Good suppliers for preparative HPLC systems

[DanielTilibit]

Hello all! This may be an a little naive question but I´m entirely new to the field and just wanted to hear some opinions. Does anyone have experience with preparative HPLC systems that could also potentially be hooked up to an MS? Currently, we have two offers: One from Agilent and one from Shimadzu. The latter one is about 100k less than the first one. Waters turned down from making us an offer in the first place. Do you consider Shimadzu a good choice here? Since their offer is so vastly cheaper, it made us a bit suspicious. Are there other suppliers apart from the three that I stated here already? I would be very happy about your feedback. Cheers!

[DR]

Basically, it's the same list as you'd use for analytical systems.

I've got to ask - why would you want to hook a prep system to a MSD?

[DanielTilibit]

Could you show me that list that you are referring to?

Is it unusual to couple a prep-HPLC to an MS? It´s a system that also could be run in analytic mode, so with a small column. I guess, therefore a MSD would be a good thing.

[DR]

Using prep systems as analytical systems can be problematic because of pump and plumbing design issues. The higher flow rates needed for prep work limit your sensitivity for analytical work. Feeding a MSD from a prep system only augments the challenge (even with dedicated analytical setups, the MSD gets fed only a small fraction of the post column eluent).

I would be inclined to check with Agilent, Waters, Hitachi, Shimadzu and, if you're not in the US, Knauer, MetrOhm, Jasco and others that service your market.
If Waters would not talk to you about this...that should tell you something about what you're requesting.

[DanielTilibit]

Thank you very much!

[TylerSmith123]

Hi Daniel,

My lab has a Shimadzu instrument that operates in that sort-of "hybrid" mode, with both an analytical portion and a preparative portion. It would take me a while to explain the intricacies of our instrument, but I would say one thing overall about having a hybrid system. Don't.

There are a lot of issues you may run into if you have a hybrid system, rather than two separate systems. The list is pretty exhaustive, and DR mentioned a few of them already, but I will name a few I have experienced. Depending on your use (mine is primarily preparative), the detector will function weirdly. In every single run on the prep, the detector is "blown-out" or oversaturated. If you do this every day for weeks, your detector will become less sensitive due to the constant over absorbance. So for accurate results on an analytical set-up, you may need to switch our the detector.

The tubing, volumes, etc. will also be an issue as you will have parts of your system, no-matter what, have incorrectly sized tubing. This can change your resolution for important samples and you may not be able to quantify the effect.

If you're thinking about using an autosampler, then you might want to decide early which side it will be used for. You do not want to switch out the autosampler loop every time you need to switch between prep and analytical, but you also don't want to shoot your 20ul analytical injection into a 2 mL sample loop made for prep.

This is the first system I have been trained on, and as I learn more and more about LC, it makes my frustration with my current instrumentation grow vastly. Additionally, Shimadzu and other companies may be able to build you something, but the support is lackluster at best because your system ends up being so unique that it may require in-person inspections and head-aches when it comes to figuring out potential issues. I could go on and on about the issues my LC faces, but if I was the one to purchase the instrument, I would certainly never do it in this fashion again (which may be the product of someone inexperienced purchasing an LC for the first time (cough-cough the prior PI)). I'm not saying it is impossible to have this system work well, however it would be much easier to just purchase as system for analytical samples and another for preparative samples. You will be making sacrifices no matter what in a hybrid system, but if your application can sustain some of these sacrifices, then hey, save the money!

TS

[lmh]

Prep LC-MS is quite normal. There is nothing weird about mass-directed fractionation.

Typically you hook up an MS to a prep LC when you want to use the MS to detect the compound you're trying to purify, and tell the fraction collector when to do the collection. Agilent are pretty good at it, but their system isn't cheap. There are good reasons why it isn't cheap.

Firstly, these systems generally assume you're doing an enormous split ratio, because they assume that you're working with a large injection of a very concentrated sample and trying to get maximum yield, so you don't want to chuck away 10% of your injection in the (destructive) MS, and in any case, that much stuff thrown into the MSD would be 100 times its maximum signal. So the Agilent Prep LC-MS systems have an active splitter, not one of the rubbish passive splitters (i.e. a couple of resistances) that you'd use in conventional lower-flow, lower-ratio splitting. The active splitter guarantees the split-ratio you specify, regardless of potential blockages and changes of pressure on each side of the split.

This also means you need an additional pump to provide the make-up flow after the split, because if you're splitting 1000:1 on a 25mL/min flow, you only have 25 uL per min going towards the detector, so your peak will get detected after it arrived at the fraction collector (and in any case is too concentrated). By adding 1mL/min of flow from a second pump immediately at the outlet of the split, you increase the flow to 1025uL/min, achieving a sensible dilution and making sure the peak gets to the detector really quickly, so you have a positive delay-time between detector and MS. This means your prep system needs an extra analytical-style pump anyway.

In Agilent's case, they can also supply the system with valves such that the analytical-style pump can run an analytical-style column, which can be used to re-analyse the collected fractions from a prep run, if you wish, or can also do analytical chromatography independently (though I'd agree it's not ideal for the job).

On the negative side, I would say that many people find that in practice their chromatography is sufficiently repeatable that once you've found the peak, you don't need to do mass-directed fractionation on a regular basis. The place where you're stuffed if you don't have a mass spec attached to your prep system is where you're purifying something that has no chromophore for PDA and you simply can't detect in the crude mix by any means but MS. How are you going to work out where to collect the fractions? If you have an MSD attached, you simply find where the product is, and either use the MSD to trigger collection, or continue to collect a window at that retention time. If you haven't, you've got to collect fractions across the whole run blindly, and then re-run the whole lot on an analytical system separately.