Chromatography Forum

I'm running an Agilent 1260 with DAD. I have found that my pressure is unstable during my runs - it also isnt reproducible. it seems there is a blockage which moves and then blocks again. A few days ago my LC switched off during the night as it exceeded the max pressure of the method - 300 bar - when i left it running it was around 230 bar. I've disconnected my detector and it does seem there is quite a big pressure drop over the DAD. I've tried back flushing (Water, ACN) and the pressure can be different, but again it is never reporducible. Sometimes i see a pressure drop of 20 bar @ 1mL/min, 40 degC, no column, pure water and then i can disconnect and reconnect and it can be 35 bar or even higher. Sometimes if i reconnect in the opposite direction the pressure drop can even be 10-15 bar different, but again it isn't reproducible.

I've had this problem for a few months where whenever i disconnect and reconnect my detector i get pressure issues. I'm obviously worried about cracking my DAD flow cell, especially if there is a particle in there. But how would i get this out??

I've included a chromatogram of the pressure vs a normal run. It's gradient elution 99% Aq increasing the % of ACN. They're blank water injections.

I've also noticed spikes on the baseline. I know this a symptom of air bubbles right? nothing is leaking and everything has been tightened enough.

It is possible to disassemble the flow cell, so maybe you could open it up and flush out anything that might be trapped inside. Agilent sells a repair kit for it that includes the wrench and new seals

https://www.chem.agilent.com/store/prod ... 3ECS_42454

You might want to give that a try.

Spikes on the baseline can have multiple reasons. If you have bubbles in system (for example due to poor degassing) you will also see pressure drops and baseline fluctuations (wavy baseline) rather than spikes. However contamination in cell can also faciliate collection of microbubbles just in cell. Do you remember when these problems started? Do you work with buffers which can crystallize in cell? If yes, you can try to clean this cell by pulling warm water (even about 60oC) through it and than you can clean cell with 20% nitric acid .

There are a lot of possible reasons for your observations, but it is not the flow cell. Examples of a few things to ask yourself. It is essential that all mobile phases be fully degassed, the solvent pickup frits clean, the mobile phase solutions are clean and have been filtered. Is this the case? Has the instrument been properly and regularly maintained (esp. Have you installed a NEW PTFE Frit in the pump, as this should be done every month)? Has the instrument been primed? Is the column NEW?

Here are some free articles which provide more troubleshooting and advice:
"Diagnosing & Troubleshooting HPLC Pressure Fluctuation Problems (Unstable Baseline)"; https://hplctips.blogspot.com/2014/01/d ... -hplc.html

"HPLC Retention Time Drift, Change, Variability or Poor Reproducibility. Common Reasons for it"; https://hplctips.blogspot.com/2015/11/h ... hange.html

I have for sure isolated it to the DAD flow cell. I ran 2 x blank gradients without the flow cell connected and the pressure was perfectly smooth and exactly reproducible.

There is someting inside my flow cell that i don't know how it got in there or what it is soluble in. My buffer is not really a buffer, but rather a pH modifier as it is only 10 mM H3PO4 so crystals are unlikely and if it was this, i'm sure i would have redisolved it by now with all the flushing i have done (warm buffer, warm ACN, warm water etc). It is something else but i have no idea what to do about it.

thanks for your suggestions

Which 1260-series DAD are you using? There are many types and variations. Some use the "Terrible" InfinityLab Max-Light cartridge cells. By chance do you have that type? If so, then it is possible you do indeed have a flow cell issue, esp if it is one of the longer path-length cells (e.g. 60 mm). Since their introduction, we have observed an extremely high failure rate and very short lifespan for the "Max-Light" cells. Unlike the nearly flawless standard cell design ("clothespin" style) which are extremely reliable, don't break and can be easily cleaned, the Max-Light cartridges are extremely prone to clogging, partial obstructions, fractures and leaks. They can not be repaired and the only solution available is to replace any defective ones with a brand new one. Flushing the flow cell with a buffer free solution is mandatory, every single day, to maintain them (*yes, 10 mM is still too much buffer and may clog a flow cell if not flushed out regularly as it can deposit inside the tiny silica capillary inside the cell). **Due to the extremely high breakage rate we have observed with these flow cells, we continue to advise our clients and also teach in our HPLC training classes to avoid these cells and the detectors that use them. We suggest instead, to purchase the Agilent detector versions which accept the std flow cell types.

