SOCl2 from CCl4, think Ill pass. Unlike most people on this forum I actually do all this stuff so I have to be very careful, and Id much
rather not have an agonising wait of 24hrs after working with phosgene to make sure im safely out of the latent period![:)](./images/smilies/smile.gif)
[Edited on 19-8-2008 by len1]
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
You know from experience that getting the third Cl onto benzaldichloride is slow. With m-xylene, exactly the same and p-xylene is no different. So
procedure to make HCMX or HCPX is no different than benzotrichloride. Choose between peroxide initiation or UV - I am set up for UV.
All the details about difficulties have to do with isomeric purity and minor amounts of PCOX but that was all because they were after the 1,3 and
1,4-bis benzoyl chlorides (isophthalyl and terephthalyl chlorides) not the byproducts SOCl2 and POCl3. I can only speak for myself but those arylacyl
chlorides are no particular use to me.
The salvation of the SOCl2 side of this is the Sinclair Oil patent. Low temp catalysis by FeCl3 of reaction between SO2 ad HCMX or HCPX and likely
benzotrichloride. Temperatures from RT up 100 C mean less or no corrosion problems in steel reactors, and at the lower end, possibility to work in
glass. (No more than 50-60 psi.) A Parr bottle ought to do it, reaction might take days. But doable.
-----------------
I found the details for UV assisted alkyl perchlorination of methylbenzenes in a pair of articles from Ind Eng Chem in 1947. The McBee group at Purdue
wanted to prepare bis-trifluoromethylbenzenes, so they started with xylenes and chlorinated these, basing their procedures partly on a 1937 US patent
to good old I.G.Farben. I have compiled the two articles and the patent into one pdf.
With the help of a medium pressure Hg arc lamp the chlorination goes rapidly, with cooling at the start and heating to not more than 110 C at the
endgame.
The hexachloro-m-xylene forms faster than the para-isomer.
The products are low melting solids. HCMX mp 34-36 C.
I want to get the most use out of benzotrichloride but the SOCl2 prep may depend on HCMX. Fortunately I am set up for UV.
[Edited on 21-8-2008 by Sauron]
Attachment: McBee.pdf (1.2MB) This file has been downloaded 741 times
Sic gorgeamus a los subjectatus nunc.
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
Here's a 1960 JOC paper from one of the US Naval labs on hydrolysis of trichloromethylarenes catalyzed by (you guessed it) FeCl3 and SbCl5.
Depont and Sinclair were reinventing this 12-15 years later. Another instance where the patent examiners should have been tarred and feathered.
This paper was not even cited.
-------------------
In reading the Ind.Eng.Chem. papers I note that a lot of time was spent evaluating mixed xylene fractions because pure isomers were unavailable or
economically unattractive. That was 1947. Now it is 60+ years on and we do not have this problem. I have a bottle ofo-xylene sitting here, m-xylene
and p-xylene are readily available. Purities are typically 99% or better. Furthermore we are just trying for the oxide by products, the purity of the
various organic acid chlorides is unimportant.
Nor is overchlorination a problem as ring chlorinated bis(trichloromethyl)benzenes are just as active in the reactions with SO2 and P2O5.
All in all, this looks quite practicable. I am continuing to explore the literature on this.
[Edited on 21-8-2008 by Sauron]
Attachment: jo01077a012.pdf (162kB) This file has been downloaded 1382 times
Sic gorgeamus a los subjectatus nunc.
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
I had another look at Kyrides.
He heated 107 g phthaloyl chloride and 2 g ZnCl2 to 100 C and bubbles a slow stream on SO2 through for 10 hours.
In the colt trap were SOCl2 and SO2, which he seperated by distillation. The yield of thionyl chloride was 66% based on phthaloyl chloride used.
The problems are obvious. SO2 is low boiling and not very long-resident (to put it mildly) in 200 C anything. So in order to react, SO2 must react
almost instantly while bubbling through the phthaloyl chloride or else react in vapor phase. To my way of thinking, this experiment is like watching a
dog walking on its hind legs, erect. It's not that the gog walks well or poorly but that the dog walks at all.
It would be interesting to rerun this experiment in the manner of the Sinclair Research patent.
SO2 in liquod phase is mixed with phthaloyl chloride and catalyst in a pressure vessel, the vessel is closed and the contents stirred and allowed to
come to room temperature.
