Sciencemadness Discussion Board

Soxhlet extraction temperature HELP

Olioli - 29-5-2012 at 01:40


How does different temperature affects the extraction efficiency of oil from oilseed using a soxhlet extraction for solvent extraction.
Since the solvent vapour condensers (about room temperature or below boiling point) unto the porous thimble holding the sample. So how does the temperature affect the extraction efficiency?

peach - 29-5-2012 at 02:16

The hotter anything is, the faster the molecules of it are moving and so the faster it tends to diffuse. At the same time, the hotter something is, the more likely it is to decompose or rearrange it's self.

The temperature is largely fixed by the BP of the solvent it's self and partly by how cold the condenser / room is. You'd want to pick a solvent based more on it's polarity since most of the volatile solvents are going to have roughly similar BP's anyway.

Olioli - 29-5-2012 at 18:35

I know that increasing the temperature directed to heating the solvent will cause the solvent to evaporate faster, but when it reaches the condensor, the temperature of the solvent will be back to about room temperature right?

So temperature only affects how fast the solvent evaporate in the round bottom flask/distillation flask?

peach - 30-5-2012 at 15:28

[Olioli, I've written this post to include explanations and some generic rules. I'm not suggesting you don't know any of it, it's only to make it of wider use.]

In regards to the power setting on the hotplate / mantle to, you'd want that set low.

The temperature of the solvent in the extractor isn't changed by the mantle setting, no.

Inside the extractor it's self, thinking about what's happening. The solvent boils at it's BP, flows up and then touches the condenser. As soon as the vapour drops below the BP, it condenses and runs down to the extractor. So the solvent will be coming off the condenser and falling back to the extractor close to it's BP. It will be a little cooler still because it takes time for drops to form on the condenser and time for them to run down; during which time they're in contact with the cool condenser.

The extractor will be above room temperature and below the solvent's BP. The precise temperature depends on a multiple of things, e.g.;
  1. How big the extractor is
  2. Room temperature
  3. Solvent BP and enthalpy
  4. Condenser temperature
  5. Mass loaded in extractor

But your mantle setting doesn't affect the extractors temperature, only the rate of solvent reboiling (and this is limited by other factors, explained below).

Soxhlets are great because they require so little human interference, they can be left alone to chug away all day, gently extracting the material into a really small volume of solvent. Their only drawback being that they involve the prolonged application of (albeit it low) heat when they're not sub-zero boiling solvents.

There is a limit to how fast solvent can boil off, set by the surface area of it in the flask below and the size of the glass joints. If you use a small (say 50ml) flask, and small glassware (say, 14/20 joints), you will be pretty much at the maximum as soon as the solvent is boiling. We have to be very careful of using high mantle and hotplate settings for reflux / distillation / extraction. Once the solvent is boiling, that's usually good enough. Whilst the boiling point of the solvent doesn't change much over the coarse of the hours, the glass and material directly in contact with it does.

Say my solvent begins to boil when I set the variac to 2, or say 50C on the front of the hotplate. If I set the variac to 3 or mantle to 60C, the solvent may boil off a little quicker because it's getting more heat energy; there is more power going into the solvent to push it through evaporation, but the temperature at which it's evaporating isn't changing.

Now say I set the mantle to 6, or hotplate to 150C, the solvent continues to boil around it's original temperature and will already be boiling off about as fast as it can. But the glass walls of the flask below will be a lot hotter than they need to be. The solvent, and more importantly the thing dissolved in it, are constantly in contact with those walls. The BP of the solvent is a 'bulk' property, that assumes it's all one thing and at one temperature throughout. But the solvent in the hot flask of your extractor is not. It will increasingly contain my extracted organic material, and the walls of the flask may be a lot hotter than the BP of the solvent.

The result is that, over the hours and hours or full days worth of running the extractor, the organic material I'm extracting may be at least partially decomposing as it touches the scorching hot walls of the hot flask. So, by running the process at a far higher setting than is needed to start the solvent boiling, I stand a substantial risk of getting 'muckier' results. This would be particularly true if I was extracting anything that is remotely sensitive.

We can think of it like frying something in the kitchen. Cooking oil boils at a really high temperature. But, before it's even boiling, it's already hot enough to scorch (brown / burn) most things that are left in it for too long. Setting a frying pan on a burner, getting the oil boiling, then increasing it even further, it will incinerate any ingredients on contact. Our goal in using a soxhlet is to do the opposite; get it out gently. This is partly why CO2 extraction is used commercially; the source material and extract only need to ever reach around room temperature, meaning there is a decrease risk of damaging the extract.

Once we are used to various solvents, various bits of glass and your heat sources, we will become accustom to setting it to around the BPs straight off. E.g. I know that to just about boil most volatile solvents between 60 and 80C, I need to set my mantle to between 1 and 2. So I can set it to 1 or 1.5, wander off, come back and check it's started.

Far, far, far too many people will set things to 9 or 10, or 'full / maximum / hottest', assuming it's fine because the BP is fixed, then leave it like that. It's not okay. This is a reason why people can end up with seriously messed up results; purely due to using the maximum heat settings.

