Amino acid analysis column life time

Discussions about HPLC, CE, TLC, SFC, and other "liquid phase" separation techniques.

37 posts Page 1 of 3
I have been analyzing for amino acids using a Hypersil Gold C18 column, gradient elution at 40 C with borate/phosphate buffer and acetonitrile/methanol/water, and automatic in-needle derivatization with OPA/MPA. I'm finding that the column seem to be failing already with less than 150 injections (retention times shifting forward, loss of resolution) and I have changed the guard column once already. Does anyone have an idea of the expected life time for this kind of analysis?

This column was used exclusively for this analysis. Every time I start a new run, the retention times shift forward and some of the closely eluted peaks start merging together. Changing the guard column helped momentarily but have now gotten worse. I have tried back flushing the column once before with no success. Because the shift seem to be happening only while column is not in use, I was suspecting that maybe the buffer is not getting rinsed out fully so after changing the guard column I started adding a 90% water rinse at the end of my runs to see if that would help, but it didn't.

Image
Black - new column
Blue - before new guard
Pink - after new guard

Image
Brown - ~20 or so samples after putting in new guard

Any suggestions for restoring resolution if possible and preventing this from getting worse in the future?

Thanks in advance.
What kind of samples are you injecting?

Can you post the full gradient method?

Is the pressure significantly different between good and bad runs?
What is the pH of that buffer? Thermo claims a pH range of 2-9 for Hypersil Gold, but I would expect lifetime to start dropping above pH 8.

At this point, you will probably need to replace your existing column. Going forward, changing guard cartridges more frequently may help, but you'll have to balance that cost against the cost of frequent column changes.

One way of looking at it is that your columns costs ($2 - $3 per sample) are probably still small relative to your time costs (how long do you spend per sample and what's your hourly rate).
-- Tom Jupille
LC Resources / Separation Science Associates
tjupille@lcresources.com
+ 1 (925) 297-5374
Rndirk wrote:
What kind of samples are you injecting?

Can you post the full gradient method?

Is the pressure significantly different between good and bad runs?


The samples are filtered wastewater.

Before I started the 90% water rinse at the end of the run, there was a stepwise pressure increase at each sequence (no increase from run to run in the same sequence) but after starting the 90% water rinse the pressure came back down and never went back up.

Image
Black - new column
Brown - last run before guard column change and starting 90% water rinse
Rest - runs after starting 90% water rinse

Image
This is the gradient.
A: 10mM Na2HPO4, 10mM Na2B4O7*10H2O, adjusted to pH 7.8 with conc HCl
B: acetonitrile/methanol/water (45/45/10 v/v/v)

Thank you for your help
tom jupille wrote:
What is the pH of that buffer? Thermo claims a pH range of 2-9 for Hypersil Gold, but I would expect lifetime to start dropping above pH 8.

At this point, you will probably need to replace your existing column. Going forward, changing guard cartridges more frequently may help, but you'll have to balance that cost against the cost of frequent column changes.

One way of looking at it is that your columns costs ($2 - $3 per sample) are probably still small relative to your time costs (how long do you spend per sample and what's your hourly rate).


The pH of the buffer is 7.8 (adjusted with HCl).

I'm in a university lab so there is no charge involved, but I totaled 134 injections, and only 83 actual samples. So that works out to about $10 per sample using 1 analytical and 2 guard columns. I guess it's not extremely costly but still seems a bit high.

I'm not sure if I'm doing something wrong that I don't realize, or this analysis just is very hard on columns...

Thank you for your help
(1) "90% water rinse" is NOT a column wash method. You should be "washing" the column after each run using a solution which is stronger than the mobile phase, not weaker than the mobile phase. Water would not wash anything off, and in fact, it probably would assist in retaining compounds on the column to foul it.

(2) Get rid if the probably worthless guard columns. They primarily exist to make money for the column vendors. *Filter your samples! **Always, always make sure your samples are 100% dissolved in the mobile phase before injecting them!

(3) Find and use an experienced chromatographer at your school and have them review your actual method. They can probably identify a number of areas (like the lack of column wash) where you can improve how you are doing things. You are just starting out with HPLC and amino acid analysis takes years to be proficient in (that is a bad method to start out with for a beginner). I doubt you have the available time just to learn this one technique. Instead, invest the time in having someone experienced supervise the analysis (or just pay them to run the analysis for you so you can work on whatever your primary project is).

