Sciencemadness Discussion Board

Gamma Alumina Dehydration Of Ethanol To Ethylene

Pantagruel - 11-4-2011 at 06:56

I feel I must first say to please forgive my lack of chemistry knowledge and experience. I have what I feel is a rudimentary understanding of organic chemistry and am trying to learn more on my own.

I'm wanting to play around with ethylene generators and have been reading a few patents, but as I'm sure you all know patents are always intentionally vague. I also do not have journal access and my research skills are not quite what I would like for them to be. The first ethylene generator I'm trying to decode is that which is in these US patents:

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sec...

http://www.freepatentsonline.com/4400291.html


The MSDS for one such ethylene generating states the liquid is predominantly ethanol with 0-10% each of isopropanol and ethyl acetate.

When I used the search engine here it turned up this link.

http://discovery.kcpc.usyd.edu.au//9.2.3-short/9.2.3_Dehydra...

From what I gather, it is a near quantitative conversion and it basically involves simply dripping ethanol on hot activated gamma alumina. However, it would be nice to have journal articles that give actual data like quantities and temperatures and someone knowledgeable, as many of you are, to help provide some guidance.

ScienceSquirrel - 11-4-2011 at 07:06

Unless your want large amounts of ethylene you would be better off making it from ethanol and sulphuric acid.

http://chestofbooks.com/food/beverages/Alcohol-Properties/Et...

Pantagruel - 11-4-2011 at 10:08

Well, I'd like a more constant and steady production without the requirement of constant observation. Slowly dripping it on hot alumina would only produce it at the rate which it is dripped. I didn't look at your link yet but I'm assuming you're talking about following the same method used for H2SO4 EtOH ether synthesis except at a higher temperature. It's hard to imagine using this for predictable steady output and reliably knowing the rate of ethene output. I'd also have to worry about ether drifting about and explosive peroxides. It doesn't sound like something I'd want to leave even briefly unattended.

[Edited on 4-11-2011 by Pantagruel]

ScienceSquirrel - 11-4-2011 at 11:06

I am not sure that I would want to leave an apparatus dripping ethanol on to hot alumina to its' own devices for too long either.
Really if you want a steady flow of ethylene for ripening fruit. etc then cylinders are the way to go.

[Edited on 11-4-2011 by ScienceSquirrel]

smuv - 11-4-2011 at 12:21

I am going to take the middle ground here. Ethylene from dehydration of ethanol IS most conveniently done via vapor phase dehydration. That being said, all this is contingent on having a decent tube furnace capable of ~400c and a setup that can feed ethanol vapor at slight positive pressures. But in terms of output, controllability and purity of gas, vapor phase dehydration cannot be beaten. That being said, dehydration with phosphoric acid is an old laboratory prep that can be realized with very little equipment.

Anyways, when I first started this hobby I used to be absolutely obsessed with all things chemoindustrial and especially ethylene.

Focus on the furnace, for making ethylene (as opposed to ether) catalysts and all that crap don't really matter much (I'll leave some refs anyways). Also, do patent searches, there is more info about this in patents than in journals, particularly older patents. Also depending on the flow rates/pressures you want to operate at, consider looking into patents about preparing spherical alumina particles (this is only for high flow rates at lower pressures).

Though, I cannot reiterate how flammable ethylene is. A balloon full of ethylene lighted with a match, makes a deafening BOOM!

P.S. Ramblings...toluenesulfonic acid loaded on a solid support is another option (I wouldnt consider the related phosphoric acid on coke an option though). Requires lower temps, but more complicated catalyst prep.

Attachment: ethanol_dehydration_alumina.pdf (439kB)
This file has been downloaded 922 times

Attachment: ethanol_dehydration_alumina_2.pdf (880kB)
This file has been downloaded 969 times

Attachment: ethylene_industrial_and_lab_production.pdf (332kB)
This file has been downloaded 1461 times

Attachment: ethylene_phosphoric_acid_overview.pdf (442kB)
This file has been downloaded 837 times

[Edited on 4-11-2011 by smuv]

Pantagruel - 11-4-2011 at 13:56

Quote: Originally posted by ScienceSquirrel  
I am not sure that I would want to leave an apparatus dripping ethanol on to hot alumina to its' own devices for too long either.
Really if you want a steady flow of ethylene for ripening fruit. etc then cylinders are the way to go.

[Edited on 11-4-2011 by ScienceSquirrel]


Well, maybe and I suppose I'd have a tendency to agree, however, there are at least two companies who sell catalytic ethylene generators for exactly that purpose. At least in the UK, ethylene generators are the preferred method of enrichment from what I've read. Perhaps I'm misinterpreting exactly how these ethylene generators work and they use a method similar to what smuv is suggesting? The patents I've seen thus far seem to suggest otherwise though.


Smuv, thank you very much for providing some literature for me to look through. I'm really liking this forum.

[Edited on 4-11-2011 by Pantagruel]

ScienceSquirrel - 11-4-2011 at 17:17

It is possible to build a safe unattended ethylene generator that uses ethanol and a catalyst but I suspect that they have a lot of computer control and detectors so a shut down occurs if something nasty happens.
It depends on your resources; if you are able to build custom systems then go ahead.
If you are doing what most home chemists do and stitch things together from plumbing parts then maybe not.

smuv - 11-4-2011 at 19:46

Why don't you specifically tell us your application?

