Refractive Index Detectors - Response Factor Consistency

[adam]

A question for the experts. Do refractive index detectors give response factors that are relatively consistent on a mass basis (like ELDS or CAD).

Any comments or suggestions are appreciated.

Thank You

[unmgvar]

RI is a concentration detector,
it means the signal response, area and height changes in direct relation to the compound concentration, as long as you are within the linear range of the compound behavior and instrument optics.

so if you have a standard compound of known weight you can create a fit curve of conc over signal and calculate the concentration of unknown samples.

ELSD is not a real straight forward concentration detector, but the software can do the calculation for you seamlessly

[mardexis]

Hi Adam,

No they do not. There is a large variability in response which is compound dependant. The response is proportional to the difference in refractive index (RI) between the observed species and the mobile phase (MP). The MP is typically water/ACN or MeOH with an RI around 1.34
so a species like bromonapthalene (RI=1.658) would give nearly ten times the response of a compound like formic acid (RI=1.37).

Detection limits are similar or a bit worse than ELSD but with much better linear dynamic range. The biggest hit is not being able to run gradient analysis by RI.

[adam]

While we are talking about this, let me ask another question.

I know that gradients don't work well with refractive index, but can you use them for scouting purposes (if you can tolerate a steeply sloping baseline). Or maybe even for analysis purposes if you are only doing the analysis once or twice and don't care about a perfectly optimized method.

[tom jupille]

(if you can tolerate a steeply sloping baseline)

It's not going to work; "steeply sloping" in this case means more than 1,000 times full scale per minute.

Assume one microgram of analyte, a peak width of 15 seconds at a flow rate of 2 mL/min. That means that your analyte is eluting in 0.5 mL of mobile phase. The apex of the peak is at (very) roughly double the average concentration, so the highest concentration of analyte is 2 x 1 microgram / 0.5 mL = 4 micrograms/ml

Assume that your gradient is 1%/min, that means that the solvent concentration is changing at a rate of 0.01g/mL/every minute (again, a *very* rough approximation), which equals 10,000 micrograms/mL every minute.

Now, assume that the response of your analyte and the strong component of the mobile phase are roughly equal. If your analyte peak is full-scale, then your drift in one minute will be 2,500 times full scale.

[AA]

In my experience, ELSD does not give relatively consistent response factors on a mass basis. There is a hugh compound dependance factor there.

This comment "ELSD is not a real straight forward concentration detector, but the software can do the calculation for you seamlessly" is of interest, what software are you talking about here?

[HW Mueller]

I have trouble with the phrase "consistent response" as used here. Could it be that you are talking about specific and/or nonspecific response? I think of a consistent response as one which doesnt change while doing the experiment/work, one that gives a linear response to boot, or, as an example, a response from a compound which is stable and follows the Beer-Lambert law.

[ivanvins]

[quote="adam"]I know that gradients don't work well with refractive index, but can you use them for scouting purposes (if you can tolerate a steeply sloping baseline).[/quote]

In fact, even mixing isocratic mobile phase by a gradient pump is not advisable unless you are interested in seing the elent composition fluctuations. Go for ELSD if you need to use gradient