PhD Candidate, Mentor left, pressure issues

[cafeguy]

Hi there,

I'm running out of resources here. Its my final year and my PI left for a different job. I'm now alone in the lab for these remaining months trying to keep it going with running the rest of the grants and company contracts. I need to get the HPLC running nicely for my last group of experiments, but since my mentor has gone, I have nowhere to turn to figure out how to fix these problems. Not sure if this is the right place to ask... but, if not, could you kindly point me in the right place?

the issue...

I'm running a microbore column to detect adenosine from microdialysis samples. Right now, I'm just trying to get standards going. Its a UV detector with a manual injector. The issue is pressure.

The column is a few years old, but I just put it into use two weeks ago (I have run it non-stop since, only having maybe a day or two where I wasn't running standards and calibrating the microdialysis probe head. I am under the impression keeping it running in 50:50 wash buffer is the best thing to do instead of letting it sit unused and not flowing at any point in time. Maybe I'm wrong about this?) and at first, my pressure with 50:50 MeOH:H2O was 15MPa. I have only run standards through this thing to calibrate the microdialysis probe, so no samples from the rat were run. Just standards in polished water.

Last week, the pressure dropped from 15MPa to 11MPa overnight. I figured out that the pump's plunger seal needed to be replaced so I ordered it. The next day, after running the system overnight to wash, the pressure was back up to 15MPa. I ran some standards again that day. The next day, the pressure was all the way up to 25MPa!!! I changed the wash buffer in case it was contaminated (degassed and filtered). I replaced the plunger seal and confirmed that it was working correctly. I completely cleaned out the manual injector by running wash buffer through the sample resevoir loop and pushing wash buffer through the injector port while in the "trigger" position. I did this while the column was off to make sure it didn't hit the column.

I reversed flushed the column for about two hours. When I put the column back in position, it ran at only slightly above normal pressures for hours.
Then, overnight, it jumped all the way up to 30MPa again!!!!! This time, it hit this pressure at a flow rate of 30ul/min. The above pressures were for flow rates of 50ul/min.

I have tested the column. The standards work. I can detect adenosine, but this pressure is getting WAY too high. Something is wrong and I have nowhere to turn to try to figure this out, and I have little money left to call out a tech support for a visit.

Since the pressure climbs overnight, I almost feel like it is something that is aggregating over time that builds up on the column entry point and raises the pressure. I think this because reverse flushing seems to fix it. I have a guard column with a new filter on it too and have always used this. I have heard that some parts of the system can sometimes wear away and cause clogs. What parts of the system do this? What do you think I should replace? This overnight pressure build up makes me think this.

Could it be worn seals somewhere I'm unfamiliar with? Worn tubing? Worn filters?

I should also mention that for nearly two weeks, I would run this column every night non-stop at a low flow rate with wash buffer. It never did this. It wasn't "slowly" getting worst and worst. It seems like in a single day, it just instantly developed this "overnight pressure" issue. Like, BAM. Here it is... every night, it will do this. I just started revrese flushing it again, and at the same flow rate... its all the way to 16MPa now. That actually speaks against it being an issue of something wearing apart and catching debris since if it was wearing apart, I would expect it to wear apart slowly...slowly developing a higher pressure each day...instead of all of a sudden, developing this issue in a single night, then doing the same every night. BUT, when I reverse flush it, it seems to fix the problem, suggesting I have pushed off any debris that has built up.

Ugh...it sucks not having a PI here, who has all this experience with this incredibly old machine, to help me fix this situation.

Thank you.

[PLCMON]

So if I am understanding you correctly, you have a guard column and a pre column filter? But back flushing your main column temporarily solves your problem?

Also, a stupid question...do you have them set up in this order?:

injector---->filter--->guard column---->main column--->detector


What is the make and model of your system?

Is there a pump frit that needs to be changed?

What are you running overnight when the pressure jumps?

[cafeguy]

So if I am understanding you correctly, you have a guard column and a pre column filter? But back flushing your main column temporarily solves your problem?

Also, a stupid question...do you have them set up in this order?:

injector---->filter--->guard column---->main column--->detector


What is the make and model of your system?

Is there a pump frit that needs to be changed?