BTW: Be warned that many equipment sellers and auction sites (e.g. ebay) routinely offer for sale these Max-Light cartridges, often described as "new" or unused, but in almost all cases they are someone else's broken one. They have no value.

Multidimensional wrote:
Which 1260-series DAD are you using? There are many types and variations. Some use the "Terrible" InfinityLab Max-Light cartridge cells. By chance do you have that type? If so, then it is possible you do indeed have a flow cell issue, esp if it is one of the longer path-length cells (e.g. 60 mm). Since their introduction, we have observed an extremely high failure rate and very short lifespan for the "Max-Light" cells. Unlike the nearly flawless standard cell design ("clothespin" style) which are extremely reliable, don't break and can be easily cleaned, the Max-Light cartridges are extremely prone to clogging, partial obstructions, fractures and leaks. They can not be repaired and the only solution available is to replace any defective ones with a brand new one. Flushing the flow cell with a buffer free solution is mandatory, every single day, to maintain them (*yes, 10 mM is still too much buffer and may clog a flow cell if not flushed out regularly as it can deposit inside the tiny silica capillary inside the cell). **Due to the extremely high breakage rate we have observed with these flow cells, we continue to advise our clients and also teach in our HPLC training classes to avoid these cells and the detectors that use them. We suggest instead, to purchase the Agilent detector versions which accept the std flow cell types.

BTW: Be warned that many equipment sellers and auction sites (e.g. ebay) routinely offer for sale these Max-Light cartridges, often described as "new" or unused, but in almost all cases they are someone else's broken one. They have no value.

Good information to know, I will definitely make sure this is not in any future quotes for equipment we may purchase. Thanks

thanks @multidimensional that is good information. It is a max light cartridge indeed. 1uL 10 mm 60bar max.

It's unusual as I've flushed and flushed with water/warm water/isopropanol/ACN everything and i cannot remove whatever is in there. it's something inoluble in organics and water. A particle of some sort.

We've ordered a new cell and I'm going to try and regenerate the broken one with something stronger (20% HNO3 for e.g) after finding some generic cleaning procedures online.

what i find unusual is that it runs fine isocratically with a 20 bar d.p @ 1mL/min water at room temp (injector and column bypassed), but as soon as i run my gradient (no column but injector loop connected) the pressure increases after each injection. The max pressure is set in my method to 80 bar (20 bar d.p over the capillaries before the flow cell) and withint 3-4 runs for sure he will tap the max pressure limit and shut off. These are zero volume injections too. I'm now not sure its definitely the flow cell but rather something coming from further up the system that gets lodged.

I'm worried if i install my new cell and run as i am running now that there will be the same issues again and again further down the line. The whole system is only ~8 months old in use.

kate

As noted earlier, the problem still sounds like it is one of the other items I mentioned (such as not replacing the pump PTFE frit) and proper troubleshooting is needed to narrow down the cause. With the extremely poor reputation that the Max-Light cartridge cells have for plugging and clogging, it is still possible that one of the connection made to the cell was done incorrectly and some of the glass particles have lodged themselves inside the capillary causing the pressure problems (in addition to buffer crystals getting stuck, glass is the #1 cause of failure with these). This is extremely common with the 60 mm path length cell and slightly less common with the 10 mm path length cell as you have (but all of them are bad, and should be avoided). You can flush all you want with any solution, but are not likely to clear the obstruction (if that is indeed your problem) since it if often glass that you are trying to dissolve/remove.
*Correct fitting selection is extremely important to avoid glass particles entering the flow cell. If you have not had training in how to make these types of connections (non-std HPLC connections), please do not connect the new flow cell to the system. Wait till after you have made sure you have the correct fittings on hand and training to know how much torque to apply when connecting them.

Do you have a standard Agilent analytical flow cell available (just the flow cell, no detector needed)? I refer to any of the Agilent clothes-pin style flow cells. If you have one available, try installing it in place of the Max-Light cartridge and run your gradient 8 times. Depending on the flow cell you select, the cell volume may be larger (i.e. 13 ul vs 1 ul), but the system back-pressure is all we are testing. Observe the method run pressure traces and compare system pressures during the gradient. This is one of the basic troubleshooting techniques we teach, to rule out one-item-at-a-time. No detector needed for the test as your complaint deals initially with back pressure. In addition to this test, you should review the two articles provided earlier for other possible causes too. If possible, remove the column from the system too, when running other test (replace it with a properly calibrated restriction capillary). Be methodical, change one-thing-at-a-time.