(4) Depending on the source of the samples (matrix) and its complexity, your column may last from 200 to 2,000 samples. Columns are consumable items. Their lifetime is a function of the purity and cleanliness of the samples, the quality of the wash solution used, the strength/types of the solvents it is exposed to and to some extent, the pressures it is exposed to (though, pressure is less important than people think).
Hi Crong,

Here is a link that provides Agilent's perspective on OPA derivatization:

https://www.agilent.com/cs/library/slid ... 0Steed.pdf

Page 17 is the most useful, perhaps. The separation conditions are similar to the ones you note above.
MattM
Multidimensional wrote:
(2) Get rid if the probably worthless guard columns. They primarily exist to make money for the column vendors.


I'd like to hear other opinions about this.

In my personal experience, even if working with filtered samples and mobile phases, it's inevitable that the column still catches undissolved stuff over time. It not only increases the pressure, but can influence the peak shape of some compounds. I'm not fond of guard columns myself (I hate replacing them), but i think they do what they promise: protect the column. Note that i'm not associated with any column vendor.
Multidimensional wrote:
(1) "90% water rinse" is NOT a column wash method. You should be "washing" the column after each run using a solution which is stronger than the mobile phase, not weaker than the mobile phase. Water would not wash anything off, and in fact, it probably would assist in retaining compounds on the column to foul it.

(2) Get rid if the probably worthless guard columns. They primarily exist to make money for the column vendors. *Filter your samples! **Always, always make sure your samples are 100% dissolved in the mobile phase before injecting them!

(3) Find and use an experienced chromatographer at your school and have them review your actual method. They can probably identify a number of areas (like the lack of column wash) where you can improve how you are doing things. You are just starting out with HPLC and amino acid analysis takes years to be proficient in (that is a bad method to start out with for a beginner). I doubt you have the available time just to learn this one technique. Instead, invest the time in having someone experienced supervise the analysis (or just pay them to run the analysis for you so you can work on whatever your primary project is).

(4) Depending on the source of the samples (matrix) and its complexity, your column may last from 200 to 2,000 samples. Columns are consumable items. Their lifetime is a function of the purity and cleanliness of the samples, the quality of the wash solution used, the strength/types of the solvents it is exposed to and to some extent, the pressures it is exposed to (though, pressure is less important than people think).


Thank you for your advice.

(1) The water is used to rinse off any buffer in the column, not to clean the column. I should clarify that this is done at the end of the sequence prior to shutting down the instrument (not at end of each sample). Before adding the water rinse, the pressure was increasing with each new sequence, hence my suspicion that buffer was still left in the column and precipitating. Since adding the water rinse, the pressure has stopped increasing, however that did not stop the continued loss of resolution. So to me that means the loss of resolution isn't related to any kind of clogging or precipitation.

(2) Samples are all filtered to 0.45 um or 10kDA through microfiltration. They are aqueous samples to start with so no issue with dissolution I would think.

(3) The method is an exact transfer from an application note provided by the instrument manufacturer (Thermo Scientific) only with a different column and flow rate and gradient changed to accommodate my column size.

(4) Well this is part of my question, for those who have experience with amino acid analysis, whether it is normal to have such a short life time and if not, what can be done to improve it? I'm no expert but I have never experienced this short of a life time on any kind of column.
mattmullaney wrote:
Hi Crong,

Here is a link that provides Agilent's perspective on OPA derivatization:

https://www.agilent.com/cs/library/slid ... 0Steed.pdf

Page 17 is the most useful, perhaps. The separation conditions are similar to the ones you note above.


Thank you so much, this is what I wanted to know. However, it appears my column lifetime is still much lower than 500 injections. Any suggestions for what could be the cause?

Other than not using a dedicated amino acid column (e.g. from Agilent or Waters), I can't see any reason at this time for why this is happening...
Hi Crong,

I agree with Tom here, the likely source of the stationary phase's longevity issue is the pH of the eluent. There's not much you can do about this, unfortunately...silica gel dissolves readily at a pH value of 8 (you're working at 7.80.)

