[tetrapyrrolic_nick]

I'm trying to distinguish two porphyrin isomers which each have two carboxylic acid groups on them (no other polar groups) using HPLC. I can't get anything to come through the column! I'm new at HPLC so I need help. I am using acetonitrile and 10mM phosphate buffer at pH 7. I can't get anything to come through the column when I run it in 100% of either solvent or a 50/50 mix. It is a 250 x 4.6 mm, 5 micron particle-size C18 column. I think it might be sticking to the column and not moving but I can't understand why it would do that. Any ideas? Should it move faster in the buffer solution or acetonitrile? Sometimes I got a big peak when I changed to solvent from acetonitrile to buffer and I can't understand that either. Please help me!

[Arkcon]

The porphyrins are very large nonpolar groups, so you may have bound them too tightly to elute.  You could try THF, or maybe a less non-polar column, for example a C-8 column.  A phenyl- based column has somewhat more specific binding for aromatic rings, so you might get some different specificity.

The buffer will also effect the separation.  If the buffer's pH is below the pKa of the analyte, you will protonate the analyte, making the molecule even more non-polar.  However, if you deprotonate with higher pH, you encounter mixed-mode separation, which will widen your peaks and make them tail badly.  The charged analyte may interact with the silica beneath the C-18 and likewise bind too strongly.  You can avoid this with a polymeric reverse phase column -- a styrene-divinyl benzene column can be pretty inexpensive, and can handle harsher reagents.  You could also try ion exchange for your charged molecules.

So many options, but I feel like I haven't actually solved your problem.   Do you know the pKa of your porphyrins?  There are tools that calculate the (approximate) pKa based on structure.  What buffers have you tried?  What pH was your buffer?  Remember, more solvent drives the pH further toward neutral -- if pH can even be considered valid for non-aqueous media.

[tetrapyrrolic_nick]

Thanks for the reply and for the advice  Yeah I think they might just be binding too tightly but I just tried to elute a porphyrin tetra acid and it won't come through either! The acid groups are on benzene rings and should therefore have a pKa somewhere around 4. Since my buffer is at pH 7 they should be pretty much fully ionised. I'm starting to suspect that my detector might not be working properly.

Do you think that THF would be more likely to pull it through the column than acetonitrile? I know it seems like basic information but I have to ask again because I can't seem to find the answer elsewhere; will the buffer or the acetonitrile (or THF) tend to pull the porphyrin through the column more quickly?

[DrCMS]

If you ran the just 100% aq buffer solution through your column with no organic solvent you have probably screwed the column. 

Try reading up on HPLC principles before you do too much more damage.

[tetrapyrrolic_nick]

If only I could learn everything simply by reading! Why does it ruin the column if you put 100% aqueous buffer through it?

The porphyrins are more soluble in THF than in buffer or acetonitrile, is it a good idea to try eluting them with THF?

[Arkcon]

*Sigh*  Once again, DrCMS: notices what I tend to miss ... you will ruin the average C-18 column if you run 100% aqueous through it, and you've implied that you have.  This is probably the most common reason why people lose all function of a C-18 column.  The water ends up ruining the wetting ability of the C18 media, and recovery is very time consuming.  I never mention it, because I was sure everyone already knew this, and surely an academic adviser, or the equipment manager tells people this.  And some modern columns can handle 100% aqueous better. 

But yes, you will need a new C18 column.  And check its functionality (the previous user may have damaged it too) -- run 50-50 methanol-buffer and inject a tiny amount of caffeine (look up exact concentration, this is a common diagnostic for HPLC function, it will also check your detector.)  You should get a sharp peak.  Then try your porphyrin sample at 50-50, and see what you get.  If you need better separation, go stepwise with more aqueous or more solvent, 75-25 or 80-20.  If you feel you need to use 90% solvent, consider that a C-18 is too strong a binding column for your analyte.  If you feel you have to use 90% aqueous, realize that that is putting wear and tear on the column media, and you will lose reproducibility as you do more runs.

If you're getting really poor peak shape, that is, it tales so broadly that you can't even detect the peak, you may have gain some improvement by adding an ion-pairing reagent.  That that complicates things for later analysis, if you're collecting peaks for NMR or mass spec.

[DrCMS]

Arkcon I only pick it up because ~15 years ago I moved to a new job in a small company who did not have a HPLC.  I convinced the owner/boss that it would be useful to buy one so he gave me the budget to buy one but he told me I'd be running it.  Before that in university and other jobs there had been an analytical group that I just handed the sample to and got the results later.  Now I had to learn all the fine details of actually running an HPLC fast.  I spent time working out what not to do so I'd not break the new very expensive machine and screw the colums etc.  Simple things like using MeOH/water flushes for the columns and auto-injectors, filtering all the solvents and degassing them, having pre-column guard filters etc. etc.  Now when I see someone make those simple mistakes I nearly did it stands out to me more than you.  Thankfully these days in a different job and company I have a very good analytical chemist who runs the QC lab and sorts out any new analysis I need.

[tetrapyrrolic_nick]

I've solved the problem now. It turns out that if I dissolve my porphyrin samples in buffer solution instead of in acetonitrile they come through! I used gradient elution using 20mM ammonium acetate at pH 7 and acetonitrile from 60% aqueous to 20% aqueous over the course of 18 minutes. The dewetting was no-doubt also a problem but running acetonitrile through it for a while seemed to fix it just fine. I also experimentally determined that they come through more quickly in higher concentrations of acetonitrile (at least up to 80%). Thanks for the help. I'm about to post another very similar question about porphyrin quarternary ammonium salts 

[tetrapyrrolic_nick]

After wrestling with porphyrin carboxylic acids that wouldn't elute I eventually discovered that if I injected the samples dissolved in aqueous buffer rather than acetonitrile that I could get them to come through. I suspect that in acetonitrile the porphyrins form aggregates which then stick in my guard cartridge.
Now I am trying to analyse porphyrin quarternary ammonium salts and I can't get them to come through the column! (not even one with four QAS groups). I have tried using buffers of pH 7 and 3 even though I didn't really expect that to make a difference (and it didn't). I have been injecting them dissolved in buffer solution. Today I will try injecting them in acetonitrile but that is just a stab in the dark. Can anyone offer me any ideas as to how I can get them to elute? Has anyone had trouble with quarternary ammonium salts before?
Thanks in advance.