SO2 at that point will be gas over liquid. If any reaction occurs the autogenous pressure will drop. If not, the temperature is raised say 25 C, the
new pressure noted and monitored. If no pressure drop occurs then SO2 is not being consumed and SOCl2 is not forming.
Raise the temperature another 25 C. At this point we are close to the bp of thionyl chloride.
I'd hold there for several hours. If still no reaction as indicated by pressure drop I would slowly raise the temperature to Kyrides' 200 C, hold for
10 hrs, then cool the reactor, chill it to 0 C and then allow SO2 to distill off through a column. Then distill out the thionyl chloride. Benzoyl
chloride is next, leaving unchanged phthaloyl chloride and ZnCl2.
If Kyrides is right then we certainly ought to obtain some SOCl2.
But, a much more likely experiment would be to do this with benzotrichloride and anhydrous FeCl3 per Sinclair. This ought to proceed slowly at ambient
temperatures, faster at temperatures still under the bp of SOCl2, and therefore be a lot easier on the autoclave.
Len1 already tried benzotrichloride and ZnCl2 with SO2 and got no SOCl2. Kyrides never claimed that benzotrichloride would work (nor did len1 say he
did.) Sinclair Research's Burk and coworkers found ZnCl2 ineffective with benzotrichloride or HCPX. If Len1 repeated his experiment with anhydrous
FeCl3 or SbCl5 I think he would obtain some SOCl2, although best results would be had in an autoclave with longer reaction time.
[Edited on 23-8-2008 by Sauron]
Sic gorgeamus a los subjectatus nunc.
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
Norman Rabjohn in 1953 JACS proposed a different avenue of attack.
He chloromethylated toluene to a mixture of o- and p-xylyl chlorides then photochemically chlorinated this mixture to both pentachloro-o-xylene and
hexachloro-p-xylene, which are readily seperable.
The work is intresting but while toluene may have some industrial economy versus xylene as a feedstock, I'd just as soon proceed from pure m-xylene or
p-xylene - much more available now than then.
The chloromethylation is done with formalin, conc HCl (aq) and ZnCl2 on toluene at 60 C while introducing anhydrous HCl for 12 hrs.
Chlorination was initiated at 100 C and allowed to rise to 145-150 C which is essentially the IGFarben parameters. Based on the assumption that the
isomer ration was 1:1 for PCOX and HCPX, the yield was claimed to be "essentially quantitative'.
My objection is a lack of much need for PCOX.
[Edited on 23-8-2008 by Sauron]
Attachment: ja01650a066.pdf (369kB) This file has been downloaded 682 times
Sic gorgeamus a los subjectatus nunc.
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
According to Kharasch and H.C.Brown, we can get to tetrachloro-m-xylene very quickly and without chlorine or UV by using 2 mols SO2Cl2 per mol xylene,
initiated by benzoyl peroxide. The reaction time for benzyl chloride from toluene for example was 15 minutes and described as vigorous and nearly
quantitative even in the dark.
This means that only the last chlorines need to be added photochemically.
I like this approach.
Fimally after some tangents here and there we have returned to being faithful to the thread title.
SO2Cl2 -> SOCl2
We make sulfuryl chloride from SO2 and Cl2 (or we buy it where possible.)
We use SO2Cl2 to place 4 of the 6 chlorines onto xylene using benzoyl peroxide initiation, then finish up with UV catalyzed elemental chlorination to
obtain hexachloro-m-xylene.
Note that the peroxide initiated chlorination released SO2, we conserve that as liquid SO2 which we use in final step
We react liquid SO2 and HCMX (or HCMPX) in presence of anhydrous FeCl3, obtaining SOCl2 and isophthaloyl chloride (or terephthaloyl chloride).
So we have converted SO2Cl2 into SOCl2.
The byproduct IPC or TPC can be exploited similarly to benzoyl chloride to prepare volatile acid chlorides per H.C.Brown.
Kyrides used benzotrichloride to prepare phthaloyl chloride, and reacted that with SO2 over ZnCl2 and obtained SOCl2 in 66% yield, benzoyl chloride
being byproduct.
The similarities in his scheme are as striking as the differences.
Both involve a trichloromethylarene.
Both involve SO2.
