Waters HPLC Binary Pump versus Quartenary Pump

[Peter Skelton]

Certainly such methods do exist, however in most applications suitable separations can be achieved with two solvents.

As a rule of thumb, I tend to look at a tri-solvent method and rule it out as being over-complicated. Certainly I'd never try and transfer such a method between multiple departments across a country or across continents.

[Vlad Orlovsky]

some people are missing the point. Quaternary pump is no for delivery of three different liquids. It is used to deliver organic solvent, water and additive/buffer. It gives you tremendous flexibility to create different rations of ACN, water and buffer without pre-mixing bunch of compositions. We never done a single method on binary pump and we are running almost 24/7 at least five days a week.

How would you create the following gradient with binary pump"
ACN gradient from 0% to 60% with 30 mmol salt buffer, or double gradient of ACN from 0% to 60% with buffer from 20 mmol to 60 mmol?

[Peter Skelton]

Such methods tend to be the exception rather than the rule though. If you have a specific need to run such methods then a quat pump offers greater flexibility. Otherwise binary pumps offer higher pressure capabilities and increased gradient accuracy.

ACN gradient from 0% to 60% with 30 mmol salt buffer - could be done by mixing your 30mM buffer with 60% ACN, then running the 0% solution on pump A, then the 60% 30mM solution on pump B

[Vlad Orlovsky]

There is a need to have more flexible system in method development. I don't see any reason to argue more about this. People who thinks that binary pump is more suitable can continue to use BP at the end you are spending money of your employers not yours on mobile phase preparations. As employer I look at all ways to improve efficiency of each stage of method development

[Peter Skelton]

But the point is it depends on what you're doing. The OP hasn't stated whether they're doing lots of complex matrix separations which require the quat's flexibility or whether they're running separations which will be better handled by binary systems, so your point about efficiency is moot, as each system will excel in different situations.

[scottythree]

I also like to have a clean-up solution on one of my extra lines.

[Peter Skelton]

Most binary pumps allow you to switch between two solvents on each pump

[lperelman]

We use both the H-class and traditional UPLC in our lab, both with great success. I agree with the comments about transferability between the quaternary system to HPLC. I also agree that having a third solvent channel for periodic cleaning is very, very helpful. It really depends on what you anticipate your methods to look like and what the nature of your analyte(s) is(are).

Good luck!

[Alexandre]

If money not a problem now and during the lifetime of the pump (maintenance cost), go for 4 solvent pump.

However from my >5 yrs experience with binary Acquity you do not need quaternary. Sub 2um chromatography is so efficient that now even isotopes are separated. The situation is that good that we are using generic ballistic gradient almost for everything and it separates. Viva La UPLC Revolución!

[remesquaddie]

High pressure mixing is the better way to go, but low pressure mixing [ pump with quaternary capabilities ] give you much more flexibility in programming , such as automated wash procedure etc. Also, high pressure mixing systems are usually more expensive.

[Alex Buske]

Certainly such methods do exist, however in most applications suitable separations can be achieved with two solvents.

As a rule of thumb, I tend to look at a tri-solvent method and rule it out as being over-complicated. Certainly I'd never try and transfer such a method between multiple departments across a country or across continents.

Disagree. Ternary mixtures offer a new dimension in method development, that can save a lot of time for complicated separations. I wouldn't use it for dissolution testing HPLC. On Impurities it is worth a look, especiall if you can't play with pH.

Method transfer isn't complicated. Just define that solvent B is x% MeOH and y% ACN. Try to develop methods for the greatest delay volume involved. That can be up to 4 ml on S**** binary high pressure gradient systems.

[Alex Buske]

High pressure mixing is the better way to go, but low pressure mixing [ pump with quaternary capabilities ] give you much more flexibility in programming , such as automated wash procedure etc. Also, high pressure mixing systems are usually more expensive.

In high pressure mixing there are two possible "points of failures" aka pumps, each including mechanics, piston seals and check valves (some of them don't like continous only-ACN mode). Additionally, with gradients from 0-100 or with mixtures 97 / 3% at least on of the pumps is running in a region, where it isn't qualified.
Last point: in LPG systems mixing starts already in the pump head, where as in HPG systems a additional mixing volume is required.

[Peter Skelton]

Disagree. Ternary mixtures offer a new dimension in method development, that can save a lot of time for complicated separations. I wouldn't use it for dissolution testing HPLC. On Impurities it is worth a look, especiall if you can't play with pH. Method transfer isn't complicated. Just define that solvent B is x% MeOH and y% ACN.

In which case you can pre-mix that ratio and run on a binary system, unless you're specifying A,B and C solvents. Again, unless you're doing something that requires a complex setup, transferring a dual solvent or even isocratic method is going to throw up fewer issues. A method with fewer failure points is by definition more robust, so it depends what your eventual goal is.

Try to develop methods for the greatest delay volume involved. That can be up to 4 ml on S**** binary high pressure gradient systems.

Delay volume of binary pumps is typically in the order of 10s-100s of uL, typical quat pumps tend to be 1-2ml. I have a quat Young Lin system with 7ml of delay volume, our quat Agilent 1200s have about 2ml of delay and the 1290 we demoed had 350ul of delay.

As you suggest, this is something that should be addressed during method development as standard proceedure, it's not a reason to pick one over the other.

In high pressure mixing there are two possible "points of failures" aka pumps, each including mechanics, piston seals and check valves (some of them don't like continous only-ACN mode)

And a LPG proportioning valve has four solenoids to go wrong. Realistically most modern HPLC systems are more than capable of running for extended periods of time without breaking down. The selection criteria should be based on the expected use of the system, if you're doing huge amounts of complex method development then a quat provides greater flexibility to run tri-solvent methods which may be useful. Running methods for most types of product development or QC type analysis may be better served with a binary system.