HTML by Rhodium
The separation or purification of liquids by vaporization and condensation is a very important step in one of our oldest professions, The word "still" lives on as a tribute to the importance of organic chemistry. There are two important points here.
Remember these. They show up on quizzes. But when do I use distillation? That is a very good question. Use the guidelines below to pick your special situation, and turn to that section. But you should read all the sections anyway.
Remember, these are guides. If your compound boils at 150.0001°C, don't scream that you must do a vacuum distillation or both you and your product will die. I expect you to have some judgment and to pay attention to your instructor's specific directions.
For separation of liquids boiling below 150°C at 1 atm from
If one of the components boils below 70°C and you use a Bunsen burner, you may have a hard time putting out the fire. Use a steam bath or a heating mantle. Different distillations will require different handling (see Chapter 18, "Sources of Heat"). All the distillations always require heating, so Chapter 18 is really closely tied to this section. This goes for enlightenment on the use of boiling stones and clamps as well (see Chapter 17, "And Now-Boiling Stones" and Chapter 19, "Clamps and Clamping").
If there is any one place your setup will fall apart, here it is (Fig. 99). When you set up the jointware, it is important that you have all the joints line up. This is tricky, since, as you push one joint together, another pops right out. Remember,
All joints must be tight!
Choose a distilling flask carefully. If it's too big, you'll lose a lot of your product. If it's too small, you might have to distill in parts. Don't fill the distilling flask more than half full. Less than 1/3 full and you'll probably lose product. More than 1/2 full and you'll probably have undistilled material thrown up into the condenser (and into your previously clean product). Fill the distilling flask with the liquid you want to distill. You can remove the thermometer and thermometer adapter, fill the flask using a funnel, and then put the thermometer and its adapter back in place.
If you're doing a fractional distillation with a column (a Class 3 distillation), you should've filled the flask before clamping the setup. (Don't ever pour your mixture down a column. That'll contaminate everything!) You'll just have to disassemble some of the setup, fill the flask, reassemble what you've taken down, and pray that you haven't knocked all the other joints out of line.
Put in a boiling stone if you haven't already. These porous little rocks promote bubbling and keep the liquid from superheating and flying out of the flask. This flying around is called bumping. Never drop a boiling stone into hot liquid, or you may be rewarded by having your body soaked in the hot liquid as it foams out at you.
Make sure all the joints in your setup are tight. Start the heat s-l-o-w-l-y until gentle boiling beg-ins and liquid starts to drop into the receiving flask at the rate of about 10 drops per minute. This is important. If nothing comes over, you're not distilling, but merely wasting time. You may have to turn up the heat to keep material coming over.
Read all about it. Ways of having fun with thermometer adapters have been detailed (see text accompanying Figs. 22-26 in Chapter 4, "Jointware").
A word about clamps. Use! They can save you $68.25 in busted glassware (see Chapter 19, "Clamps and Clamping").
Make sure the entire thermometer bulb is below the side arm of the three-way adapter. If you don't have liquid droplets condensing on the thermometer bulb, the temperature you read is nonsense. Keep a record of the temperature of the liquid or liquids that are distilling. It's a check on the purity. Liquid collected over a 2°C range is fairly pure. Note the similarity of this range with that of the melting point of a pure compound (see Chapter 12, "The Melting Point Experiment").
Always keep cold water running through the condenser, enough so that at least the lower half is cold to the touch, Remember that water should go in the bottom and out of the top (Fig. 98). Also, the water pressure in the lab may change from time to time and usually goes up at night, since little water is used then. So, if you are going to let condenser cooling water run overnight, tie the tubing on at the condenser and the water faucet with wire or something. And if you don't want to flood out the lab, see that the outlet hose can't flop out of the sink.
It is important that the tubing connector remain open to the air; otherwise, the entire apparatus will, quite simply, explode.
Do not just stick the vacuum adapter on the end of the condenser and hope that it will not fall off and break.
This is foolish. I have no sympathy for anyone who will not use clamps to save their own breakage fee. They deserve to pay.
The receiving flask should be large enough to collect what you want. You may need several, and they may have to be changed during the distillation. Standard practice is to have one flask ready for what you are going to throw away and others ready to save the stuff you want to save.
Why everyone insists on loading up a bucket with ice and trying to force a flask into this mess, I'll never know. How much cooling do you think you're going to get with just a few small areas of the flask barely touching ice? Get a suitable receptacle-a large beaker, enameled pan, or whatever. It should not leak. Put it under the flask. Put some water in it. Now add ice. Stir. Serves four.
Ice bath really means ice-water bath.
Say you place 50 mL of liquid A (bp. 50°C) and 50 mL of liquid B (bp 100°C) in a 250 mL R.B. flask. You drop in a boiling stone, fit the flask in a distillation setup, and turn on the heat. Bubbling starts, and soon droplets form on the thermometer bulb. The temperature shoots up from room temperature to about 35°C, and a liquid condenses and drips into the receiver. That's bad. The temperature should be close to 50°C. This low-boiling material is the forerun of a distillation, and you won't want to keep it.