The catalysts are different and so are the conditions.
Can Kyrides' method be improved to run at lower temperature with use of a better catalyst and perhaps a better yield? Well, perhaps. I think I will
try to reduce the new process to practice and only then see about getting the Kyrides reaction up off its knees.
This looks like fun.
[Edited on 25-8-2008 by Sauron]
Attachment: ja01877a048.pdf (946kB) This file has been downloaded 869 times
Sic gorgeamus a los subjectatus nunc.
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
KYRIDES REDUX
L.P.Kyrides of Monsanto obtained three US patents in 1934 pertaining to the material described in his JACS article I have posted previously.
The first concerns the catalyzed reaction of SOCl2 with phthalic anhydride in the melt (200 C) producing SO2 and phthaloyl chloride. As noted this
reaction is reversible; SO2 bubbled through phthaloyl chloride at same temperature gives a 66% yield of SOCl2. In both cases anhydrous ZnCl2 is named
as preferred catalyst.
The other two deal with reactions of benzotrichloride withacid anhydrides especially phthalic anhydride to give phthaloyl chloride and benzoyl
chloride almost quantitatively.
Kyrides found that by increasing the amount of AnCl2 catalyst he could reduce reaction temperature to 110-120 C. As there are no gaseous products, and
the sole liquid reactant and both products are all high boiling, this reaction can be run in glass with just a reflux condenser and a CaCl2 gyard
tube. This is a very convenient way to obtain the phthaloyl chloride needed to make SOCl2 from SO2.
Len1 has shown how to prepare benzotrichloride. It is also commercially available.
Here are the three Kyrides patents rolled into one pdf. Unlike his JACS paper they have not been posted previously on this forum anywhere I know of.
They make a most interesting read alongside that paper.
[Edited on 28-8-2008 by Sauron]
Attachment: kyridespatents.pdf (362kB) This file has been downloaded 592 times
Sic gorgeamus a los subjectatus nunc.
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
Remember:
Under same conditions (200 C and in presence of ZnCl2)
Forward reaction:
SOCl2 + phthalic anhydride -> phthalyl chloride and SO2
Reverse reaction:
Phthalyl chloride and SO2 -> SOCl2 + phthalic anhydride
Here is the reverse Kyrides reaction to prepare thionyl chloride from SO2 and bhthalyl chloride::
A mixture of 107 g 94% phthalyl chloride and 2 g anhydrous zinc chloride is heated to 200 C. Dry SO2 is bubbled through the heated mixture in a slow
current for ten hours and the distillate collected through a condenser cooled with tap water. In this manner 44.3 g distillate was obtained.
Redistillation obtained 39 g SOCl2 amounting to 66% of the theoretical based on phthalyl chloride.
No attempt was made to analyze unreacted SO2 for thionyl chloride but since the pot residue (phthalic anhydride) was practically chlorine free it is
likely that a substantial amount of SOCl2 was lost with unreacted SO2.
My comments:
I'd either use a dewar condenser to collect the unreacted SO2 along with the thionyl chloride or place a cold trap cooled with dry ice-acetone (or
alcohol) after the main receiver.
I would investigate the effect of increasing the amount of ZNCl2 employed, on the yield.
And I will scale this up to a 1 liter basis, with hope of obtaining 400 g SOCl2 or better in a 10 hr run.
As previously discussed I would run this in a tall form cylinder with gas inlet at bottom and heat by external heating tape. This would extend the
residence time of the SO2 in the phthalyl chloride, compared to the shallow pool of chloride in a RB flask.
By these methods, unreacted SO2 (or SO2 from decomposition of SOCl2) can be used in next run and no SO2Cl2 product lost.
I do not regard 66% as an unworthy yield, but improvement is always nice.
The phthalic anhydride left in post can be reconverted to phthalyl chloride using benzotrichloride, and used to prepare more SOCl2.
If this reaction were done in a sealed system, it would be in equilibrium, but since in both cases one of the products is in vapor phase and is
stripped from the reaction, in an open system both reactions can go to completion.
[Edited on 30-8-2008 by Sauron]
Attachment: Kyrides_phthaloyl.pdf (318kB) This file has been downloaded 888 times
Sic gorgeamus a los subjectatus nunc.
|
|
Klute
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 1378
Registered: 18-10-2006
Location: France
Member Is Offline
Mood: No Mood
|
|
You might want to be carefull with any eventual over pressure using such a cylinder... How do you plan on preventing the melt from entering the gas
inlet? A tap near the inlet?
