# How to measure the percent weight of NaCl in a potato chip

I want to determine the weight percentage of sodium chloride in a potato chip. I can see from this answer that atomic absorption spectroscopy (AAS) can be used to quantify sodium chloride in a solution. I propose the following measurement scheme:

1. Dissolve the potato chip in a sample of water with known mass
2. Homogenize the solution until it has a uniform composition
3. Draw 10 aliquots from this solution
4. Use AAS to measure the weight percentage of salt in each aliquot
5. Use the mean and standard deviation of the weight percentages from the 10 aliquots to generate an estimate of the true weight percentage with a confidence interval

My questions:

1) What is the best way to get the mass of the aliquot that belongs to the potato chip? I want to minimize measurement uncertainty.

2) There could be compounds other than sodium chloride that contains sodium or chloride in a potato chip. I fear that the AAS won't detect sodium chloride, but sodium and chloride ions separately. What is the best way to measure the concentration or mass of sodium chloride with AAS? If AAS is not the best way, can you name an alternative method?

I'm somewhat skeptical that atomic absorption spectroscopy (AAS) would work as you outlined the procedure. The "solution" would have starches and fats floating around in it.

(1) I'm not sure that you could get a homogeneous solution.

As far as the homogeneous solution, think of some parts of the solution settling. So what happens if the solutions sit for an hour before going into the AAS?

You could test this by homogenizing a sample and analyzing it immediately, then analyze it after an hour. Do you get the same result?

(2) Even if you did I'm not sure that all the gunk would burn correctly (uniformly) in the AA flame.

Using a plasma spectrometer would solve this problem.

(3) Depending on what sort of accuracy you want there is also the problem of moisture in the chip. How "dry" is a potato chip?

(4) There is a difference between salt added to the chip and $[\ce{Na^+}]$ and $[\ce{Cl-}]$ in the chip. I'd guess that $[\ce{Na^+}] \approx [\ce{Cl-}]$ in a biological system, but I'm sure that it isn't exactly $[\ce{Na^+}] = [\ce{Cl-}]$.

Such are the joys of method development in analytical chemistry.

Your statement about the linked answer, and AAS for this application in general, is not quite right; you are not measuring "sodium chloride", but sodium.

I would modify the preparatory part of your procedure to do a clean water extraction of the potato chip, which mainly amounts to adding a filtration step after homogenization of the chip/water mixture. This will leave you with nearly a one-phase solution of water, with some oil floating on top. As sodium chloride is far more soluble in water than in water-immiscible oils, you will want to sample only the water. This is much simpler than analyzing a 3-phase (aqueous, oil, solid) system and also adds some very small degree of selectivity (for water soluble sodium only, like sodium chloride).

Unfortunately, there is no good way to measure only the sodium that came from sodium chloride via AAS; you will simply be measuring all soluble sodium, which is the vast majority of sodium compounds. The assumption I would make in the case of a potato chip is that the vast majority of the sodium comes from the sodium chloride added, usually in good excess, for flavoring.

Regarding your question about the best way to get the mass of the aliquot that belongs to the chip: just assume that you've extracted all of the salt from the chip into the water. In this case, say you used 200 mL for the extraction and then took 10 mL aliquots for analysis; this means each aliquot represents 10/200 of a chip.

If you want to use AAS you will need to make calibration functions. Instead, you can do better and faster. I would recommend using Na-selective reagents like Crown esters for example and do the complexometric titration.

• I fear that Na-selective reagents would only help to quantify sodium, not sodium chloride. Other sodium-containing compounds could exist in the potato chip, so measuring the amount of sodium via Na-selective reagents may overestimate the quantity of sodium chloride. Am I correct in this assessment? If not, please tell me why. – A Scientist Oct 15 '15 at 16:47
• I believe you're correct. I'd look into running cations and anions by HPLC to determine the ratio of sodium to chloride, which will tell you which one is in excess - if the slope of the plot is far from unity, that will confirm other sources of the ions in question. – Todd Minehardt Oct 15 '15 at 17:18