What could be the reason for this HPLC analysis?

[Sandra7]

Last Thursday, I prepared a standard mixture of 5 compounds of with known concentration (50 ug/mL) in MeOH and optimized a method development.
The compounds are
(1)
(2)
(3)
(4)
(5)


The method is
SP: C18
MP: H20 and ACN
Flow rate: 1ml/min; pressure limit=250
Gradient program is
At 0 min, 76% H20 & 24% ACN
At 8 min, 70% & 30% ACN
At 13 min, 64% H2O & 36% ACN
Total time=13 min; post time=3 min

The order of elution and RT are:
(2)
(1)
(3)
(5)
(4)

Today, using the same EXACT method I analyzed SPE extraction of the same 5 compounds dissolved in H2O (5 ug/mL each). I used 50ml of the aqueous std sample and used 5 ml MeOH in extraction. So, it is 50 ug/mL each.

When I analyzed it with the same EXACT method, the retention times for all 5 compounds were much shorter with the last peak coming out at 8.129 min. Here are the RT:
The order of elution and RT are:

(2)
(1)
(3)
(5)
(4)

Note that H2O used in MP is supposed to have 1% MeOH in it to prevent mold forming. Noticing that the bottle containing H2Ois full, the prof said that whoever refilled it may have put more MeOH in it.

Right away, I tried to adjust the second part of MP in gradient program like show below (giving total run time to be 9 mins; post time 3 min) but the last peak didn't come out until 1 1/2 min after post time. It came out at 13.5 min. RT for others were changed too though the first part of the gradient program was kept the same.
Gradient program is:
At 0 min, 76% H20 & 24% ACN
At 8 min, 70% & 30% ACN
At 9 min, 69% H2O & 31% ACN
Total time=9 min; post time=3 min
RT are:
(2)
(1)
(3)
(5)
(4) (didn't come out during the analysis but I saw a peak coming out 1 1/2 min after post time. I took a screen shot.

So I tried adjusting the gradient program again like this:
Gradient program is
At 0 min, 76% H20 & 24% ACN
At 8 min, 70% & 30% ACN
At 12 min, 65% H2O & 30% ACN
Total time=13 min; post time=4 min
(My logic of keeping ACN to be 30% is that in the previous run, the last peak came out 1 1/2 min after post time of 3 mins, meaning the MP had gone back to original strength? I know that I should have given post time to be 5 mins
RT are:
(2)
(1)
(3)
(5)
(4) (didn't come out during the analysis and I don't know whether it did during post time or after ward because I wasn't near the computer.

Any suggestion what I should do to shorten the analysis time?
Note: The first part of the gradient program, i.e. form 0 min to 8mins can not be changed to ensure that the critical pair, namely A and B , come out with Rs>1.5.

Tomorrow, I would like to improve this method to shorten analysis time before I run my unknown. I would also re-analyze the first standard prepared using MeOH (without SPE extraction). May be I should re-analyze that first tomorrow using the same exact optimized method that gave good separation between A and B tomorrow before analyzing the SPE extracted standard mixture and SPE extracted unknown to determine whether the RTs have changed for the 5 compounds. IF the RTs have changed, that means so changes that I cannot control has happened between last Thursday and now.

Any idea how I can shorten the analysis time given the current info.

[Khanom]

Hi Sandra7,

Have you thought about the use of a column heater chiller. This will keep your column temperature constant, on a daily basis.

Kind regards,

Marketing
Cecil Instruments Limited
Milton Technical Centre, Cambridge CB24 6AZ
United Kingdom

email:- ade.kujore@cecilinstruments.com
telephone:- +44 (0) 1223 420821
fax.:- +44 (0) 1223 420475
web site:- http://www.cecilinstruments.com

[Peter Apps]

Hi Sandra

Why mess about trying to recreate the separation with a mobile phase composition that is compromised by the possible addition of extra methanol ? Just mix up a new 1% methanol solution and run the original gradient, or have you done that already and you now want to speed up the analysis ?

There is no way that the SPE step can affect the rention times.

Peter

[Consumer Products Guy]

Hi Sandra7,
Have you thought about the use of a column heater chiller. This will keep your column temperature constant, on a daily basis.

What - Sandra's not using temperature control for the column ? Then good luck getting reproducibility. Why use room temperature like in the 80s (I mean the 1980s !!!) ? I hope you're also degassing your solvents.

Also, I wouldn't bother with 1% methanol in the aqueous part of the mobile phase. You will get some evaporation as that reservoir vents and sits, so its composition actually changes a little as time goes by. If you and "prof" are worried about microbes growing in plain water, just change the water every few days. We don't add anything to our water mobile phase here (30 years) unless it's a pH modifier. We don't have microbe issues.

[WillyOne]

Another possibility is that the outlet check-valve is not working really well.
Perhaps you should clean the lines with 100% water at high flow (10 ml/min) AFTER have disconnect the column

[Sandra7]

Hi Sandra

Why mess about trying to recreate the separation with a mobile phase composition that is compromised by the possible addition of extra methanol ? Just mix up a new 1% methanol solution and run the original gradient, or have you done that already and you now want to speed up the analysis ?

