volumetric vs. gravimetric dilution

[LALman]

...Once we got 4-place electronic analytical balance (I pushed for this, we had the first in the whole company) we were able to weigh much quicker, so just used a dropper into a volumetric flask. Sometimes, we would actually add some solvent to the volumetric flask, tare that, then add some sample...

I misunderstood what the OP was desiring to do until I saw this post.

I do this to dilute my Internal Standard Surrogate mix. The mix comes in methanol in a 1mL vial but I never get 1mL out. I have a spreadsheet set up. I weigh and then tare a clean 40mL VOA vial. Then I add about half of the mass I expect I will need of P&T methanol as a "keeper' and record that weight. Then I tare again and crack the vial and add the contents and get that weight into the spreadsheet. I tare again and add methanol untill I reach the weight my spreadsheet tells me, having kept track and calculated the amount needed to get the target ppm. I like to dilute my 2000 ppm ISS stock to 70 ppm with this method. 5uL in my 5 mL gastight syringe is 70 ppb at the instrument. Since I split at 40:1, that would be 8.75 ng for my BFB.

[Peter Apps]

Thanks to all who replied to this. In my case, I am not so much worried about the increased precision of gravimetric vs. volumetric - I would prefer to do gravimetric dilutions because it uses less solvent and there is no glassware cleaning necessary.

James and Peter discussed the area I am concerned with - whether or not gravimetric dilutions can introduce error based on solvent density. For example, there should be less mass of analyte per unit volume in (1mg analyte / g acetone) solution than a (1mg analyte / mL acetone) solution because acetone's density is less than 1g/mL. Assuming that the same volume of standard and sample are injected, does volume 'fall out' of the calculation?

Yes. Try a dummy calcualtion and see.

Peter