Chromatography Forum

I am having an issue with system overpressure on UPLC.

I am trying to transfer a method from a kinetex pfp 2.6µm 100x4.6mm column to kinetex pfp 1.7µm 100x2.1mm column, the 1.7µm is a new column. I've used it for a few days without a problem, rougly about 10 injections so far.

However, the other day when I tried to run my mobile phase through the pressure builds up and the system shuts down, the pressure exceeds 15,000 psi. At first I thought it was the guard column that was blocked, but that wasn't the issue. Then I attached the column but disconnected it from the detector and ran some solvent through. The pressure was still high (almost 10,000psi) but the solvent was being pushed through. The pressure seemed higher when running aqueous phase than organic phase. As soon as I connected it to the detector, the system pressure built up again over the 15,000psi limit.

I swithed the column back to my old one, and it seemed to run fine with max pressure around 4000psi. So I'm assuming it's the problem with my new column, but I don't know what the issue is.

The mobile phases I use is (A) 5% formic acid in aqueous and (B) 5% formic acid in 80% ACN.
The samples I run are MeOH extracts of plant material. At the end of each run the column is conditioned with (B) for 20min.

Help?

Do you see your pressure building up with each injection?
For me it looks like an incompatibility between sample and mobile phase.

Ace

With that "old" column that runs fine at 4000 psi, do you mean and old 2.1mm 1.7µm column (EXACTLY the same you're having issues with) or do you refer to the 4.6mm 2.6µm column?
With the flow rate scaled to the internal diameter change, the 1.7µm column will give you about 2.3 times the back-pressure than the 2.6µm column. 4000 psi x 2.3 = 9200 psi.

Do you see your pressure building up with each injection?
For me it looks like an incompatibility between sample and mobile phase.

Ace

Hi,

No there is no incompatibility between sample and mobile phase. The method is already developed just needs tweaking. I've already ran samples through using this column few days before. I'm not sure what happened.

The pressure builds up as soon as I tried to equilibrate it with solvent. I don't actually get to the sample analysis part.

The odd thing is when I run through aqeuous phase (just pumping through) the pressure builds up but when I run through organic phase it's much less.

With that "old" column that runs fine at 4000 psi, do you mean and old 2.1mm 1.7µm column (EXACTLY the same you're having issues with) or do you refer to the 4.6mm 2.6µm column?
With the flow rate scaled to the internal diameter change, the 1.7µm column will give you about 2.3 times the back-pressure than the 2.6µm column. 4000 psi x 2.3 = 9200 psi.

I have two columns:
a 4.6mm 2.6µm which has been used for a awhile and 2.1mm 1.7µm column which is spanking new - I'm trying to transfer my method from 4.6mm 2.6µm to 2.1mm 1.7µm.

I've reduced the flow-rate but still working on adjusting the gradient to get the best peak resolution.

I'm new to method development so I'm still learning about these things.

Can someone please explain to me why my new column had worked with the exact same method for about 3-4 runs and now it doesn't?

So you lowered the flow-rate to account for the reduced internal diameter of the columns? What was the original flow-rate (on the 4.6mm column), what's the flow-rate for the 2.1mm column?
As I already said, if the flow-rate was scaled down according to the change in column diameter, the 1.7µm column will give you roughly 2.3 times the back-pressure of the 2.6µm one. So if you have ~4000 psi with 2.6µm you MUST have ~9000-10000 psi with 1.7µm one under the same conditions. That's the drawback of using smaller particles .

Concerning the flushing: The more acetonitrile you have in your eluent, the lower its viscosity will be and the lower the backpressure generated by the column. Pure acetonitrile will generate MUCH lower backpressure than purely aqueous eluent. Again, this is a MUST, not a problem with your system or the column. Never watched the pressure while running a gradient? (Methanol is a different story, though...)

When you first used your brand new 1.7µm column, did you really equillibrate it with your mobile phase? If not, I'd guess it was still filled with the shipping solvent which is usually high in organic when you made your first injections. So the pressure building up was actually the column equillibrating with your mobile phase...