Do you plan on using some kind of mecanical stirring? I'd love to see a diagramme or picture of the setup when done!
\"You can battle with a demon, you can embrace a demon; what the hell can you do with a fucking spiritual computer?\"
-Alice Parr
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
Overpressure? It's not a closed system.
I'd use the inlet/outlet assy from a wash bottle (gas washing bottle) in a test tube shaped pyrex cyclinder with a round bottom and a standard taper
joint at top to match. I would use a wash bottle, except that I'd be afraid that the flat base would not take the thermal stress.
Nothing exotic about this set up. The phthalyl chloride is not a melt. It's a high boiling liquid. Only at the end of the reaction, when all the
phthaloyl chloride is gone and the phthalic anhydride has replaced it, is it a melt. So when no more SOCl2 collects, it's time to stop SO2 flow and a
good idea to withdraw the inlet tube from the melt. Although clearing out a little phthalic anhydride is not a big deal. I'd use a plain inlet not a
frited one. I hope this is clear?
See Corning 1760. I'd have it made without base, and capacity 1.5 L so it will take a 1 L charge of phthalyl chloride. 29/42 at top so it will accept
one of the inlet/outlet assys I already have.
[Edited on 30-8-2008 by Sauron]
Sic gorgeamus a los subjectatus nunc.
|
|
Klute
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 1378
Registered: 18-10-2006
Location: France
Member Is Offline
Mood: No Mood
|
|
ok, this has cleared things up. I thought phtaloyl chloride was a solid, that can be distilled under vacuum (Kyrides article you joined distill it
IIRC).
Looking forward to hear more!
\"You can battle with a demon, you can embrace a demon; what the hell can you do with a fucking spiritual computer?\"
-Alice Parr
|
|
Sauron
International Hazard
![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif) ![*](images/xpblue/star.gif)
Posts: 5351
Registered: 22-12-2006
Location: Barad-Dur, Mordor
Member Is Offline
Mood: metastable
|
|
Phthaloyl chloride bp 2690271 C at normal pressure. So sitting in an oil bath which is regulated to 200 C it is probably around 75 C below reflux.
In the forward reaction (making phthaloyl chloride from SOCl2 and phthalic anhydride) Kyrides dripped thionyl chloride onto the melt of the anhydride
(200 C) which was admixed with ZnCl2 catalyst. He did not introduce it below the surface, so the reaction was at the surface. SO2 evolved, Kyrides did
not attempt to trap and quantify it. He used 300 g phthalic anhydride and 241 g SOCl2, the theoretical quantity required, adding an average of 24 g
per hour. Temperature 220 C. The ZnCl2 was 1.5 g.
He obtained 380 g product by distillation in vacuo 119-122 C @ 4.5 mm. Of this 22 g seperated out as phthalic anhydride and was filtered. The
phthaloyl anhydride analyzed as 94% pure, balance being still dissolved anhydride. Taking into account the recovered anhydride and dissolved
anhydride, the yield of chloride was practically quantitative on the basis of anhydride used.
Len1 objected to Kyrides' 200 C temperature in the reverse reaction, citing lit. comments that SOCl2 is unstable at such temperatures. But be that as
it may, the essentially quant. yield in this procedure belies his objection. Obviouslt SOCl2 is stable enough for the time needed to react.
In the reverse reaction SOCl2 is formed from SO2 and phthaloyl chloride at 200 C over same temperature. Krides recovered 66% purified SOCl2. Lem 1
apparently ignored or choose to overlook Kyrides comment that he did not try to recover SOCl2 from the SO2 stream but thought that its presence was
significant there. That is why I want to cold-trap that and measure the true yield of SOCl2 by this method. Also doing this on a scale 10X larger
should be instructive.
Making phthaloyl chloride from SOCl2 is now of only academic interest since the benzotrichloride method is better. But making SOCl2 from SO2 and
phthaloyl chloride, that is of practical interest to anyone who wants this product but does not care to make or work with SO3 or oleum.
Sic gorgeamus a los subjectatus nunc.
|
|
Pages:
1
2 |