3
$\begingroup$

I'm not a chemist by trade so my experience in the lab has probably not been as in-depth as most of you. I've been tasked with investigating a proposed reaction in which an organic mixture (mostly formic acid) is to be heated at reflux for several hours.

The complication being that a pressure of approx. 4 bar is required during the reaction. How would one go about achieving this? The reflux condensors I've browsed through are all open to atmosphere on one end.

Is there some sort of pressure regulator I can attach on the output of the condensor to maintain the pressure at 4 bar?

I'm told that the reaction may produce $\ce{CO2}$ and $\ce{NH3}$ by-products which means pressure is expected to increase and fluctuate as the reaction proceeds. Ideally I'd like to occasionally vent these to drive the reaction equilibrium forward.

The formic acid is quite corrosive to anything not plastic or glass.

$\endgroup$
  • $\begingroup$ In a mostly formic acid solution I can't imagine that much $\ce{NH3}$ is going into the gas phase. Essentially all the $\ce{NH3}$ will be converted into $\ce{NH^+4}$ by the formic acid. // Would venting the $\ce{CO2}$ drive the reaction to completion? $\ce{CO2}$ isn't very soluble in an acidic solution. $\endgroup$ – MaxW May 2 at 19:02
  • $\begingroup$ Another question. Do you need just a limited quantity, or is this for industrial production? $\endgroup$ – MaxW May 2 at 19:30
  • 1
    $\begingroup$ Conceptually 4 bar isn't a problem. I have no idea what kind of setup is available commercially, but maybe some sort of teflon lined stainless steel with teflon gaskets. A chemist can work out the experimental conditions, but to build the reactor for industrial scale production you'll need the expertise of a chemical engineer. $\endgroup$ – MaxW May 2 at 19:42
  • 1
    $\begingroup$ @MaxW you are most likely correct, most NH3 would most likely end up as ammonium formate, the low solubility of CO2 would need to be vented occasionally to drive the reaction forward Im required to only make a small quantity in the lab to prove the reaction and determine what kind of yields are possible, the industrial scale would then be handled by a chemical engineering team $\endgroup$ – SimpleJack May 3 at 5:44
3
$\begingroup$

You could theoretically boil at the higher temperature but why would that be a requirement? Why would you require reflux specifically if you can actually heat the formic acid to higher temperatures under pressure? Heating should work on its own and degassing to drive the equilibrium should work with a back pressure regulator. However, most lab glassware will not take that pressure. You normally need some stainless steel pressure equipment. One way might be to use plug flow reactors rather than bombs.

$\endgroup$
  • $\begingroup$ I cooked a number of things under reflux since the reaction rate was slow. The 4 bar limit might be some limit imposed by the equipment. In other words to get the reaction to go fast you'd need maybe 20 bar, but the equipment could only handle 4 safely. // I'd agree that ordinary glassware won't take 4 bar safely. $\endgroup$ – MaxW May 2 at 19:35
  • 1
    $\begingroup$ Reflux is just a condition that basically means, "heat to the boiling point and ensure the liquid can condense back into the flask/reactor etc." Heating to reflux is not that informative either, especially if you have azeotropes. Operating under atmospheric pressure limits you to the boiling point of the mixture. Working under pressure does not (depending on the temperature you need of course). $\endgroup$ – Beerhunter May 2 at 19:44
  • 2
    $\begingroup$ But the reflux instruction or requirement under pressure serves no purpose, and certainly not in most glassware, especially no need for condensers $\endgroup$ – Beerhunter May 2 at 19:47
  • 2
    $\begingroup$ The OP's task, given as written, serves no purpose. They just need to heat so that nothing boils off at 4 bar or above and use a gas-liquid separator. It's not a standard procedure for regular lab glassware. Specialist kit is needed $\endgroup$ – Beerhunter May 2 at 20:06
  • 1
    $\begingroup$ @Beerhunter - Given that CO2 would be created and must have pressure relief, just releasing gas would release CO2 saturated with formic acid. Having a pressure relief above the condenser would reduce the amount of formic acid in the gas significantly. $\endgroup$ – MaxW May 4 at 13:49
0
$\begingroup$

First of all I would consider the necessity, price and complexity of the pressure and try to carry out the reaction in standard laboratory glassware for longer time (checking the content of the desired product every duration period pointed in your instruction, e.g. 3 hours, 6 hours, etc.). If you get a good yield at ambient pressure this may save your company some money and dealing with pressure equipment at the price of little money and a working day in a simple experiment.

If you have to stick to the elevated pressure at laboratory scale you could use a vertical 1 - 1 1/2 inch quartz tube with the lower end welded or PTFE tube in a steel sleeve. I would recommend you to use a 1-1.25 m long tube but not less to have enough space for the vapor. I would not suggest to fill the tube more than 1/8 of its total height(=volume). Do not forget to cover the bottom of the reactor vessel with broken glass/ceramic balls to avoid sprinkling.

Please make sure to take precaution measures in case the reactor ruptures, which could not be excluded. Carrying out the experiment in a fume chamber and having some non-burning mesh/cloth around it is a must/standard laboratory technique. Another standard safety measure is to have weak top cover so in case something goes wrong the risk of injuring something/somebody around will be diminished.

$\endgroup$
  • $\begingroup$ Trying to make some kind of homemade rig is a huge safety problem in my mind. There must surely be some company that sells such reactor vessels for industrial research. // The OP made it quite clear that the vessel needed to be pressurized to test the feasibility of the full size industrial setup. $\endgroup$ – MaxW May 6 at 0:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.