I have seen Graham condensers set at an angle (perhaps as low as 30 degrees from horizontal). I guessed this wouldn't be a problem because siphoning would clear the condensate. But I recently saw an argument that Graham condensers should only be used at near-vertical angles. Thinking harder this makes sense to me, because as soon as enough liquid condenses in the first loop to form a trap the condenser can't work on any other vapor. Absent some vacuum on the receiving end the only way it could clear itself is if there happened to be enough vapor to eventually close a loop and start the siphon effect, or if enough pressure built up on the boiling end to push the liquid through which would be bad!

But I still see diagrams and photos of Graham condensers set at shallow angles. Here's an example from the first few Google image results:

Bad Graham condenser setup?

So when and why, if ever, is it considered a good practice to run a Graham condenser in a non-vertical orientation?


3 Answers 3


I have no idea why there are so many photos showing Graham condensers in a non-vertical position. Maybe just because they look so cool and photos tend to be landscape. The only time that might be safe is if there were a vacuum pulling to the collection side.

Graham condensers should be used vertically, precisely for the reasons you stated: Particularly because otherwise they will bubble, bump, and potentially explode.

  • $\begingroup$ One might add that for non-vertical use, the plain old Liebig-type is the only suitable choice. I wonder why anyone calls this Graham-thing a condenser, it's just a cooler. $\endgroup$
    – Karl
    Nov 29, 2015 at 22:14

As far as I can tell, the main reason that graham condensers are shown in near horizontal position is because the setup required to put it properly in a vertical position looks odd to the untrained eye, and requires quite a tall support. The boiling flask, heat source, and so on are elevated as high or higher than the top of the condenser.

The Graham is really only useful as a condenser in the vertical position, and even then, it will tend to "spit" as condensate traps gas above it. To my knowledge it is almost exclusively used in vacuum distillation, for that reason. In spite of numerous drawbacks, it does put a great deal of coolant around the coil. As mentioned, it would be downright daft to use this as a reflux column.

Why this very mediocre piece of glassware should be so common (about half of all distillation apparatus kits seem come with these), is beyond me. Even a plain Liebig condenser is of greater value for any common use that a graham might be used for (vacuum distillation aside). The technique for making graham condensers (to outward appearances) seems like it would be extremely similar to that for making coil condensers, which are far, far more useful, both as reflux columns and distillation condensers, so why the Graham would be so much cheaper is beyond me.


Your reasoning seems sound. For myself, I wonder why anyone would want to use this type of condenser at all. It has a very small bore which reduces the effective surface area. There are a number of other (better) types of condensers to choose from. Was in the lab for 20 years and never found a reason to use this type of condenser.

  • $\begingroup$ Maybe they're only used for photos (because they look so neat)? The small bore does seem problematic, but in simple terms of effective surface area my guess is the coiling actually maximizes that for a given condenser length. $\endgroup$
    – feetwet
    Aug 28, 2015 at 13:41
  • $\begingroup$ I came here through the VLQ queue. While I believe this answer has enough quality not to be treated with the trash we deal with in this queue, I think you should be a little bit more elaborate. Shrugs Some might think this is better as a comment. $\endgroup$
    – M.A.R.
    Aug 28, 2015 at 16:44
  • $\begingroup$ Perhaps this answer could be improved by explaining when and why you use the other condenser types? Sort of a process of elimination? (Because it didn't take long to convince myself that a small-diameter but proportionally longer (coiled) bore has the maximum surface area as compared to other liquid-cooled glass condensers of the same length of the Liebig or Allihn style. Spiral cold-fingers like a Dimroth, on the other hand, would appear to be yet more efficient, though a Friedrichs would appear to have the same perils as I cite for the Graham.) $\endgroup$
    – feetwet
    Sep 11, 2015 at 7:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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