I've done a fair amount of amino acid work in my past, including FMOC derivatization, Waters Accu.Tag, Phenomenex' EZ:Faast and Dionex using integrated pulsed amperometric detection, amongst others, though not OPA. It seems to me that you're already doing what you can. As I recall, Accu.Tag was the most gentle toward the stationary phase--I swapped out guard columns every 150 injections and a typical column back then lasted 1500 injections or so. I would not even attempt to use those NovaPak materials at pH 7.8, though...it may not be so easy to switch stationary phases without changing the derivatization, though. I don't know what options are open for you here...what do you think?
MattM
Depending on 'your' amino acid analysis samples... we do know of plenty of methods were the column are useless after as few as 200 samples. As I said, 200 to 2000. It depends. You may be one of the '200', and if so, please consider using a better column wash system (you did not tell us what your column wash is) to prolong the column lifetime, but do not expect it to double. Columns are consumable items and the method application that you are using is very harsh, so the weak point will be the column.
mattmullaney wrote:
Hi Crong,

I agree with Tom here, the likely source of the stationary phase's longevity issue is the pH of the eluent. There's not much you can do about this, unfortunately...silica gel dissolves readily at a pH value of 8 (you're working at 7.80.)

I've done a fair amount of amino acid work in my past, including FMOC derivatization, Waters Accu.Tag, Phenomenex' EZ:Faast and Dionex using integrated pulsed amperometric detection, amongst others, though not OPA. It seems to me that you're already doing what you can. As I recall, Accu.Tag was the most gentle toward the stationary phase--I swapped out guard columns every 150 injections and a typical column back then lasted 1500 injections or so. I would not even attempt to use those NovaPak materials at pH 7.8, though...it may not be so easy to switch stationary phases without changing the derivatization, though. I don't know what options are open for you here...what do you think?


Hi Matt,

Dissolution of the stationary phase could definitely explain what I'm seeing. Although I do wonder if that is the case, I should be expecting to see retention time shift forward with each sample right? For me the Rt only shifts forward when the instrument is shut down and restarted next time. During continuous runs in the same sequence the Rt is stable. That just makes me think I'm doing something wrong in the shut down process...

I'm looking into Agilent's columns since their derivatization process is pretty much the same, and Waters AccQ Tag seems very expensive. I recently watched a webcast from Agilent about their amino acid columns and it does advertise chemical modifications to the silica for high pH resistance. However in the same webcast they also recommend to recalibrate for retention times and response factors weekly, which to me says - expect peaks to shift forward with use. So how much better is it really?!

Thank you again for your insight.
Multidimensional wrote:
Depending on 'your' amino acid analysis samples... we do know of plenty of methods were the column are useless after as few as 200 samples. As I said, 200 to 2000. It depends. You may be one of the '200', and if so, please consider using a better column wash system (you did not tell us what your column wash is) to prolong the column lifetime, but do not expect it to double. Columns are consumable items and the method application that you are using is very harsh, so the weak point will be the column.


I hope I understood correctly - during each run the gradient is ramped up to 100% B (45/45/10 ACN/MeOH/H2O) and held for 3 min before ramping back down. This is pretty much following the application note here adjusted for my column size (100mm x 2.1mm x 3um)

http://tools.thermofisher.com/content/s ... 849-EN.pdf

What would you recommend for a better column wash?

Thanks again.
This applies to all HPLC methods: A proper "wash" should be stronger than the mobile phase and do a better job of fully dissolving everything injected.

"100% B (45/45/10 ACN/MeOH/H2O) and held for 3 min". So that means your column never "sees" more than 45% ACN / 45% MeOH (basically, 45% ACN since that is the strongest solvent). Therefore, you should use a wash solution which exceeds those values.

Perhaps 10% Water plus 90% ACN (or 5% Water/95% ACN); or a mixture of 80% ACN plus 15% MeOH plus 5% water. You should experiment to see what works best.
37 posts Page 1 of 3

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 1117 on Mon Jan 31, 2022 2:50 pm

Users browsing this forum: No registered users and 1 guest

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food & Beverage, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry