U-HPLC, UV/VIS or DAD (PDA) detector

[bobokiki]

Dear All,
I have to take a decision about purchasing new LC equipment for our bioanalytical lab. I wonder between the U-HPLC (ultrahigh-pressure chromatography) and standard HPLC system (pump-up to 400 bar), our analytical field is in the clinical chemistry- we analyze biological (plasma and blood samples, extracted by liquid –liquid or other extraction technique. We are also in doubt about UV/VIS or DAD (PDA) detector due to the Noise level of the detector Please recommend us which detector to choose.

[juddc]

I would strongly recommend a PDA over a tunable UV-Vis because you gat so much more information with a PDA and the sensitivity advantage of a tunable single wavelength UV-vis has been reduced significantly in recent years. Modern diode arrays detectors are much quieter than their predecessors.

The main advantages of UPLC over HPLC are speed and (to a point) sensitivity. If you're in a high-throughput lab and you need ultimate sensitivity but only will use UV detection, then go for UPLC. If you need a bit more flexibility in detection (i.e fluorescence, conductivity, RI, etc) then standard HPLC might be the way to go. Others can comment in more detail on method trnasfer and other issues...

[lmh]

I don't know if this helps, but most manufacturers have systems that can work just like an ordinary hplc, but offer some uhplc-like facilities, including higher pressures. For instance, Thermo's Accela is very much like their older Surveyor, but offers a high-pressure mode with minimal alterations needed, should you want to experiment with high-throughput. Similarly the Agilent 1200 pumps are capable of more than the old 1100. These systems save you being forced into a decision quite yet!

Waters are a bit sniffy about these competitor efforts, pointing out that Waters are the only people who've produced a complete, optimised uplc system offering the "real deal" on high-pressure, short run time. But I've heard rumours Agilent have an equally good system in the pipeline.

[HW Mueller]

Just curious: I used to have huge matrix problems with blood samples using normal columns (4.6x250mm), thus I wonder how these much smaller columns handle this. Is more pre-cleaning necessary?
On the detectors: I have mentioned this before, namely, that I am using a Linear (something ele now?) detector which can run a spec. in a fraction of a second with 1nm wavelength resolution. It is apparently also more sensitive than any PDA. In fairness I must say that some people seem not to like this detectors method of running at two wavelength simultaneously: It swings (the grating) mechanically back and forth between the two wavelength.

[aceto_81]

Just curious: I used to have huge matrix problems with blood samples using normal columns (4.6x250mm), thus I wonder how these much smaller columns handle this. Is more pre-cleaning necessary?

I think this is matrix dependend, but on a regular column with a prefilter, you could built in some extra space like pumping at 2000psi, and when the prefilter becomes clogged, you end up with 3500psi.
For the U-HPLC likes, the buildup is much higher, so you can clog up less before running into trouble.
But you have a lot more sensitivity, so you can dilute a lot more, so the matrix effect is less.

But as said, everything depends on the application.
Another thing: the needle of an alliance take the sample at the side of the needle, while the UPLC actually takes the sample with the needle tip. For precipitates samples in your vials, this means the HPLC takes a sample just above the precipitation, while the UPLC.... right!

Ace

[oscarBAL]

Hello All; I just want to add something to this dicussion.
Why when we talk about UPLC everybody talks about speed?
I think that works with sub 2um particle its really helps to resolution!!
In the case of dificult samples Resolution could be more important that speed.
and better resolution also could means better LOD. please correct me if I am wrong.

[aceto_81]

You're absolutely right about the resolution and speed: Every body thinks about speed, but you certainly can use the UPLC to maximise resolution.
If more resolution gives you a better LOD, that's not always true: assume a HPLC method with 2components, barely separated with 50% MeOH. Now you only take 20% MeOH, you resolution may be more, but also your retention time may be 50 minutes or so. How about the LOD you think?


Ace

[Hfranz]

Dear All,
I have to take a decision about purchasing new LC equipment for our bioanalytical lab. I wonder between the U-HPLC (ultrahigh-pressure chromatography) and standard HPLC system (pump-up to 400 bar), our analytical field is in the clinical chemistry- we analyze biological (plasma and blood samples, extracted by liquid –liquid or other extraction technique.

It is simply not required to decide for or against an UHPLC system. My suggestion is a system that handles both: Dionex' UltiMate 3000 RSLC system. Many of our customers buy an RSLC today to be prepared for ultra-high pressure applications in the future. For the time being, they use their standard HPLC applications with 4.6 mm columns of up to 30 cm length. Recently, we have also expanded our pump offering, now offering binary, quaternary, and dual-gradient pump versions for up 800 bar (11,600 psi) and 5 mL/min.

[Bryan Evans]

I would suggest talking about these systems with colleagues doing similar research as yourself. Ask them what successes / failures they have had with sub 2um particles.
These U-HPLC instruments are designed to be able to pump solvent through a column packed with sub 2um columns. It addresses the "N" part of the resolution equation.
But you still have to fight with method transfer, sample prep, k', alpha, peak shape, ect. There are practical issues that need to be considered.

I'd also suggest looking at our website. You'll find hundreds of separations (using 3um particle) that
a). cannot be done with sub 2um columns, and
b). do not require U-HPLC systems

http://www.imtaktusa.com/library/technical_information/

[unmgvar]

bobokiki,

your decision will have to be divided first according to those demands:

1. application needs. what are you doing what do you need to achieve?
2. cost to efficiency ratio- you can get faster, more sensitive, more reliable, ease of use, robustness, cheaper, but never all of them bundle together especially if you need to answer 1 as well.
3. System wide specs- check flow rate range, higher pressure limit, autosampler range of injection, type of injection, column oven range and column max size can add a pre-eluent or not. detector speed, amount of channels, system modularity, software capcities, GLP (if required), ease of use, calcualtions capacities
3. how much budget do you really have
4. tech/application support of your vendors. you do not want a white elephant in your lab.

when you answer that you will know if to go 1st for UHPLC or HPLC, and then 2nd if to go for a UV or PDA

as for your specific questions:
first of all remeber that there are still more column phases avalaible in analitical range and far more vendors providing possibilities. going UHPLC means less choice right now (it gets better every year)

not all vendors give the same specs for their UHPLC system, especially on 3 things. higher limit pressure of the system ; type of injection; speed of data collection of the detectors.
one interesting physical restriction is the 800 bar limit. that is where "plastics" become more "liquid" then solid. to be able to run at higher pressure means other materials (mostly ceramic- using buffers and working with "dry" system bring higher ware faster).
another interesting aspect of working above 800 bars is precision. precision of flow rate, precison of injection. i have seen an Accela not do 1 ml/min at 1000 bars, could do only 0.7 ml/min. you never know this unless you chekc it.

so even if you do go for a "super" resistant system, robustness is an issue needed to take into account. i have heard of UPLC customers that told that their Waters tech adviced them not to go above 800 bars because of the exponent ware it creates.

remember that sensitive is a double edge sword. sensitive results are achieved by more sensitive system with all of what it means.

Type of injection:
there are 2 injection types currently used that i know of in UHPLC systems. the difference affects performance, way of use and effective injectin range
split loops
pulled loops

split looped are the better ones, more precise better RSD. more expensive, require good engenering for low carry over (increases expenses), a lot less sample loss during injection, very fast injection time

pulled loops are less expensive, simpler engenering, if well used give by far the lowest carry over. but very high sample loss, need to be used as partial loops or full loop. very slow injection cycle (very important if you do very short runs like 1-2 minutes). RSD is less then in split loop but it does not mean that it won't be good enough for you, it depends on what you are doing

detector UV or PDA:
UV detectors on a single wavelenght give at least 7 times less noise(many times more even) then PDA- if you seek low detection limits this is an issue
UV detectors have a far better behavior of Refractive index value then PDA. this can mean smaller front peaks effects at zero dwell volumes, a lot less drift due to gradient methods.

PDA gives you 3D spectral information, althou in words it is short, but in information it means alot.
for both types of detectors you will need at least 40-60 Hz for a UHPLC system. do not bother with less if you go for 1-2 minutes runs and very narrow peaks

i agree with Bryan Evans on his suggestion and also advise you very strongly not to believe all of those sales people that will tell you that transfering from HPLC to UHPLC is easy. it requires validation. on the other hand if you start from nothing then it is not an issue you validate from scratch anyway.

[Uwe Neue]

As somebody from a group that is routinely working below and above 800 bar, and who has never seen an issue or heard about an issue of 800 bar being some kind of a limit, I highly doubt unmgvar's statement. While I agree that different plastics have their limitations, you can bet that the engineers that design a 1000 bar system will have selected components that are - at the minimum - good to 1500 bar, or more typically to 2000 bar.

Let us not spread crazy rumors that are clearly not based on observable facts with an UPLC system.

[aceto_81]

800 bars limit?
I don't know about that, we frequently work higher when needed, with no trouble.

But when it's not needed, we won't push things to the limit.
You can run an UPLC method at 14500psi, but when your inline frits get clogged, you will soon run out of pressure.

But this is the same with an HPLC method, if your system limit is 4000 psi, who runs all the time at 3800 psi?

Whenever possible we try to make our methods robust, even for pressure increase.

Ace

[mbicking]

Couldn't pass up this chance to put in my two cents worth!

Hans, to answer your question about the small columns, I have not done a lot of work with them, but my experience is that, yes, you can get very fast separations with very good resolution. But there is always a price to pay. With the smaller dimensions you have less sample/loading capacity. With the smaller particles, you have smaller frits, which are then more sensitive to particulates; filter to 0.1 um rather than 0.2 um. All that said, unless you are very careful with your procedures, column lifetime is likely to be somewhat shorter with these columns.

My opinion is that UPLC (or whatever we want to call it) is needed if you really need to push the limits for speed (<1 minute run times) or high resolution (>100 components). If you aren't in this category, you can probably do well enough with a good quality system (available from many different manufacturers).

[HW Mueller]

OK, sort of what I would expect. Highly generalized: It takes a bit more disciplin, etc., than "normal" HPLC.
When UPLC just hit the market we discussed whether it is worthwhile. At least a factor of 2 was claimed as resolution improvement. My conclusion was that this would certainly justfy the introduction for some separations. If I had to set up a lab I would go for both normal and ultra systems. For instance, normal analytical columns can be useed conveniently for prep. separations in nuclear medicine. Normal HPLC should also be useful as a pre-purification for UPLC.