So you lowered the flow-rate to account for the reduced internal diameter of the columns? What was the original flow-rate (on the 4.6mm column), what's the flow-rate for the 2.1mm column?
As I already said, if the flow-rate was scaled down according to the change in column diameter, the 1.7µm column will give you roughly 2.3 times the back-pressure of the 2.6µm one. So if you have ~4000 psi with 2.6µm you MUST have ~9000-10000 psi with 1.7µm one under the same conditions. That's the drawback of using smaller particles .

Concerning the flushing: The more acetonitrile you have in your eluent, the lower its viscosity will be and the lower the backpressure generated by the column. Pure acetonitrile will generate MUCH lower backpressure than purely aqueous eluent. Again, this is a MUST, not a problem with your system or the column. Never watched the pressure while running a gradient? (Methanol is a different story, though...)

When you first used your brand new 1.7µm column, did you really equillibrate it with your mobile phase? If not, I'd guess it was still filled with the shipping solvent which is usually high in organic when you made your first injections. So the pressure building up was actually the column equillibrating with your mobile phase...

The original flow rate of the other column was 0.7mL/min. You're right the pressure sky-rocketed and hence system overpressure. But still confused why it didn't happen the first few times I used it.

When I first used my new column I did equilibrate it with mobile phase, as well as running through some blanks injections. All seemed fine at the time - even at 0.7mL/min.

I did watch gradient pressure, your explanation makes sense as when the aqueous phase was pumped through the pressure is much higher than my organic phase containing 80% ACN. I learn something new everyday.

In terms of flow rate for my current column, I first scaled it back to about 0.4mL/min, and it was operating about 8500psi. My peak resolution didn't turn out so good but that's because I was fiddling with the gradient time. I've now scaled it right down to 0.15mL/min and also decreased the injection volume from 5µL down to 1µL. But keeping the same gradient time - running that one now so will see how it turns out.

How do I maximise peak resolution without extending run-time which is already really long?
I am analyzing multiple analyte so I need to make sure the peaks are not too wide and co-eluting.

welcome to my world!...ive spent the last 2 months with this same problem. so i know what you are talking about. Get your column and remove the inlet head off it and replace the frit in side the head. but before you replace it reattach the head without the frit and pump 100% slovent through it for a amout 30 mins now take the inlet offf again and replace frit with new one. I gurantee that it will work! but you going to have find whats causing the blockage......most probaly inline filter contamination....wrong filters used....growth in water....not refridgeuating your hplc water after open...etc etc...let me know how you get on and make sure your preparing new mobile phase every 1-2 days. also get rid of your guard column you dont need it and get rid of your inline filters

No, no, no, don't take that column apart! If I understood correctly, there is no blockage and the "excess" pressure was just due to going to lower particle size. BTW "get rid of your guard column and your inline filters" is a strange advice for UHPLC, especially since you yourself just suffered from a blocked column. Rather than taking an UHPLC column apart, risking to create a void I'd spend a few bucks for a new guard or a filter. I'd certainly go with at least an inline filter on UHPLC.

@atw9: Please don't take this as offense, but I think there are more basic things to learn for you than how to maximize peak capacity. It's hard to develop efficient and robust methods when you're just at the stage of being puzzled by the different backpressures of water and acetonitrile. Method development is a high art, mastered by few (me not including) and there are quite a few things to consider if you want to prevent frustration with the developed methods.
There are some good books out there, "Practical HPLC method development" is a classic, "High-performance gradient elution" another very good one. Consider also courses and seminars like those from LCResources or others.

the inlet frit is blocked i am 100% sure and and if he/she doesnt replace the frit the column it will stay blocked. the reason why its blocked cause you went to lower particle size thats why! and something that wasnt effecting your 2.6um column is now effecting your 1.7um column. you wont get a column void and i have waters engineers tell me to do what im saying

the inlet frit is blocked i am 100% sure and and if he/she doesnt replace the frit the column it will stay blocked. the reason why its blocked cause you went to lower particle size thats why! and something that wasnt effecting your 2.6um column is now effecting your 1.7um column. you wont get a column void and i have waters engineers tell me to do what im saying

