Department of Energy Argonne National Laboratory Office of Science NEWTON's Homepage NEWTON's Homepage
NEWTON, Ask A Scientist!
NEWTON Home Page NEWTON Teachers Visit Our Archives Ask A Question How To Ask A Question Question of the Week Our Expert Scientists Volunteer at NEWTON! Frequently Asked Questions Referencing NEWTON About NEWTON About Ask A Scientist Education At Argonne Salt Crystals
Name: Melissa
Status: other
Grade: other
Location: PA
Country: USA
Date: Fall 2010

We made crystals from non iodized salt and iodized salt to compare differences for my sons science project due soon. The iodized salt made perfect cubes the size of a pencil eraser, while the non iodized made cubes much smaller and "snowflake" looking crystals. My question is why the difference if the iodine is an impurity that should drop to the bottom of the solution? Should they not be even due to the impurities separating?


Iodized salt will contain a small proportion of sodium iodate or sodium iodide, along with the normal sodium chloride. In any case, the iodine (in whatever form) will be soluble in the solution so it will not be an "impurity that should drop to the bottom of the solution". It does not contain any elemental iodine, although this could form in tiny amounts by the oxidation of the iodide ion, or reduction of the iodate.

As you evaporate your solution to form crystals the sodium chloride even though it has a higher solubility) will certainly reach its solubility limit first, as it is such a large proportion of the total solute, so the crystals will be sodium chloride for both the normal and iodized salt.

The amount of iodide or iodate involved is so small that I find it hard to believe that it will affect the crystal formation of the sodium chloride. But crystal formation is famously fickle, so I could be wrong. However, it is much more likely that other factors are influencing the crystal formation pattern. I would suggest that you repeat the study, paying particular attention to controlling all the variables that could affect crystal formation: initial concentration of salt solution, volume of solution, surface area, conditions and time of evaporation etc etc.

If you can repeat these results under controlled conditions, they would be an interesting result! It is possible that the iodide or iodate ion is influencing the crystal structure (it only takes very small amounts of impurities to change pure corundum to sapphire or ruby, for example...)

Best wishes,

Tom Collins

Melissa -

I understand your confusion; the thinking is correct as far as it goes. Impurities should mess up a crystal if they are included (not make it grow bigger/neater/faster), and do nothing if they aren't included.

A small point: impurities not included in the growing crystal may either precipitate elsewhere (drop to the bottom), or remain dissolved. In principle, remaining dissolved is what happens first, and later if the concentration builds up precipitation may happen ( and may interfere with the crystal too).

There is an in-between behavior called "habit-modifier". Some substances that do not fit into the crystal nonetheless hang around temporarily on the growing surface and are cloying enough to have effects, such as:

- changing the relative growth rates in various crystal directions

- changing the crystal shape by changing the balance between growth-rates of the directions of facets vs. edges vs. points.

- suppressing the nucleations that mess up the crystal,

- assisting the kinds of nucleations that must occur regularly for the crystal to keep growing fast (such as starting a new row along a growth-step on a surface, or starting a new layer on a finished layer)

Both surfactants such as soaps and partly-interchangeable ions (such as iodide vs. chloride in this case) can have such effects.

Of course, to be fair, you should make sure that your salt-without-iodide has high purity, i.e., is equivalent to the salt-with-iodide minus the iodine. It would have to come from a chemical company with specs. Adding your own pinch of iodide to this pure sodium chloride would make the fairest experiment.

Any consumer table-salt will often have sodium silicates in it, to cover each grain with a hard, water-resistant thin skin, to make it less easily glued to its neighbors by humidity. This silicate stuff definitely settles to the bottom in a crystal growth experiment, if you're lucky. Otherwise, it may mess up and force small messy crystals such as the snowflakes you mentioned. If you dissolve some salt in more water than it needs, and the water is still faintly cloudy instead of perfectly clear, that's probably the silicates. Silicates can be dissolved in basic solutions, so if you add a pellet of sodium hydroxide (such as drain-cleaning lye) and the cloudiness disappears, I would take that as a sign that the cloudiness was in fact silicates. Not sure I would actually grow crystals with that batch, but you could try it and see.

To make sure there are no silicates in your salt, it needs to be bought from a chemical company (or pharmacy drug counter), or else you need to crystallize it once and wash the crystals with rubbing alcohol, before dissolving it again to grow the crystals you really intend. It is work and time, but crystal substances are distinctly cleaner and grow different crystals the second and third time around.

Jim Swenson

I would have expected this experiment to work more or less equally well with both kinds of salt, so I am really not sure. Generally the things that determines the size and quality of the crystals grown are concentration, temperature, cooling rate, and the availability of nucleation centers. These include scratches on the inside of the flask, dissolved impurities in the water, and even vorticies created by stirring in highly supersaturated solutions.

Usually to get big crystals you want slow cooling and heterogeneous nucleation, i.e., scratches in the flask, a piece of string, etc. For sure, if the crystals form too suddenly from a supersaturated solution they will be small and flaky.

The iodine in iodized salt is not in the form of pure I2, so no I2 goes to the bottom of the beaker as an impurity. Different manufacturers use different sources of iodine but in the US only sodium iodide and cuprous iodide are allowed. Sodium iodide is by far the most common. All four of these are pretty soluble and will be in small enough concentration that I would be surprised if they interfered with the experiment. They should instead stay in the liquid above the crystals (chemists call this the "supernatant liquid"). According to the American Salt Institute's web site, US manufacturers use 0.006% to 0.01% KI by weight in iodized salts. Also, either of your two salts may also have a small amount of fillers and/or anti-caking agents like dextrose and other compounds.

In the end, I recommend approaching this with the scientific method. Try different salts from different manufacturers. Also, you might try switching flasks and trying again with the same two salts.

Best, Dr. Topper


Iodine is added to salt only as an extra added health benefit. The human thyroid gland needs a supply of iodine to properly function, So salt manufacturers "iodize" their salt to provide this benefit.

Here is an online article that tells you all you would ever want to know about Iodine:

Refer to section 9 of this article for the biological role of iodine and the thyroid gland.

So you ran a crystal experiment and one ended up with cubes and the other ended up in a snow flake texture. I do not think the cause is the presence of iodine. Neither do I believe that the iodine should drop to the bottom of the solution because there just is not that much iodine added to the salt. (Iodine solubility in water is discussed in section 8.1 of the Wiki article)

One way to test for the presence of iodine is to rub the substance under test on a potato and if that place on the potato turns dark purple then you have confirmed the presence of iodine. (See the "Analytical chemistry and bioanalysis" section of the wiki article. Sec 8.6) You probably have a lot of impurities in your salt samples as well.

I would run the experiment again and be extra sure that the parameters are all the same.

Good luck.

Sincere regards,

Mike Stewart

NEWTON is an electronic community for Science, Math, and Computer Science K-12 Educators, sponsored and operated by Argonne National Laboratory's Educational Programs, Andrew Skipor, Ph.D., Head of Educational Programs.

For assistance with NEWTON contact a System Operator (, or at Argonne's Educational Programs

Educational Programs
Building 360
9700 S. Cass Ave.
Argonne, Illinois
60439-4845, USA
Update: June 2012
Weclome To Newton

Argonne National Laboratory