My suggestion (finding the minimum):
  1. Set up your extractor (do this minus any organic material but with solvent present)
  2. Pick a suitably sized heat source and setup
  3. Turn the heat source on at the lowest possible setting and get on with something else for an hour
  4. Check back and look to see if there are any signs of bubbles moving in the solvent. If there aren't, nudge the setting up very slightly (e.g. if it's numbered 1,2,3,4 etc and you're on 1, go to 1.2) , get on with something else for 30-60 minutes, check back
  5. When you can see vapour moving up to the condenser and condensate falling down to the extractor, mentally note this setting. Try decreasing it ever so slightly towards the last setting you were at and see if the process stops. If it does, you will now have a rough mental idea of what's just about enough to get it going. To give an idea of 'rough', if it's numbered 1,2,3,4 etc to ten, a 'rough' setting would be half one of those (say, somewhere between 1.5 and 2)

Pushing it:
  1. With the solvent just boiling away on that minimum setting, time roughly how long each of your syphon cycles are taking
  2. Saying your mantle has settings labelled as 1,2,3,4 etc and your cycle has started at 1.5 try increasing it to 2
  3. Allow the process to come into thermal stability (half an hour to an hour), and then time a syphon cycle
  4. Try going to 2.5, half an hour to an hour, time another cycle

What you will find is that constantly increasing the temperature has a diminishing effect on cycle time; the solvent simply can't boil off any quicker.

It's important to leave the thing alone for at least half an hour between changes when finding the settings. It should be more like an hour. Read something interesting, tidy up etc. It really does take about that long to stabilise. Once the rough settings are known, it becomes much easier to go straight to those points in the future. It can actually be much easier, quicker and more gentle to do this manually with prior knowledge as opposed to letting a really expensive digital controller do it (which can take three hours to warm up every time, overshoot or sway back and forth around the set temperature).

For virtually every application, be it reflux, extraction or distillation, the results are always better and the process quicker and easier in the long run if the heat settings are closer to the minimum required to effect boiling. This information is basically generic to practical chemistry, and would make for an entry in a stickied advice thread, in my opinion. It needs ramming home even to university students (see example below).

By increasing your mantle temperature, it may decrease the cycle time slightly (with diminishing effect). It won't increase the temperature in the extractors body. It will accelerate the decomposition of the organic material in the hot flask. As the cycling process is automatic and can be left to run unattended, it makes more sense to use a lower temperature, let it run for a few extra cycles and yield less brown organic material.

How to increase the extraction efficiency / rate without increasing the heat input:

Easy, wrap the extractors body with a layer of glass wool (house insulation works fine). This will keep the extractor it's self closer to the BP of the solvent, without the scorching hot flask below. You only need to wrap the section where the sample sits. If you want to be really gentle with the heating, you could also wrap the top of the hot flask and the section the vapours travel up through.

Here's a similar point being made by a Mr. Derek Lowe (PhD Organics, drug discovery researcher), with regards to a solvent distillation:

Summer students are showing up at academic and industrial labs around the country right about now. A certain percent of them will blow something up within the next three months, and that percent will be several standard deviations above the ka-boom rate of the other lab members. I'm not trying to say mean things about summer students. I merely speak the truth.

I had a summer undergraduate working with me for a while in grad school, and he taught me several lessons, of varying utility. One day he needed some dry benzene for a reaction, so I helped him set up a still in my hood. One-liter round-bottom flask, some benzene, a little sodium. My intern, who I'll refer to as Toxic John, put a heating mantle on the thing and turned it on.

A little while later, I walked past my hood and noticed that the stuff was boiling merrily. A bit too merrily, actually - it was really hopping around in there. I turned down the Variac (basically a big dimmer-switch type AC transformer that's used to step down the voltage to equipment like heaters) and went on my way. But I came back a little while later, and it was still rolling away in there.

If anything, it was worse. I turned down the Variac again, wondering just what was going on, and why my guess about the inital setting had been so wrong. A few minutes later, things hadn't improved much. The benzene was really leaping around, splattering and erupting. I looked a little more closely at the Variac this time, and noticed something that had escaped me: the heating mantle wasn't plugged into the damn thing at all.

Nope, it was plugged right into the wall socket, as some of my experienced readers will have guessed. As soon as I noticed that, I dropped the lab jack that was holding the heating mantle, which gave me a good look at the glowing red coils showing through the woven glass lining. I could feel it on my face like a sun lamp. Cursing, I pulled the thing out of my hood and heaved it into the hallway, right into a shopping cart that we kept out there for visits to the stockroom.

I went looking for Toxic John as the mantle popped and clicked. It was cooling down, but I wasn't. "John!", I shouted (I was pretty crabby back in grad school), "you plugged the mantle into the wall! No wonder it looked like a volcano in there!"

"What's the matter," he asked me. "Benzene doesn't burn, does it?" "Doesn't. . .burn. . ." I said slowly, as a nearby post-doc put a warning hand on my shoulder. "Well," said John, making his case, "it's inert to bromination!" That line of reasoning didn't impress me much, and as I recall, I told him that if he had any more insights like that we were going to find out if he was inert to bromination himself. Then I went off looking for the professor who'd just taught him sophomore organic chemistry, to let him know that his work, once again, had been in vain.

This solvent still was designed specifically to get the solvent in the still head closer to room temperature. The coil from the condenser drops down into the pool of solvent condensing in the head, cooling it to the condensers own temperature. If that lower part of the coil isn't there, the pool of solvent tends to be warm. This 'coil within the condensed solvent' is missing from a soxhlet, as the solvent should be warm in that application.

[Edited on 31-5-2012 by peach]