He gets about 9.000 psi with the 1.7µm column where he had 4000 psi with the 2.6µm column at the same linear velocity. That's not a blocked frit, that's what's to be expected.
And opening a column in order to exchange the frit IS critical with UHPLC columns. You can easily create a void and thus destroy efficiency. Especially the Kinetex he's using is not meant to be opened, I heard that from the guy who designed the column hardware. Of course, if the column is certainly blocked, you may try this as the column would be dead anyway...

the inlet frit is blocked i am 100% sure and and if he/she doesnt replace the frit the column it will stay blocked. the reason why its blocked cause you went to lower particle size thats why! and something that wasnt effecting your 2.6um column is now effecting your 1.7um column. you wont get a column void and i have waters engineers tell me to do what im saying

He gets about 9.000 psi with the 1.7µm column where he had 4000 psi with the 2.6µm column at the same linear velocity. That's not a blocked frit, that's what's to be expected.
And opening a column in order to exchange the frit IS critical with UHPLC columns. You can easily create a void and thus destroy efficiency. Especially the Kinetex he's using is not meant to be opened, I heard that from the guy who designed the column hardware. Of course, if the column is certainly blocked, you may try this as the column would be dead anyway...

@seamoro
I won't touch the column for now, see how I go with some runs by reducing flow rate and extending gradient time.

@HPLCaddict
Thanks for your advice no offense taken. However I'd like to point out that this is a learning curve for me I know the theory stuff, I've read the books, but I think it's more useful to have practical experiences. I've done calculations etc but sometimes you learn more by doing practical things. It's a project I'm working on and I am being supervised (except the supervisor is away...All things go wrong when they are away!) and I was following directions right up to the part that things suddenly don't work. It's a complicated sample to analyze with many many analytes. I guess I enjoy the challenges!

P.s I'm a lady

Ma'am:

Depending on funds, I would do a couple of things:

1. Buy a new column to be a replacement or a backup to your 1.7um column - columns don't last forever, and it's a terrible thing to be without a backup.

2. Go to http://www.hplctransfer.com and input your analysis conditions from the 2.6um column to scale them to the 1.7um column. Inputting the values you've given agrees with what others have posted - ~9000 psi is about what you would expect. What temperature are you running the column? Running at elevated temperatures with small particle columns is usually a must due to the increase in backpressure, and running at elevated temperatures will reduce your backpressure and will help stabilize your retention times (as long as you aren't heating the column beyond manufacturer's limits or beyond what makes sense for the mobile phase and stationary phase combined).

3. Buy an inline filter for the HPLC system (preferrably, a few). I like to run filters after the mobile phase mixer and before the autosampler and after the autosampler and before the column, to keep particulates from the pumps/mobile phase away from the injection valve/column, and to keep particles from the sample/injection valve away from the column. I'm using Restek's UltraLine UHPLC In-Line Filter with my LCMS, and have also used the Optimize Technologies EXP 0.2-micron Stem Filter system. The Restek 0.5-micron filters are indeed cheap insurance for an expensive column at $12/filter.

I agree with the above - do not open the Kinetex column, or really any column. I've tried it in the past with 5-micron traditional columns, and I have almost always ended up throwing the column away. You must keep in mind that a column, while expensive, is a consumable - it will perish. Think of it like tires on a car - with proper care, you can get a lot of miles out of them, but at some point they need to be replaced.

Dear Mrs. atw9 ,

things always go wrong when the supervisor is away . Anyway, you've got the right attitude seeing this as a challenge to enjoy. And, yes, practical experiences are always better, but a sound understanding of the underlying theory never hurts, and for HPLC method development it's inevitable. You've just made the practical experience that sub-2-micron particles will create LOTS of backpressure...

Concerning your transfer/optimization problem: What's the reason for turning to the 1.7µm column? What was wrong with the original method on 2.6µm? Are you supposed to increase the resolution? Then going to 1.7µm and remaining the column length is logical. Are you supposed to speed things up? Then a shorter column with 1.7µm would be the logical step.
BTW running that 4.6mm 2.6µm core-shell column at 0,7mL/min is like driving a Ferrari in second gear. You might actually increase efficiency by increasing the flow rate (and at the same time decrease run time)...

One little noob question:
What HPLC system are you using and is there any stainless steel capillaries?