What are you running overnight when the pressure jumps?

Its a Shimadzu machine. The pump is LC-10ATvp. The injector is a Rheodyne Injection Valves 7125.

I must say, I am a novice at this. Actually...its not a guard column from what I am to believe. The lab always called it the "guard column", but it really doesn't have a column component to it. It is more like two caps with a filter element that sits between those two caps. It looks more like this (http://www.futecs.com/product/hplc_acc/img/hplc209.jpg) with a filter put inside there. The column directly hooks up to this.

So I would say...

injector---->filter--->main column--->detector

I am not sure if the pump frit would need to be changed, but I am to understand that the pump frit prevents particulates from pump seals that are falling apart, but if the the pump seals are new, then it wouldn't be them right?

I am pumping 50% MeOH in polished water (filtered and degassed) overnight.

Another interesting note is that when I run this through the column reversed, the pressure is at 16MPa. If I stop the pump, and flip the column to the correct orientation, the pressure goes to 19MPa and is slowly climbing. I just did this. Not sure if flipping the flow so often is bad.

I also reversed the order of the ferrels on the column, in case one ferrel is overtightening the column at a certain point.

I'm almost at the point of just whipping out my own credit card and buying a new column.

[uzman]

It looks like , the ''fines'' ( crushed silica particles inside the column ) collected at the detector side of the column and blocking the outlet frit gradually.

When you reverse the column , they allow the flow , detaching from the frit .

If you have no other choice , you may continue your work by using the column reversed and see if it is giving the same results.

[PLCMON]

The logic of uzman sounds like it could be correct. Probably the easiest thing to try.

I am not to familiar with the particular model pump you are using, but for the ones I have worked with you need to change the pump frit once or twice a year minimally. Also, if you didn't change the frit after you changed the pump seals...then it might be a good thing to do regardless.

If you can't run the column in reverse like uzman suggested, maybe there is someone else at your institution that could lend you a column to at least see if that is the problem. Then at least you would know if you bought a new column it would solve the problem.

[M_Farooq]

Sorry to hear about your ordeals. Anyway the problem may not be as serious as you may think since everyone faces these HPLC issues quite frequently. It was a bad lab practice to place the "inline filter" between the injector and the column. Secondly, the inline filter may reduce the efficiency of the peaks if placed between the column and the injector.

Assuming you use clean samples (pre-filtered samples, if needed) the in-filter, which your lab called as guard column should be between the pump and the rheodyne. It is usually the pump seal parts which clog the column inlets on old hplc pumps. Sometimes, in worst cases, it is the worn out rheodyne seals (see the test below). For a quick try, open that "guard column" and notice the colour of the filter. If it has turned black from its metallic colour, the seals are dying. Sonicate the filter frit for 10-15 minutes in hot water, MeOH, and in dichloromethane separately. It is very likely that the pressure will become normal and life will be back to normal.

16-19 MPa is not a big deal at all on a microbore column as long as the column is alive that too with 1:1 MeOH-H2O. This is a very viscous eluent more viscous than water and methanol. What is your flow rate?

Additionally, the movement of fines is possible but not very likely in this case. Fines do not move from end to the other by reversing the flow since these microbores are packed at crazily high pressures, so the bed is pretty stable and in the most compact form possible. Assuming it is a commercial column, the fines must have been removed before packing the microbore column. Try the simplest idea first, sonicate and clean the inline frit.

I also work my way backwards up step by step for pressure issues. Remove the column, note the pressure. Remove the rheodyne tubing going to the column and note the pressure. Remove the injection loop and note the pressure. If at any place you see an abnormal rise in the pressure, that tubing could be the culprit too. Discard it and attach a new tubing. If at every place the pressure is normal, even after sonicating and cleaning the inline frit, use the column in the reverse direction. The world will not come to an end after reversing the column flow direction and if it is working fine till you graduate, no problem

Good luck.