I agree SS, for a safe autonomous system a lot is required. But if all that is needed is a device that can be left in steady state while the operator is free to focus on other things while staying nearby, no computer automation is required. Though arguably the same could be achieved with well setup condensed phase dehydrations.

watson.fawkes - 12-4-2011 at 05:27

Quote: Originally posted by ScienceSquirrel  
It is possible to build a safe unattended ethylene generator that uses ethanol and a catalyst but I suspect that they have a lot of computer control and detectors so a shut down occurs if something nasty happens.
Such systems have both sensor and actuators. It's the actuators where the bigger costs are. It's much cheaper to set up a system that raises an alarm for manual shutdown than to set up the system to performs a shutdown.

The computer control, at this point, is cheaper than the sensors.

smuv - 12-4-2011 at 11:45

I don't understand, If you can raise an alarm, why can't you just instead actuate an optotriac-> triac and turn off the system. Thats only 5 bux worth of components.

But anyways, for a long time, I have thought about taking a smoke detector and interfacing it with essentially a relay, to kill power to my lab, in the event of a fire. That being said, in most cases, if there is a fire, you are already screwed no matter if power is on or off (especially when you are making ethylene).

ScienceSquirrel - 12-4-2011 at 12:45

If Pantagruel tells us the scale he is thinking of working on and the degree of unattended working, etc that he expects from his apparatus then we can come to some sensible suggestions.
He could be planning on pumping ethylene in to a big greenhouse to ripen tomatoes and if he gets it wrong he might be making a giant gas bomb.

watson.fawkes - 12-4-2011 at 20:32

Quote: Originally posted by smuv  
I don't understand, If you can raise an alarm, why can't you just instead actuate an optotriac-> triac and turn off the system. Thats only 5 bux worth of components.
Turning off a gas-phase reaction means having solenoid-activated valves to shut off supply and to purge the reaction vessel. Just turning off power to, say, a heater, doesn't by itself put the reactor into a safe state; that takes work.

smuv - 12-4-2011 at 21:00

Watson, +5 years ago I'd agree, but with a microcontroller things like this are a cakewalk. If you can do chemistry, you can program an arduino no problem.

The fact that many people in the home distilling community have setup automated and continuous stills/fermenters is really proof of point.



[Edited on 4-13-2011 by smuv]

watson.fawkes - 13-4-2011 at 04:38

Quote: Originally posted by smuv  
Watson, +5 years ago I'd agree, but with a microcontroller things like this are a cakewalk. If you can do chemistry, you can program an arduino no problem.
I've never argued that the controller is the problem. It's not the controllers where the money is, it's not the sensors (mostly), it's the actuators. The issue is not control when things go OK. It's automatic response when they don't. Response means some thing has to respond, and that means mechanical response. Have you priced solenoid-controlled relays? They're not cheap. You can occasionally scrounge good deals, but the rate of those coming through isn't particularly high. And if you need special materials (like stainless steel valve bodies), forget it.

smuv - 13-4-2011 at 07:59

I guess our experiences just differ. :)

But let me say:

I have controlled 1.5kw AC loads with a microcontroller controlled triac, both ON/OFF and 256 step dimming. Cost per channel to do dimming ~$6

I have interfaced a thermocouple with a microcontroller using a maxim chip that does cold junction correction. The thermocouple costs more than the chip.

I have controlled a servo with a microcontroller. And god knows I have bought TOO many SS valves from ebay. Though I have not tried to make my own manually actuated valve (yet).

I would say I am 3/4 to full automation, and I certainly haven't spent much. I have invested some time but all things considered, really not that much. But anyways, I like you watson, I don't want to argue and have bad blood between us, but you have to understand what you are saying kinda goes against the ethos of a project/dream of mine; so I can't avoid disagreeing. :)

watson.fawkes - 13-4-2011 at 09:58

Quote: Originally posted by smuv  
I would say I am 3/4 to full automation, and I certainly haven't spent much. I have invested some time but all things considered, really not that much. But anyways, I like you watson, I don't want to argue and have bad blood between us, but you have to understand what you are saying kinda goes against the ethos of a project/dream of mine; so I can't avoid disagreeing. :)
We're mostly in agreement. Half my point, really, is that when you tally up the costs for building a fully automated rig, that the mechanical actuators are very likely going to be the biggest line item in the budget. And this seems be exactly because they're not made of electronics. The other half of my point is that if you want a fully automated system to be safe, you will need these electro-mechanical parts. In many cases, this is true; in all the cases I've thought about that involve fire hazard, it's definitely true.

Back to the original point, then, it seems like a good compromise to build an alert system first, and then build fail-safes later. This reduces the cost of entry significantly, because you don't really need electro-mechanical parts for the alert system.

I don't want to pose this as a blocker, to say that these kinds of ideas are impractical. They're extraordinarily practical, and cheap micro-controllers are central to it. I'm all over micro-scale chemical plants, which to me means they'll fit within a canonical 4' x 8' footprint in a garage. But there is a reasonably big jump in price at a certain point in the automation process.

S.C. Wack - 13-4-2011 at 14:26

Quote: Originally posted by Pantagruel  
When I used the search engine here


Quote: Originally posted by Pantagruel  
it would be nice to have journal articles that give actual data


It's all here somewhere.
http://www.sciencemadness.org/talk/viewthread.php?tid=1959&a...
(((Computer access for the public is much more restrictive at the library now since back then, in 2005.)))
Quote: Originally posted by ScienceSquirrel  
I am not sure that I would want to leave an apparatus dripping ethanol on to hot alumina to its' own devices for too long either.


Assuming that the alumina long remains active in that setup, with the ethanol's purity and the other conditions used, I don't see why not as long as you fill the dripper every day. High and low temperature limits to open the furnace circuit would be nice to have, and I think my Watlow 988 (that I'd be using as furnace control anyways) will do this. I'd have to test it to see if the alarm would reset if the power went off, then came back on.