Agilent LC 1100 HPLC Not Showing Standards, Strange Peaks.

[PeterSDSU]

Hello everyone, so I'm trying to use an Agilent LC 1100 HPLC to separate and quantify the different types of vitamin B12. I've been at it for a few weeks now but I can't get my standards to show any peaks. I'm dissolving my standards in water. I'm using a method I found from a paper linked here (Click View PDF below, PDF link was massive):

https://www.sciencedirect.com/science/a ... via%3Dihub


Here's my info:
Solvent A: 0.1% acetic acid buffer titrated to pH 3.5 with NaOH
Solvent B: acetonitrile with 0.1% acetic acid
Flow rate: 1mL/min
Column: Agilent Zorbax SB C18 4.6x250mm 5 micrometer particle size
The method is:

0-2 min: 5% B
2-14 min: 15% B
14-19 min: 18% B
19-32 min: 35% B
32-33 min: 60% B
33-35 min: 5% B

I've ran this a few times and all I see are big peaks right around 5 minutes at the beginning of the run, and they fade away into smaller peaks. I was wondering if this was from my standard but then I ran it with a vial of water instead of my standard and I saw the same peaks. The standard I injected was 0.1mg/mL cyanocobalamin, and I injected 5 microliters. When I ran the plain water I also injected 5 micrometers. Pictures are below.

Cyanocobalamin try 1:


Cyanocobalamin try 2:


Water (Ran directly after cyanocobalamin try 2):


I'm also noticing pressure fluctuations during the run of about 5-10 bar. The pressure decreases generally as the percentage of pump B increases. At the beginning it might go from 74 to 79 bar over the course of 6 seconds, and at 60% B I will be going from 39-50 bar over he course of a few seconds.

Not sure if it means anything, but when I'm priming the pump after turning it on, and I'm using 100% A, I can see the solvent/ little air bubble (It forms after turning the machine off for the day in the clear tubes going into both outlet valves.) going into the active inlet valve on pump B actually getting pushed backwards a small amount after each pump. After switching to 100% B to prime that side this air bubble eventually goes away and comes out the purge valve and the pump runs normally.

There's no leaks that I can see, outlet ball valves were recently changed, piston seals looked good, and were changed on one side where needed.

It's very strange because this machine was working before I changed the column, and solvents. It successfully separated and detected formate but this method only used one solvent. Any help or insight is greatly appreciated!

[Multidimensional]

It appears that you are using HPLC for the first time with no professional training or experience. Sorry to sound so blunt, but it is obvious and does not allow us to provide useful information to assist you (as you need industrial training to understand how to operate the system and collect data, This is not a UV/VIS spectrophotometer). Acquiring the several years of training to have just a basic level of understanding is a waste of time if you want to do this now. Instead, hire someone that already has the training to avoid the collection of invalid data. PLEASE contact a professional chromatographer for local assistance. You need someone with experience to assist you. This will move your project forward... Don't try and figure it out yourself, that will not work.
'There are so many things wrong in the info provided, too many to list here, plus so much basic key missing information (such as exactly which column, with what dimensions are you using).

The reference link you provided is just an abstract, not the actual paper so no information is available to us w/o purchasing the article (instead of using a link, include the full method when asking for free help. Do not expect people to spend time looking up the basic details that YOU should be providing to help you).

BTW: (1) Please determine K prime first. This basic fundamental is key to starting and understanding any HPLC analysis. (2) Running "plain water" makes no sense. If you want to check the system, inject a blank (a blank is the mobile phase). (3) PLEASE turn OFF the UV/VIS REFERENCE WAVELENGTH feature that you have ON. It should always be "OFF". Signal B and D are irrational (you subtracted the signal you are monitoring). Signal A and C are made invalid by using the reference signal. You have it set to automatically SUBTRACT all signals between 310nm and 410nm from the output (the data is destroyed forever). Training is required to use a DAD. Failure to adjust the settings and data collection to valid values results in invalid data collected.(4) If you purchased this HPLC system used from one of the many lab equipment flippers, then the degasser is almost certainly broken and not degassing your mobile phase (instead, it is contaminating it) causing pump cavitation, bubbles and baseline noise. Before running samples: Have the HPLC pump (esp the MCGV) and A/S (if available) serviced too by an experienced professional to insure they meet the manufacturer's performance specifications too. Using a damaged HPLC wastes time and money plus can be very frustrating as you are unaware of which problems you are causing vs which are due to the broken instrument. (5) For free professional HPLC related training articles, please go to this link: https://hplctips.blogspot.com/

[PeterSDSU]

@Multidimensional Thank you for your reply, yes I am a beginner. I have been working on this and learning for about a year now. I have updated the post with the column I'm using, I apologize for forgetting that. It's an Agilent Zorbax SD C18 4.6x250mm column. The article is available as a PDF by clicking the link below the abstract, I didn't want to paste that link since it was so long.