There is no way that the SPE step can affect the retention times.

Peter

Here are the 4 analyses I did today.
(1) Analysis (re-analysis really) of the SPE extracted version of the 5 std compounds (50ug/mL each) using the original optimized method as done yesterday.
Method is
SP: C18
MP: H20 and ACN
Flow rate: 1ml/min; pressure limit=250
Gradient program is
At 0 min, 76% H20 & 24% ACN
At 8 min, 70% H2O & 30% ACN
At 13 min, 64% H2O & 36% ACN
Total time=13 min; post time=3 min
(b)Conclusion:(/b) Got the same weird result as yesterday, i.e.earlier retention time like yesterday.

(2) Analysis of my unknown compounds using the same exact method
(b)Conclusion:(/b) Couldn't decide until I get result of no.1 to have the same ret time as in the std mix prepared in MeOH.

(3) Analysis of a standard mixture of 5 compounds of with known concentration (50 ug/mL) in MeOH; this has been analyzed last week already
(b)Conclusion:(/b) Got the same result as before, i.e. last week when method optimization was achieved.

(4) Repeat no. 1, i.e. another analysis (re-analysis) of the SPE extracted version of the 5 std compounds (50ug/mL each) using the optimized method.
(b)Conclusion:(/b) Now, this time, I get the same ret time as in the analysis above, i.e. the Std Mix prepared in MeOH and used in method development.

So, I was happy to see SPE extracted std mix finally giving the same retention times as the std mix prepared in MeOH. So, I am leaving my method as it is.

Willyone suggested this: "Another possibility is that the outlet check-valve is not working really well.
Perhaps you should clean the lines with 100% water at high flow (10 ml/min) AFTER have disconnect the column".

Thanks.

[Sandra7]

Another possibility is that the outlet check-valve is not working really well.
Perhaps you should clean the lines with 100% water at high flow (10 ml/min) AFTER have disconnect the column

Do you mean disconnect the column after cleaning the lines or clean the line and then disconnect the column?

[bisnettrj2]

WillyOne means disconnect the column before attempting the cleaning process. Depending on your system, you should be able to open a valve to divert your pump flow to waste while attempting the cleaning.

One question: What are the dimensions of your column? I see that you are running at 1 mL/min, but I do not believe you have stated the dimensions of your column. I'm wondering if you are re-equilibrating your column for a long enough time after the end of your gradient. 3 minutes post-time, depending on your column dimensions, could be only 1-2 column volumes, which may not be sufficient to completely return your system to the original starting conditions of the gradient.

My advice would be to use 100% water in the A reservoir (as others have suggested - just replace it when you are going to run your samples, but be sure to clean the reservoir before re-filling), and to add at least 5 column volumes of post-time to your run, to rule out any re-equilibration issues. Run a couple system blanks, then run a couple standards, and check the retention times.

Another question - are you injecting 100% MeOH? If so, what is your injection volume?

[WillyOne]

Sorry my english level is quite bad.
I propose the possibility that the outlet check-valve of your HPLC system is dirty. It means that small particles or cristalizations into it frustrates the closure of the valve.
The outlet check valve impede that the flow returns to the pump when the piston retracts.
That problem presents as typical sympton erratic RT and also, in severe cases erratic pressure readings.
Sometimes the seat is damaged and the only solution is changing the valve, but before that perhaps you can try to clean it.
Just disconnect the column from the system and store it. Or better than that disconnect the flow line just before your injection valve
Then pump water for 15 min at high flow i.e. 5 to 10 ml/min directing the flow to a lab vessel. After that rebuild your system and try again your Analysis

[Sandra7]

WillyOne means disconnect the column before attempting the cleaning process. Depending on your system, you should be able to open a valve to divert your pump flow to waste while attempting the cleaning.

Thanks.

One question: What are the dimensions of your column? I see that you are running at 1 mL/min, but I do not believe you have stated the dimensions of your column.

Column dimension is not given in this handout but some info was stated in the first HPLC lab which used the same instrument. Column is 25 cm long with 5um spherical beads. I do not have the diameter.

I'm wondering if you are re-equilibrating your column for a long enough time after the end of your gradient. 3 minutes post-time, depending on your column dimensions, could be only 1-2 column volumes, which may not be sufficient to completely return your system to the original starting conditions of the gradient.

I see. That's what the prof was trying to conclude yesterday (Wed) except that the previous day (Tuesday), after analyzing the 1st aliquot of SPE extracted Std Mix first, I analyzed the 2nd and 3rd aliquot (meaning plenty of time for this re-equilibration to happen), before reanalyzing the 1st aliquot two more times using slightly adjusted method (the first part of the method was left the same, i.e. the gradient program was kept the same from 0min to 8min, adjusting only from 8min on) to see whether the last peak could be eluted a little earlier. Still, no changes in that last 2 runs of 1t aliquot of SPE extracted Std Mix (except that the last peak didn't elute during the program time).

