Allowance for undissolved solids

[AdrianF]

If say 1g of inactive solid is in a 100ml volumetric flask the active is only dissolving in 99ml.

does anyone out there make a correction? e.g when analysing tablets with insoluble exipients

[zokitano]

Very interesting question!

If the procedure for e.g assaying tablets says "put each tablet in to a 100 mL volumetric flask, add diluent, mix (or shake or sonicate) and then fill up to the mark with for e.g diluent". As Adrian pointed out you are practically dissolving the component of interest in to a volume less than 100 mL (the volume of 100mL is a sum of the volume due to the diluent added and the volume of the undesolved tablet constituents), but in the calculations, in the method, 100mL is used as final volume of dilution (the actual volume available for dissolution of components is less than 100mL).

Thus one makes calculations with concentration of the component of interest which is less than the actual concentration in the 100 mL flask, right?

And also this method produces a systematic error "as is". Instead of the declared concentration of the component (e.g. 0.05 mg/mL) one has slightly higher actual concentration in the final solution (>0.05 mg/mL).

[Consumer Products Guy]

I would accurately add 100 ml solvent, and mix with that.

[andy]

I would try a recovery experiment to see if the undissolved solids are causing a problem. If the recoveries are good, I would just dissolve the tablet in the 100 mL volumetric flask.

[juddc]

Interesting question, though if you're terribly concerned with this why not simply measure everything by mass and be done with it? You can accurately determine the difference in diluent added this way and see if the difference is going to impact your calculations in any meaningful way.

Also, in your example you're only diluting in 99 ml if the specific gravity of your solid is 1.00 and it's completely insoluble.

[Dan]

I was involved with a method for soft gelatin capsules where the soft gelatin didn't disolve in the solvent. The capsules were cut and placed into the volumetric flask, shells and all. We determined that the soft-gelatin shells had a volume of about a 2-3 mL. So, instead of 100 mL we only had 97-98 mL of solution. We used a correction factor.

Over the years, I have seen many methods that used a volumetric flask in the sample preparation without consideration that there is undissolved material in the flask which results in volume errors for which no correction is ever made. There is also a problem in that a volumetric flask is not very suitable for the sample preparation when the sample is tablets or capsules (try it with a cream or toothpaste product; what fun!). Also, there are a host of extraction issues to deal with.

The conclusion/suggestion: CPG already said it. Use a different container (erlenmeyer flask, bottle, vial, etc.) and add in the desired amount of solvent. It's easier, faster and can improve accuracy and precision.

Regards,
Dan

[juddc]

Creams and toothpastes are fairly easy to prepare in volumetrics: Use a positive displacement device such as a 1 ml disposable syringe and make sure the tip is clean before dropping it into the flask.

Sometimes the fun is getting undissolved (frequently polymeric) materials out of the flasks...

[danko]

Hi Adrian

I would suggest the following exercise: Fill a 100 mL volumetric flask up to the mark with the relevant solvent and then add 10 tablets (or whatever number you choose in order to minimize the uncertainty).
Then mix well (according to your procedure) and remove the redundant liquid so that the mark and the liquid surface are aligned in the same manner as prior to the tablets addition.

The volume of the removed liquid (divided by 10 or the number of tablet added) will indicate how large is the assumed error. Then you can correct the results accordingly.

Because I’m not sure that you can assume that 1g of the matrix will cause 1 mL volume increase. It depends both on the density of the matrix and on the amount/volume of the liquid that penetrates the unsolved particles.

Best Regards

[zokitano]

Interesting question, though if you're terribly concerned with this why not simply measure everything by mass and be done with it? You can accurately determine the difference in diluent added this way and see if the difference is going to impact your calculations in any meaningful way.

I agree with Juddc. Simple expression of the concentration of the analyte as molality instead of molarity will minimize the effect of the undissolved tablet constituents. That can be done by simple measuring of the solvent mass, used for dissolution of the sample (tablets).
And also, if one decides to do it this way, one should express the concentration of the standard and sample solution as molality.

[Bruce Hamilton]

Your standards are supposed to match your samples, so add the excipents to them and make to volume or use weight - whichever you like.

I used to analyse Vitamin E in soft gelatine capsules, and we used special widemouthed, calibrated volumetric flasks for ease of cleaning, and tossed similar number of blank gelatine capsules into the standards.

During method validation, we added blank gelatine capsules to 1 standard and also analysed the equivalent standard without the capsules. We corrected for the capsule volume as we confirmed that the presence of the capsule didn't affect the analysis, but then no longer worried about capsule volume correction.

Please keep having fun,

Bruce Hamilton

[HW Mueller]

If I remember this correctly then molality is moles per 1000g solvent, not per 1000mL.
The correct way to do this would be to extract, filter the extractant, place the filtrate and washings in the volumetric, fill to the mark.
(As a student in Denver it was questioned that I would ever be a chemist after I perpretated a much lesser sacrilege than having solids in a volumetric flask.)

[AdrianF]

Thanks for your comments. The method I am referring to is a method accepted by the regulatory authorities - not my idea.

I have tried the experiment suggested by Danko and discovered that the final volume was over 101ml under assay conditions so the error is 1%+ ;

Most tablets are analysed in this way. In this case the tablet has a low content of active so a large proportion of excipient so the problem is more evident.

Mr Mueller - it is impractical to filter the extract. In many cases it is difficult to filter and there are losses on filtration.

The way round it is either to apply a correction or as in the USP for this tablet add a known volume of mobile phase to the sample in a centrifuge tube - extract by ultrasonication, centrifuge and use supernatant for analysis.

This problem has crossed my mind before but I have never taken the time to think it through or question the status quo. Now the fun begins!

[zokitano]

If I remember this correctly then molality is moles per 1000g solvent, not per 1000mL.

That's exactly what I meant when I said: that can be done by simple measuring of the solvent mass, used for dissolution of the sample (tablets)

Best regards

[HW Mueller]

AdrianF, certainly one can do both a filtration as also a centrifugation wrongly, I just think that washing a filter can be more efficient than washing a precipitate.
Zokitano, some suggested adding 1000mL solvent. That would give uniits which are not defined in the literature, as far as I know.

[AdrianF]

It is not practical to use molality:

1. because of the use of mixed solvents eg water/mehanol- you would need to make up the solvent exactly the same for each sample.

2. The problems with subdilution

3. Yhe problem of weighing flasks with large volumes of liquids.

As proof of the difficulty I doubt that anyone out there uses molality in routine analysis.