Keep letting liquid come over until the temperature stabilizes at about 49°C. Quick! Change receiving flasks now!
The new receiving flask is on the vacuum adapter, and the temperature is about 49°C. Good. Liquid comes over, and you heat to get a rate of about 10 drops per minute collected in the receiver. As you distill, the temperature slowly increases to maybe 51°C and then starts moving up rapidly.
Here you stop the distillation and change the receiver. Now in one receiver you have a pure liquid, bp. 49-51°C. Note this boiling range. It is just as good a test of purity as a melting point is for solids (see Chapter 12, "The Melting Point Experiment").
Always report a boiling point for liquids as routinely as you report melting points for solids. The boiling point is actually a boiling range and should be reported as such:
If you now put on a new receiver, and start heating again, you may discover more material coining over at 50°C! Find that strange? Not so. All it means is that you were distilling too rapidly and some of the low-boiling material was left behind, It is very difficult to avoid this situation, Sometimes it is best to ignore it, unless a yield is very important. You can combine this "new" 50°C fraction with the other good fraction. For liquid B, boiling at 100°C, merely substitute some different boiling points and go over the same story.
OK, you set all this stuff up to do a distillation. Everything's going fine. Clamps in the right place. No arthritic joints, even the vacuum adapter is clamped on, and the thermometer is at the right height. There's a bright golden haze on the meadow, and everything's going your way. So, you begin to boil the liquid. You even remembered the boiling stone, Boiling starts slowly, then more rapidly. You think, "This is it!" Read that temperature, now. Into the notebook: "The mixture started boiling at 26°C."
And you are dead wrong. What happened? Just ask-
Is there liquid condensing on the thermometer bulb?? No!
So, congratulations, you've just recorded the room temperature. There are days when over half of the class will report distillation temperatures as "Hey I see it start boiling now" temperatures. Don't participate. Just keep watching as the liquid boils. Soon, droplets will condense on the thermometer bulb. The temperature will go up quickly, and then stabilize. Now read the temperature. That's the boiling point. But wait! It's not a distillation temperature until that first drop of liquid falls into the receiving flask.
For separation of liquids boiling above 150°C at 1 atm from
Why vacuum distill? If the substances boil at high temperatures at 1 atm, they may decompose when heated. Putting a vacuum over the liquid makes the liquid boil at a lower temperature. With the pressure reduced, there are fewer molecules in the way of the liquid you are distilling. Since the molecules require less energy to leave the surface of the liquid, you can distill at a lower temperature, and your compound doesn't decompose.
If you want to measure the pressure in your vacuum distillation setup, you'll need a closed-end manometer, There are a few different types, but they all work essentially the same way. I've chosen a "stick" type (Fig. 101). This particular model needs help from a short length of rubber tubing and a glass T to get connected to the vacuum distillation setup.
Suppose, by luck of the draw, you've had to prepare and purify 1-octanol (bp. 195°C). You know that, if you simply distill 1-octanol, you run the risk of having it decompose, so you set up a vacuum distillation. You hook your setup to a water aspirator and water trap, and you attach a closed-end "stick" manometer. You turn the water for the aspirator on full blast and open the stick manometer. After a few minutes, nothing seems to be happening. You pinch the tubing going to the vacuum distillation setup (but not to the manometer), closing the setup off from the source of vacuum. Suddenly, the mercury in the manometer starts to drop. You release the tube going to the vacuum distillation setup, and the mercury jumps to the upper limit. You have air leaks in your vacuum distillation setup.
Air leaks can be difficult to find. At best, you push some of the joints together again and the system seals itself. At worst, you have to take apart all the joints and regrease every one. Sometimes you've forgotten to grease all the joints. Often a joint has been etched to the point that it cannot seal under vacuum, when it is perfectly fine for other applications.
You've found all the leaks, and the pressure in your vacuum distillation setup is, say, 25 torr. Now you need to know the boiling point of your compound, 1-octanol, this time at 25 torr, and not 760 torr. You realize it'll boil at a lower temperature, but just how low? The handy nomographs can help you estimate the new boiling point.
This time you have the boiling point at 760 torr (195°C) and the pressure you are working at (25 torr). Using Fig. 102a, you
So a liquid that boils at 195°C at 760 torr will boil at about 95°C at 25 torr. Remember, this is an estimate. Now suppose you looked up the boiling point of 1-octanol and all you found was: 9819. This means that the boiling point of 1-octanol is 98°C at 19 torr. Two things should strike you.
Now we have a case of having an observed boiling point at a pressure that is not 760 torr (1-octanol again; 98°C at 19 torr). We'd like to get to 25 torr, our working pressure. This requires a double conversion, as shown in Fig. 102b.
So, we've estimated the boiling point to be about 105°C at 25 torr. The last time it was 95°C at 25 torr. Which is it? Better you should say you expect your compound to come over at 95-105°C. Again, this is not an unreasonable expectation for a vacuum distillation.