Regarding your responses, (1) I will look in to the K prime and find out what it is. (2) The reason I said "plain water" was just to specify that I didn't take a pipette from my mobile phase to make a true blank. My first mobile phase is water with 0.1% acetic acid so I thought it would be similar. I'll try taking my actual mobile phase and injecting a blank. I think the question of what those peaks are at the beginning still stands though and I don't think it's from the water or sample I'm injecting since it happens with both.

(3)As for the reference wavelength, Chemstation requires you to set a reference wavelength when setting up the DAD signals. I talked to the professional chromatographer at my school and he said he never changes it from 360, but it must be set to something. According to multiple sources I've read it should be set to a wavelength that whatever you're looking for doesn't absorb at, so the DAD has a baseline absorbance of your mobile phase. That makes sense for B and D why it's just a straight line then, I was just trying stuff to see what happens.

(4) The machine was working previously for others in my lab. However their methods only required one solvent in channel A at a constant flow. Now that I've switched columns, solvents, and started using channel B as well with a more complex method I'm getting these weird results. I did fix the degasser (he tubing was cracked) and leaky outlet ball valves on both the pump heads.

(5) Thank you for that link, that seems like a very helpful website!

While it would be nice to just call someone or scrap this, I want to figure this out for myself.

[Consumer Products Guy]

As for the reference wavelength, Chemstation requires you to set a reference wavelength when setting up the DAD signals. I talked to the professional chromatographer at my school and he said he never changes it from 360

Strange, but depends on the wavelength used for the sought-for analyte.

For example, when we assayed sunscreen actives, we changed the reference wavelength from what we might use for analytes absorbing at 280nm.

According to multiple sources I've read it should be set to a wavelength that whatever you're looking for doesn't absorb at.

Yes.

[Multidimensional]

SO many untrained users and misinformation on the web and at your school. Trying to figure this out on your own will fail and is unscientific. You do not know what you do not know. You can not learn what you need to know in one year (or five, in this case). It is naive to think that you can sort this out on your own. Our company receives calls and emails from people in your same situation every week. They do not understand that what they are doing makes no sense and one day, one week or one year long class will not fix it.

Thank you for providing the basic HPLC column information.

"(1) I will look in to the K prime and find out what it is." The fact that you are using an HPLC system for one year and do not know what this is or have not been using it is frightening. Again, you can not begin to use HPLC without professional training. You will not get this training at your school!

"(3)As for the reference wavelength, Chemstation requires you to set a reference wavelength when setting up the DAD signals." - THIS IS 100% FALSE. You do not understand how to use the software or the DAD (as noted earlier). Click on the box next to it to TURN IT OFF!. Please refer to this authoritative article to learn more about this optional DAD software feature and why you should turn it OFF; [ https://hplctips.blogspot.com/2011/03/r ... -hplc.html ]. *The article was written by my boss who is a worldwide expert "in all things" chromatography.

"I talked to the professional chromatographer at my school and he said he never changes it from 360, but it must be set to something." Do not ask for advice from this person. They clearly have no professional HPLC training (and no training on ChemStation). "360" is just a filler value used to fill the field in the software by the manufacturer when it is turned on (fields can not be left empty if turned 'ON'). They never meant for anyone to use it and understood that a trained user would of course know this. Just click it OFF, problem solved. I wonder how many other settings are also wrong in the method?.

[Chromavore]

Welcome to the hellish world of chromatography! Unfortunately if you didn't pick up an engineers level of fluid dynamics, a physicists understanding of light, the statistical analysis skills of a mathematician and a software programming degree while studying chemistry then good luck troubleshooting a damn thing.

I'm joking, and I don't think it's ridiculous to think that you could recreate a method from a published journal or that the process of performing experiments, trying to interpret the data and then asking for help when it doesn't make sense is unscientific but you have somewhat fallen backwards into the trap that is hplc method transfer/development.

The reason people on this forum go nuts for details is we have seen an "insignificant change" like buying the same product from a different manufacturer or small changes to a piece of tubing destroy months of work. Or yeah having the wavelength setting wrong on the detector. There is also a healthy disrespect for hplc in a lot of labs iv been in where a lot of people just think we push a button and the big box gives us peaks so we get salty about it when it comes to the fundamentals.

The link you posted does just circle back round to a paywall, you probably have some kind of institution access set up since you mentioned school, how I very much miss that sweet institution access. If you could post more details of the original method, ANY changes you have made to that and why, what you think might be going on and what you have tested and what direction that's pointing you in maybe we can be of some help.

Good luck,
Chromavore

[AnalyticalWisco]

I'm joking, and I don't think it's ridiculous to think that you could recreate a method from a published journal or that the process of performing experiments, trying to interpret the data and then asking for help when it doesn't make sense is unscientific but you have somewhat fallen backwards into the trap that is hplc method transfer/development.

I was relieved to see this response. The tone of the previous responses was hard to stomach.

I'm interested to hear if @PeterSDSU made any progress on this. My first impression of the situation is that he may have chosen a tough starting point for this project. Running a gradient with Acetate buffer with detection at 215nm is going to be really noisy (acetate is also not really the best buffer choice for pH 3.5). The following application may be a better starting point.

"https://www.agilent.com/cs/library/applications/(FD9)5988-6365EN.pdf"