I have a problem with a method development. The drug to determine is rapamycin, macrocyclic lactone (mol wt 914,2). The blood concentrations providing efficacy are 5-20 ng/ml.
I have got Waters PDA 996 detector with baseline noise 0,00005 AU.
Rapamycin and I.S.(32-desmethoxyrapamycin: DMR) exhibit two peaks in the chromatogram, becouse in solution these compounds exist as an equilibrium mixture of inter-converting isomers A,B,C. The major component is isomer B. So in the chromatogram is the first peak rapamycin B, the second RAP C (3 - 10% of RAP B account), the third DMR B (higher) and the fourth DMR C (smaller).
I have tested two RP C18 columns and I started with the mobile phase consisted of ACN and water.

1. Column: Symmetry 250x4,6 mm, 5 um particle diameter
Mobile phase: 80% ACN, 20% water
Flow: 0.4 ml/min
Column temp.: 60°C
Inj. vol.: 50 ul

Capacity factor for RAP B: 2.4
EP Resolution: RAP C/RAP B 1.71, DMR B/Rap C 1.76,DMR C/DMR B 2.22
USP Tailing Factor: RAP B 1.05, RAP C 1.08, DMR B 1.08, DMR C 0.98
Peak width (50%) at 484 ng/ml: 0,45 for RAP B, 0,35 RAP C, 0,43 DMR B, 0,42 DMR C

2. Column: LUNA Phenomenex 150x4,6 mm, 3 um
Mobile phase: 73%ACN, 27% water
Flow: 0,5 ml/min
Col. temp.: 55°C
Inj. vol.: 50 ul

Capacity factor for RAP B: 5.7
EP resolution: RAP C/RAP B 1.8, DMR B/RAP C 1.75, DMR C/DMR B 2.4
USP Tailing Factor: RAP B 1.06, RAP C 1.08, DMR B 1.08, DMR C 0.98
Peak width (50%) at 484 ng/ml: RAP B 0.47, RAP C 0.38, DMR B 0.42, DMR C 0.40

When I decrease concentration of rapamycin in the sample, the lowest concentration at which I can distinguish the rapamycin spectrum is 24.2 ng/ml (0,00015 AU - 100ul inj. vol.) on both columns. But I need LOD at least 5 ng/ml, better 2.5 ng/ml. What can I do for it? To change column diameter? Try to take another column? How should I continue?

Many thanks for your answers!

Jitka

Assuming that you need to assay the drug in plasma (or whole blood?), you should need to perform a sample treatment (SPE, liq-liq extr., etc. depending of your particular drugs) in order to isolate and concentrate your analytes.
If you started with 1 ml of plasma by liq-liq with an org. solvent that you can evaporate to dryness, you could resuspend the drug in 100-200 ul of a suitable solvent for injection of 50-100 ul. In that way you get a 10x concentration factor (assuming recovery from plasma 100%)that means, 2 ng/ml in plasma would be 20 ng/ml in the final extract. Try to use a solvent containing less organic modifier (ACN)than the mobile phase to resuspend the dry extract. This improve peak shape when higher aliquots are injected.
Regarding chromatogr. conditions, I prefer 100-125 x 2.1 mm ID columns, 3 um particle diam., 0.25-0.3 ml/min.
Have you tried high temperatures (80°C)? What's the wavelenght you've selected?

Yes, sample treatment is necessary.
You also should check your HPLC system for some optimization. What kind of flow cell is in your 996? If you have the analytical flow cell with path length 10mm and cell volume 8µl there is also a microbore flow cell with path length 3mm and cell volume 2,6µl available. This small cell volume you will need when using small ID columns.

For sensitivity enhancement you already did the right thing. You went from 5µm to 3µm particles using a shorter column. But still there is something you can do for optimization. Here are some "golden" rules for optimization:

decreasing column length will give you reduction in analysis time and solvent consumption

decreasing column diameter and length will give you increase in mass sensitivity and reduced solvent consumption

increasing ratio of column length to particle size will give you increase in resolution

increasing column diameter will give you increase in loadability

So, my recommendation is to go with
1. small flow cell volume
2. small ID of your SS capillaries
3. shortest way from column out to detector in
4. smaller column ID (2mm or 1mm ID)
5. smaller particle size (2µm particles are available on the market)
6. go with a 2mmID x 10cm column packed with 2µm ODS particles

Good luck.
Gerhard

Dear MFB and Gerhard!

Thank you very much!

So I start to concentrate my samples (whole blood) by liq.-liq. extraction, that´s true, it will be helpful. What a simple idea! I expect, that endogenous components from blood will complicate me next testing...

I found out that higher temperatures up to 55°C or 60°C hadn´t improved resolution, but height of peaks. As high temperatures as 80°C I haven´t still examined.

Maximum wavelength of rapamycin is in Merck index 277 nm. Somebody measures at 276 or 278 nm. I tried all - it´s the same.

I have a standard flow cell in my 996. You write, Gerhard, that I will need a microbore flow cell for small ID columns - it means also 3,9 mm and 3,0 mm or smaller than 2 mm?

I am afraid of extracolumn effect ( I have no experience, I´ve only read about it), when I use a microbore or narrowbore columns. I know - I must shorten tubing, decrease flow cell size...Say me, please: is my fear reasonable?
When I use smaller column size, I suppose I must decrease injection volume - so is it so helpful becouse of increasing sensitivity to decrease column ID?

Jitka