USP on Mobile Phase

[shaun78]

The USP suggests that, in order to make their assays work, one is allowed to adjust the mobile phase by 2% absolute or 30% relative, which ever is larger. However, the change in any component can not exceed 10% absolute, nor can the final concentration of any component be reduced to zero.

I just don't get it.

I have two bottles: a). water; b). acn.
Initial conditions: a). 82%; b). 18%
gradient over 20 min to: a). 40%; b). 60%

I want to increase the amount of water initially .... Would that mean I can only adjust the composition of, say my starting conditions, by 2%; or can I get away with 17% (can't reduce b to zero); or can I only run with 10 absolute variation%.

To me, I would be working in absolute, so I would think I can only change by 2% .... but we all know that really buys one very little if anything at all.

Like I said, I just don't get their terminology here.

[Mark Tracy]

absolute: |%ACN(original) - %ACN(new)|
In your case 2% absolute means ACN(new) = 16% or
10% absolute means ACN(new) = 8%

relative: |%ACN(original) - %ACN(new)|/%ACN(original)
In your case 30% relative means ACN(new) = 12.6%

Clearly 30% relative is a greater change than 2% absolute, but less than 10% absolute. So you may change to as little as 12.6% ACN.

As a contrasting example suppose that your initial composition were 5% ACN. 2% absolute would mean ACN(New) = 3%, 10% absolute would mean ACN(New) = -5%, and 30% relative would mean ACN(New) = 3.5%. In this situation the greater change would be 2% absolute, and you would not hit the zero rule. So you could adjust to ACN(new) = 3%.

The third example would be 90% ACN initial. 2% absolute means 88%, 10% absolute means 81% and 30% relative means 54%. While 30% relative is greater than 2% absolute, it hits the limit of 10% absolute, so the maximum change would be to ACN(new) = 81%.

[shaun78]

Mark,

Thank you for your exceptionally informative reply. I now think I understand what they are getting at with this.

Thank you!

Shaun

[Mark Tracy]

I forgot one detail: these rules apply to every component of the mobile phase, so you need to do the math for both water and acetonitrile to determine the limits.

[shaun78]

will do.

Thanks again!

Shaun

[zokitano]

The USP suggests that, in order to make their assays work, one is allowed to adjust the mobile phase by 2% absolute or 30% relative, which ever is larger. However, the change in any component can not exceed 10% absolute, nor can the final concentration of any component be reduced to zero.

Shaun78 has asked about mobile phase that contain two solvents (binary mixture).
What's happens when one deals with three solvents, for instance:
10% THF: 30%ACN:60% Buffer X?
Does it mean that you're allowed to do the same changes (in the specified limits given in Pharmacopoeia) for every solvent?
Or what about when one uses 4 different solvents?

Has anyone reasonable answer about this question?
I just wanna be sure about this.

Thanks

[tom jupille]

First of all, the suggested default adjustment limits appeared in Phamacopeial Forum in 2004. As far as I know, they have not yet appeared in the USP as such.

The FDA has similar limits for their use: http://www.fda.gov/ora/science_ref/lm/v ... _04_05.pdf

This does cover ternary but not quaternary blends.

[zokitano]

Thank you Tom for the reference. It clarifies some of the doubts that I had.

Best regards

[shaun78]

The USP brought it back to the PF around the time that I posted this question. Once again, it got shot down.

Because it did not become official, I had to quickly write a validation protocol and validate a method because I changed column length (150 mm to 250 mm) and mobile phase comp by 2%.

Incredible waste of time in my humble opinion.

[tom jupille]

Incredible waste of time in my humble opinion.

Yeah, but otherwise we'd be unemployed!

Seriously, if the situation is likely to recur, what I would do is to write an SOP essentially duplicating the FDA document (and referencing it for justification).