It is definitely the flow cell. if i disconnect it and run the gradient the backpressure is fine with no spikes or jumps and 100% reproducible. I bypassed everything including the injector valve etc and just connected the pump directly to the flow cell.

With regard to the improper fitting, where does the glass particle come from??
I'm using proper agilent fittings, stainless steel tubings and then hand tight peek fittings on the outlet. I don't understand where i would get something glass from here.

kate

"I bypassed everything including the injector valve etc and just connected the pump directly to the flow cell."

- I hope I misread your above post because if you connected that tiny 1 ul volume capillary flow cell directly to the pump, then doing so could easily over-pressure it, fracturing the glass capillary inside (Max-Light cells are made up of a glass capillary). When a flow cell is connected to the outlet of a column, the back-pressure it should see on the flow cell inlet port should be at or near ATM. Connecting the inlet port of a flow cell directly to an HPLC pump would result in extremely high pressures being generated at turn-on to the flow cell inlet port through the "flow cell" (just a capillary in this case) to the narrow outlet port. What I suggested was to run the entire HPLC system in a normal manner (with column, injector etc installed), but instead of connecting the Max-Light cell, connect a conventional flow cell and monitor the system pressure during an analysis run. That would give you a scientific comparison to evaluate with only the flow cell being the test variable.

The glass particles come from the Max-Light flow cell. Incorrect swaging of the fittings (i.e. wrong depth using PEEK fingertight or too much torque) can crush the end of the fused silica capillary inside releasing glass particles into the tiny capillary that makes up the "flow cell".

Yes you read correctly - but i inserted tight backpressure restrictions on my pump and evaluated pressure with an without connections. No overpressure was applied and i'm still able to measure samples and nothing is leaking.

As for the glass - that seems like a possibility. if the capillary was damaged surely it would leak?

kate

If by "restrictions" you mean you used the high pressure alarms on the pump, then they would be unable to protect the flow cell from such quick and dramatic pressure changes caused by connecting it directly to the pump. The pressure builds so quickly that the buffered readout from the sensor will not be able to act fast enough to shut down before damage is done. Having a column inline (normal operation) acts like a pressure buffer. That buffer would reduce the pressure seen by the flow cell substantially. Unless you have a splitter in front on the cell, never connect a flow cell directly to a pump, esp to flush it out (unless it is rated for the same pressure or has a relief feature like the std clothespin style Agilent flow cells have.

When fractured, these "special" cells, may or may not show a leak. When the tubing is internally fractured, glass particles partially obstruct it resulting in poor performance and/or higher than normal back-pressures. (2) If the fracture is large enough, then liquid will fill the inside of the plastic cartridge housing over time and eventually leak out (depending on what the liquid is and the flow rate used, this may take some time to observe). This is why they are normally thrown in the trash when damaged. They can not be repaired. We know of many users who have gone through several cartridges in a short period of time ($$$$) while they learn to use them with their system. It can be a very expensive learning curve and if not careful, could exceed the cost of the detector. Many labs have drawers filled with broken ones.

So my new detector came and the pressure drop is about 8 bar @ 1 mL/min 99% aqueous.

i opened the casing of the other flow cell to visually inspect and there it no evidence of leaking at the inlet or outlet and the capillary is intact too.

Any suggestions on solvents i can use to try and dissolve whats in there? I was thinking of a 20 % HNO3 or Formic acid solution.

8 bars with 100% pure water at 1.00 mL/min seems a bit low for the 1 ul flow cell... If you had 20 bars under the same conditions for the "suspect" cell, then it may or may not have some internal obstructions. Perhaps the real question should be can you run the same analysis 30 times in a row, on the same column and HPLC w/o having the pressure slowly climb over time?

Flow Cell Wash solution: Did you see my earlier comments? "When fractured, these "special" cells, may or may not show a leak. When the tubing is internally fractured, glass particles partially obstruct it resulting in poor performance and/or higher than normal back-pressures." - So no, a leak is not required to have a flow cell with glass particles stuck inside it. Glass particles will not be dissolved by the wash solutions. *This is why some of these flow cells end up in the trash can.

You comments still seems to indicate the problem observed may not be related to the flow cell. The intermittent changes in system pressure over time still imply another issue, as previously noted or just a fouled column.