scaling up from analytical to preparative HPLC

[vyncyi]

Dear all,
i'm very new to HPLC, my project require me to scale up to preparative to obtain enough samples to proceed on to the next step. i've done some literature search and come to this conclusion, i can convert the flow rate and maximum loading in a straight forward method if i'm running an isocratic system. however, i'm very confused about the gradient part...can someone please enlighten me on how to convert it?
thank you very much

sincerely,
vyn

[tom jupille]

Sample load scales the same way as it does in isocratic chromatography: in direct proportion to the column volume. In order to keep "equivalent" separation chemistry, you need to keep "k*" (average k') constant. k* is proportional to:

(tG x F) / Vm

where:
tG is the gradient time
F is the flow rate
Vm is the column volume

(this assumes that you are keeping the gradient range constant and that you are running a linear gradient). You can juggle those in any combination; so long as everything cancels out, you will have equivalent separation.

[vyncyi]

dear Tom,
thank you for your help. i've tried it out but i still have the same prob, i.e the peaks are very tiny and broad...almost can't be be identified and they don't come out at the same retention time (not even close at all).
is there any thing else that i should have take into consideration?
thank you

[tom jupille]

Tiny peaks usually are not the problem in prep LC!

How about more details about what you are doing:
- sample mass(es)?
- column dimensions and proposed flow rates in the analytical and prep systems?
- gradient conditions (initial and final %B, time)
- type of separation (reversed-phase, normal-phase, etc)
- what is your sample dissolved in?
- detector and cell configuration?
Any (or all) of those factors could have an effect.

[vyncyi]

initially when i was developing the method using analytical, i injected 60ul (1mg/ml) of sample (dissolved in methanol) using a gradient of 95%water to 100% acetonitrile in 30minutes. the flowrate was 1ml/minute, column is RP-C18e (diameter: 4.6mm, length:100mm), PDA detector... but i'm not sure about the cell configuration.
when i started using the prep column (diameter: 25mm, length: 100mm), i increased the sample injection to 1.8ml (1mg/ml), flow rate was 29.5ml/minute for 30minutes with the same gradient condition and time.
when i got very small and broad peaks, i thought maybe i didn't inject enough sample, so i increased it to 1ml(50mg/ml), but i still can hardly detect the peaks. then i tried lowering the flowrate to 20ml/minute instead but still with the same results.