The pressure-temperature nomograph (Fig. 103) is really just a simple, graphical application of the Clausius-Clapeyron equation. If you know the heat of vaporization of a substance, as well as its normal boiling point, you can calculate the boiling point at another temperature. You do have to assume that the heat of vaporization is constant over the temperature range you're working with, and that's not always so. Where's the heat of vaporization in the nomograph? One is built into the slope and spacings on the paper. And, yes, that means that the heat of vaporization is forced to be the same for all compounds, be they alkanes, aldehydes, or ethers. So do not be surprised at the inaccuracies in this nomograph, be amazed that it works as well as it does.
A Claisen adapter so you can vacuum
distill and take the temp simultaneously
Multipurpose setup with a three-neck flask
having one neck stoppered
For separation of liquids that are soluble in each other, that boil fewer than 25°C from each other, use fractional distillation. This is like simple distillation with the changes shown (Fig. 106).
Fractional distillation is used when the components to be separated boil within 25°C of each other. Each component is called a fraction. Clever where they get the name, eh? This temperature difference is not gospel. And don't expect terrific separations either. Let's just leave it at close boiling points. How close? That's hard to answer. Is an orange?
That's easier to answer. If the experiment tells you to "fractionally distill," at least you'll be able to set it up right.
If one distillation is good, two is better. And fifty, better still. So you have lots and lots of little, tiny distillations occurring on the surfaces of the column packing, which can be glass beads, glass helices, ceramic pieces, metal chips, or even stainless-steel wool.
As you heat your mixture it boils, and the vapor that comes off this liquid is richer in the lower boiling component. The vapor moves out of the flask and condenses, say, on the first centimeter of column packing. Now, the composition of the liquid still in the flask has changed a bit - it is richer in the higher boiling component. As more of this liquid boils, more hot vapor comes up, mixes with the first fraction, and produces a new vapor of different composition - richer yet in the more volatile (lower boiling) component. And guess what? This new vapor condenses in the second centimeter of column packing. And again, and again, and again.
Now all these are equilibrium steps. It takes some time for the fractions to move up the column, get comfortable with their surroundings, meet the neighbors... And if you never let any of the liquid-vapor mixture out of the column, a condition called total reflux, you might get a single pure component at the top - namely, the lower boiling, more volatile component all by itself! This is an ideal separation.
Fat lot of good that does you when you have to hand in a sample. So, you turn up the heat, let some of the vapor condense, and take off this top fraction. This raises hell in the column. Nonequilibrium conditions abound - mixing. Arrrgh! No more completely pure compound. And the faster you distill, the faster you let material come over, the higher your throughput - the worse this gets. Soon you're at total takeoff, and there is no time for an equilibrium to get established. And if you're doing that, you shouldn't even bother using a column.
You must strike a compromise. Fractionally distill as slowly as you can, keeping in mind that eventually the lab does end. Slow down your fractional distillations; I've found that 5-10 drops per minute coming over into the receiving flask is usually suggested. It will take a bit of practice before you can judge the best rate for the best separation.
But you have to watch out for the deadly azeotropes.
Once in a while, you throw together two liquids and find that you cannot separate part of them. And I don't mean because of poor equipment, or poor technique, or other poor excuses. You may have an azeotrope, a mixture with a constant boiling point.
One of the best known examples is the ethyl alcohol-water azeotrope. This 96% alcohol - 4% water solution will boil to dryness, at a constant temperature. It's slightly scary, since you learn that a liquid is a pure compound if it boils at a constant temperature. And you thought you had it made.
There are two types of azeotrope. If the azeotrope boils off first, it's a minimum-boiling azeotrope. After it's all gone, if there is any other component left, only then will that component distill.
If any of the components come off first, and then the azeotrope, you have a maximum-boiling azeotrope.
You should be able to see that you have to be really careful in selecting those chaser or pusher solvents mentioned. Sure, water (bp 100°C) is hot enough to chase ethyl alcohol (bp 78.3°C) from any column packing. Unfortunately, water and ethyl alcohol form an azeotrope and the technique won't work. (Please see Chapter 34, "Theory of Distillation.")
Mixtures of tars and oils must not dissolve well in water (well, not much, anyway), so we can steam distill them. The process is pretty close to simple distillation, but you should have a way of getting fresh hot water into the setup without stopping the distillation.
Why steam distill? If the stuff you're going to distill is only slightly soluble in water and may decompose at its boiling point and the bumping will be terrible with a vacuum distillation, it is better to steam distill. Heating the compound in the presence of steam makes the compound boil at a lower temperature. This has to do with partial pressures of water and organic oils and such.
There are two ways of generating steam: externally and internally.
In an external steam distillation, you lead steam from a steam line, through a water trap, and thus into the system. The steam usually comes from a steam tap on the bench top. This is classic. This is complicated. This is dangerous.
You can use many types of steam traps with your distillation setup. I've shown two (Fig. 107a/107b), but these are not the only ones, and you may use something different. The point is to note the steam inlet and the trap drain, and how to use them.