Exploring the Relationship between Concentration & Refractive Index

[Serenie]

< Mentor Note -- two threads merged and new user reminded not to multiple-post threads at the PF >

i'm experimenting about refractive index and i can't figure out the relationship between concentration of the solution and Refractive index.

What i found about this is that RI is connected to density, but the relation is not linear.

Then how does HPLC measure concentration of the solution based on RI?

 

[Serenie]

i'm a student in high school and I'm having a problem figure out the relationship between RI and concentration of the solution.
What i found suggest that there is no linear relation. However, HPLC does allow us to know the concentration of the solution based on RI.
My question here is that how does HPLC know the concentration when measured RI?
Does it just experimentally measured and put into the computer? or does it is induced by theoretically induced?

 

[sophiecentaur]

My question here is that how does HPLC know the concentration when measured RI?

I have never used HPLC but do you not need to know the substance that you are analysing before you can tell the concentration from its RI?

 

[Serenie]

I have never used HPLC but do you not need to know the substance that you are analysing before you can tell the concentration from its RI?

We are using NaCl solution.

We are just using another way to measure RI. During our research, we knew that HPLC can measure the concentration of the solution.
But what we found via wiki told me that there is no linear relations between concentration of the solution and RI.
So we are wondering in what way the HPLC know the concentration

 

[sophiecentaur]

We are using NaCl solution.

We are just using another way to measure RI. During our research, we knew that HPLC can measure the concentration of the solution.
But what we found via wiki told me that there is no linear relations between concentration of the solution and RI.
So we are wondering in what way the HPLC know the concentration

You know the substance and I presume that you have to declare it to the program. The computer will have data about all common substances. It's a monotonic relationship so it will just use a look up table to give concentration from RI. I imagine if you created some exotic substance you could input data, based on some calibrating measurements you would have to make.
Have you done any research into the HPLC method?

 

[Serenie]

Oh, so is there a table about the RI and concentration?
i didn't know that.

i think we didn't do research about the HPLC method.
But i think, for measuring RI, it will measure like spectrophotometer.
if not i want to know how i can research the HPLC method

 

[sophiecentaur]

Oh, so is there a table about the RI and concentration?

From what wiki has to say, it seems that they use HPLC to split the mixture up and then (sometimes) use the RI measurement to find the concentration with a look up table. Storing and retrieving Data is so easy these days and there is no point in trying to find an algorithm to 'calculate' it.
I know nothing about it but I know to Google "HPLC" and to put in terms like "method" or "using HPLC".
If you are a high school student I doubt whether you really need to concern yourself too much with something this specialised for Exam purposes. HPLC is very much an Industrial process and you are unlikely to have the equipment in your Lab (?).

 

[Nik_2213]

I used a bunch of HPLC detector types, though never RI. Still, before each run, we'd do a reproducibility check in the expected concentration range, do wary stats on those 5~10 injections. Also, each sample or short group of samples would be bracketed by multiple standards, again in the expected range. Several times a year, and after service or repairs, we'd run a 'staircase' of standards to check linearity across a wider range. We'd also check for 'carry-over' using 'wash-vials'.

When dealing with 'swabs', 'residues' and other wild-cards, we'd run a 'staircase', plus additional standards or sample dilutions to confirm we were working well within the detector's linear response range. If working near the lower end of sensitivity, often with 'dirty' samples, we'd use 'wash-vials' to catch slow-eluting 'Tail End Charlies', plus several 'wash-vials' on the end, to be sure, to be sure...

One 'gotcha' doing eg product transfer from research lab to production support is research teams get all the best toys, but may forget the rest of us don't.

Once upon a time, we received a *fully registered* test procedure for a new, capsuled med. Rather than cut open each capsule and wash the contents into a flask for known dilution then vialling, they simply dropped a capsule into each flask, agitated to dissolve...
"But won't the synthetic gelatin gum the HPLC column ?"
"We had no problems. Just follow the procedure."
So we did.
We were supposed to run 3~4 batches of 10 capsules plus standards per auto-sampler tray. Night after night, the HPLC column gummed up, the pump went over-pressure and halted after a dozen samples. Expensive column ruined, pump & sampler strained, leaks everywhere. Total nightmare.
We queried the procedure, was told they had no problems.
We ruined several more expensive columns, were being 'leaned on' because testing had been delayed.
Again, we queried the procedure, was told they had no problems.
"But what about the gum ? We've ruined $10k of custom columns because they gum up after a dozen capsules. How do you get around that ??"
"Uh, we back-flush after each sample; don't you ??"
{{ Face Palm... }}
"No, we don't. First, we're not set up for back-flushing. Second, and most important, THAT IS NOT IN THE REGISTERED PROCEDURE YOU SENT !"
"Oops..."

 

[Ygggdrasil]

Usually you would calibrate your measurements to a set of standards with known concentration. Even if the relationship is not linear, you can still interpolate to estimate the concentration of your unknown from the standard curve.

 

[Nik_2213]

"HPLC is very much an Industrial process and you are unlikely to have the equipment in your Lab"

Um, we used 'industrial strength' isocratic & gradient HPLC at 150~~200 Bar, with chunky s/steel valves, 10-32 micro-bore fittings and 1.5~2.0 ml/min flows of toxic methanol/water or acetonitrile/water blends through s/steel columns at 40~50'C. But, I saw ads for 'Bio' tech that ran at much, much lower pressures. Really, this was an automated or semi-automated version of 'gravity eluted' chromatography. Transparent columns, plastic pump and valves, benign solvents, faster flows. Truly, 'Visible Science'. Sounds like fun !

 

[DrDu]

i'm a student in high school and I'm having a problem figure out the relationship between RI and concentration of the solution.
What i found suggest that there is no linear relation. However, HPLC does allow us to know the concentration of the solution based on RI.
My question here is that how does HPLC know the concentration when measured RI?
Does it just experimentally measured and put into the computer? or does it is induced by theoretically induced?

So what exactly are you trying toeasure and with which equipment, detectors etc?

 

[Kevin McHugh]

RI detectors suck for HPLC work. They must be thermally isolated (column, tubing, solvent, etc.) in order to obtain decent results. RI is very sensitive to temp changes. Drifting baseline is a problem with temp fluctuations. The most widely used application for HPLC with RI is analysis of polyolefins with GPC columns (varying pore size in tandem). Toluene or dichlorobenzene the mobile phase. Everything is in thermal equilibrium at 130C. RI is linear in concentration in this application..