My advice would be to use 100% water in the A reservoir (as others have suggested - just replace it when you are going to run your samples, but be sure to clean the reservoir before re-filling),

Since the reservoir is pretty full, the dept. or the prof wouldn't be happy if I try to refill.

and to add at least 5 column volumes of post-time to your run, to rule out any re-equilibration issues.

I would get the radius (diameter) of the column to calculate the column volume. Since column length is given in cm (it is 25 cm), I will calculate for volume in cm^3 since currently flow rate is 1ml/ min which i believe is the same as 1cm^3/min, right?

So, once I calculate the volume of the column (in cm^3), I divide column volume by the current flow rate to get post time, right?

Run a couple system blanks, then run a couple standards, and check the retention times.

What does running system blanks mean? Just leave the system for about the same time of the analysis? Today, I shall do blank runs, then run the std. mix (would it matter if I use the one prepared directly in meOH vs 1st aliquot of SPE extracted version?), before proceeding with runs for quantitative analysis.

Another question - are you injecting 100% MeOH? If so, what is your injection volume?

With limited instrument use time, I haven't done that yet. After running the system blanks, I will run 100% MeOH today prior to running Std. Mix. By then the lamp would have been on long enough for quantitative analysis as well.

If I don't get column diameter info:, should I increase post time to 5 mins?

Thanks.

[Sandra7]

Sorry my english level is quite bad.
I propose the possibility that the outlet check-valve of your HPLC system is dirty. It means that small particles or cristalizations into it frustrates the closure of the valve.
The outlet check valve impede that the flow returns to the pump when the piston retracts.
That problem presents as typical sympton erratic RT and also, in severe cases erratic pressure readings.
Sometimes the seat is damaged and the only solution is changing the valve, but before that perhaps you can try to clean it.
Just disconnect the column from the system and store it. Or better than that disconnect the flow line just before your injection valve
Then pump water for 15 min at high flow i.e. 5 to 10 ml/min directing the flow to a lab vessel. After that rebuild your system and try again your Analysis

Thanks.

[bisnettrj2]

My advice would be to use 100% water in the A reservoir (as others have suggested - just replace it when you are going to run your samples, but be sure to clean the reservoir before re-filling),

Since the reservoir is pretty full, the dept. or the prof wouldn't be happy if I try to refill.

and to add at least 5 column volumes of post-time to your run, to rule out any re-equilibration issues.

I would get the radius (diameter) of the column to calculate the column volume. Since column length is given in cm (it is 25 cm), I will calculate for volume in cm^3 since currently flow rate is 1ml/ min which i believe is the same as 1cm^3/min, right?

So, once I calculate the volume of the column (in cm^3), I divide column volume by the current flow rate to get post time, right?.

They would have a problem with filling the reservoir with clean WATER every day? It's just water! It's not like you're dumping out the acetonitrile! In any case, exchanging the water is more precautionary than anything. However, if the reservoir was 'topped off' with either water or methanol (as your prof posited), the only solution is to discard the contents of the reservoir and re-fill with a known liquid (either 1% MeOH or 100% water).

As for calculating the column volume, there are several calculations you can use to get to approximately the same number. However, you can just refer to the chart in this link for the volume associated with your column dimensions, after allowing for the pore volume of the packing material.

http://www.chiralizer.com/colvol.htm

And you are correct, divide by your flow rate to get the time in minutes to complete a column volume. You probably have a 250x4.6mm column (pretty standard dimension), which means it should take about 3 minutes for one column volume to pass through the column. If you want 5 column volumes, that means you need 15 minutes of post-time at the original conditions of your run.

[Sandra7]

As for calculating the column volume, there are several calculations you can use to get to approximately the same number. However, you can just refer to the chart in this link for the volume associated with your column dimensions, after allowing for the pore volume of the packing material.

http://www.chiralizer.com/colvol.htm

And you are correct, divide by your flow rate to get the time in minutes to complete a column volume. You probably have a 250x4.6mm column (pretty standard dimension), which means it should take about 3 minutes for one column volume to pass through the column. If you want 5 column volumes, that means you need 15 minutes of post-time at the original conditions of your run.

Yes, it is 4.6 mm.

Today, first, I ran 100%MeOH before doing any analysis (different conc: of the same std compound). I used 5 mins for post time. It worked out fine. Next time, I will do the same, i.e. run 100% MeOH using the same method before I do any analysis.

Thanks.

[Sandra7]

They would have a problem with filling the reservoir with clean WATER every day? It's just water! It's not like you're dumping out the acetonitrile! In any case, exchanging the water is more precautionary than anything. However, if the reservoir was 'topped off' with either water or methanol (as your prof posited), the only solution is to discard the contents of the reservoir and re-fill with a known liquid (either 1% MeOH or 100% water).

The water level in there was low. I refilled it today.

[Sandra7]

Another question - are you injecting 100% MeOH? If so, what is your injection volume?

10ul.