[Gerhard Kratz]

I agree with M_Farooq. Maybe you have still time to book an online crash course, or better ask around to find an experienced colleague. You can kindly offer him some beers to compensate his help. In my case you would be bankrupt afterwards...............
Good luck

[lmh]

You are using polished water and pumping continuously - which is good. How often do you replace the water? I don't know how they do it, but bugs grow in LC-MS-grade super-clean 0.2u filtered water. Personally I'm not averse to having some sort of guard or filter between the injector and the column. Yes, it can reduce the efficiency of the peaks, but if you choose your filter well, as one sold for minimal mixing volume and intended as a column guard, then the effect is minimal. If you're only 99% certain that your samples are absolutely free of any precipitate, you can be 100% certain that the standard you run will be the 1% that contains some dust.

[cafeguy]

Thank you all for your help and suggestions!

Fortunately, after back-flushing the column for awhile, I went in and cleaned everything with the 40% acetonitrile, then 100%, then 40% again. I did this at longer intervals and it actually seemed to do the trick and everything is running at a nice pressure now!

Another issue though...

The D2 bulb I was using was very very old... at 5,000 hours of its recommended 2,000 life. Before ordering a new one to change it out, I noticed some odd peaks that would come up during my 10 minute runs. Most of the time, it wouldn't obscure my sample peak so I gave it no thought. I thought it was just the bulb flickering or something to due with an old bulb.

It wasn't.

The same problem is happening with the new bulb. The peaks are HUGE in comparison to my sample, always have the same shape and size.

I ran a 60 minute run just to see if they happen at a specific rate, and it seemed to happen at semi regular intervals of 10 minutes (not always perfect intervals) or so. It happens if I don't inject anything and instead, just record. It happens if I run the wash buffer through it. It happens if I take the column off and just watch the system.

the flow rate is about 80ul/min.

I am making new mobile phase again to see if there is some sort of contamination (with newly made polished water).

Any idea what it might be? Some troubleshooting guides say peaks could be due to air bubbles or something with the pump... but the curious thing is that EACH and EVERY peak is identifical. There is no significant variation.

A zoomed out image of this can be seen here... https://www.dropbox.com/s/m0qjg7lkanqdn ... 2.jpg?dl=0

A zoomed in image of one of the peaks can be seen here... https://www.dropbox.com/s/jq3fs7p3l4h8a ... 3.jpg?dl=0

And like I said... these peaks are HUGE. WAY out of range of my standards.

[M Farooq]

There are simple rule of thumbs which might help you in finding the source of problem

1. If the noise is periodic, with large to medium amplitude fluctuations and "low" frequency corresponding to the pump strokes, this is due to the bubble in the pump head. There could also a bubble in the detector. Eliminate option #1 first before beginning complex solutions

2. Noise which has high frequency, low amplitude, without a drift, this is coming from a detector with bad settings (too fast of the detector time constant)

3. Spikes are tricky: Since these are periodic, with a split, it might be possible that there is some electrical interference in the power lines nearby. Try a different power line or turn off something close something which is consuming lot of power. See if it helps.

[cafeguy]

Thanks for the advice! Its a great rule of thumb.

I purge the pump every time I move to a new buffer so I don't think the bubble is in there if it is a bubble. I am trying to find a syringe adapter to clean/remove potential air bubbles from the flow cell right now.

As for the electrical thing, I could play around with the orientation of things... another note to add... I did move the whole system a couple months ago to a new room since my PI left and that is the first time I noticed the "peak" occurring.

[M Farooq]

A poor man's solution to remove a detector bubble is to gently place the thumb at the detector outlet tube without blocking the flow. Try a very gentle tapping action. Another milder way is to intentionally introduce a air plug via syringe which washes off the tiny annoying bubble sitting in the detector. The electrical interference can come from anything used nearby may in another room, if the power line is the same.

[cafeguy]

Thank you every one!!!!

The pressure is down on the column. I think it has something to do with the packing of the column since it hasn't been used before, but was sitting in storage for over five years. Right now, if the pressure goes up, I find that just running 40% acetonitrile through it for 30 minutes reduces the pressure. So that's the plan from here on out...just running this whenever I notice the pressure has climbed.

As for that weird peak, I determined it was a bubble in the detector. I manually pushed some isopropyl alcohol through it to clean it and to push out any bubbles and it seemed to remove that repetitive peak!

So, here we go! Time for what hopefully turns out to be a lynchpin of an experiment for my thesis!!!!!

THANK YOU!