April 2012

Salt tenderizes.
Salt preserves.
Salt enhances flavor.

And too much salt can ruin a dish.

Its easy to over salt. Sometimes you forget the ham, just tossed into the pea soup, was cured in salt as well as sugar. Or, you confuse "Tbs" with "tsps". Or, fail to remember that boiling down a sauce to concentrate the flavors will also concentrate the salt, rivaling the Dead Sea in salinity.

But never fear, help is waiting in the pantry. A perennial kitchen "hint" suggests that by simply adding a raw potato to the broth, it will act like a sponge, sucking up the extra salt and saving the day.

But don't put away your take-out menus quite yet-- unfortunately, this hint is a miss.

Why? Well, first of all the potato can remain in the hot broth or stew for only so long before it begins to disintegrate and ruin the dish. So it may have only thirty minutes to work its magic. And, in a thick broth, the liquid hardly convects, so the potato "sponge" only absorbs salt from within its immediate vicinity. Plus, if the potato absorbs salt AND flavor AND water in the same proportions as the stew, its no different than ladling out a bit of liquid. To work as promised, the potato must suck in salt but not flavor.

Like in the myth, a potato will absorb salt until the salt level inside the starchy tuber matches the salt level in the soup. Remove the potato, and you remove some of the salt. If you want to remove half the salt, you'd need to drop in potatoes equal in volume to the entire batch of stew! But the biggest problem is diffusion, or lack thereof. Qualitatively, a potato will capture salt. But, quantitatively, the amount removed is a drop in the ocean.

To see why, I performed the following experiment: This pot contained 16 cups of water and enough salt to bring the concentration up to 2%. Which is on the high side for consumption- most soups are closer to 0.5%, and the ocean is at 3.5%. Two percent of 16 cups of water is 2.5 oz, or about 15 tsps of table salt.

I then peeled and quartered two large Idaho potatoes, added to the light brine, and simmered for about a half hour until just tender. Often, the potato sponge myths suggests simply dumping in just half a spud- so this is aggressive remediation.

After removing the potatoes and cooling under a fan (to stop any of the absorbed salt ions from diffusing further), it was time to figure out just how far the salt had absorbed into the potato. I used two methods, both of which agreed quite closely.

First, I simply cut out small plugs from surface and the interior, and tasted. Inside the potato, no salt flavor at all. In the top few mms, plenty salty. Try this yourself at home.

I also employed a visual salt indicator on a slice of cooled potato (the indicators are normally used in pools and aquariums, but are also a great technique for kitchen science- see here for more details). The indicator turns white at high salt concentrations, and salmon colored with no salt. As you can seen, the eye and tongue are in close agreement:

Knowing the dimension of the two potatoes and the depth of the salt diffusion¹, its easy to calculate how much salt was removed. The salinity was reduced by about 4 parts in 100, or about a ½ tsp out of 15. At best.

Pretty unimpressive. Which begs the question, are there simple ways to remove salt from an over salted broth?

You might increase the potato's surface area by dicing into cubes, and wrapping in cheesecloth to make it easier to remove. Or use rice instead of potatoes². Which will help, but some of the flavor is also absorbed, tossing the baby out with the bath water.

Salt is easily, and strongly, dissolved in water. To preferentially remove excess salt requires a chemical or mechanical technique which more powerfully attracts salt than the OH ions in the water. And is non-toxic, foolproof and inexpensive. Which eliminates most electrochemical and chemical techniques.

But there are two methods that work.

The first is best for broths- solvent induced precipitation. Counter-intuitively, you continue to evaporate and concentrate the broth, raising the salinity. Then add in pure ethanol (vodka), which forces salt crystals to precipitate and fall to the bottom. Filter out the crystals, flame off the alcohol, and rehydrate. More science than cooking, but it does work.

The second techniques applies to either broths or thick stews, and harkens back to the early days of dialysis. Place the salty brew in a "semipermeable membrane". This is a thin film riddled with tiny pores, sized to easily pass salt ions, yet small enough to block much larger flavor molecules. Pour the salty stew in a bag made from the semi-permeable membrane and drop into a large bath of plain water. Over time, the salt will randomly diffuse out bag into the water, diluting the salt but not removing any flavor. The diffusion continues until the salt concentration is equalized between the bag and the bath.

I used sheep intestine³ (e.g. sausage casings) as the bag, and measured the salt reduction in two ways. At room temperature, I used a floating hygrometer to measure the increasing salt concentration in the water bath. At 160F I periodically measured the salt concentration with a reflectometer- both devices purchased at the local salt water aquarium store.

This photo shows a bag of Heinz gravy in a bag, to which I added 4.5 extra tsps of salt (about 20 gms). The hygrometer is the glass tube bobbing in the center. A day later, the water was very slightly dyed with color from the gravy and the hydrometer has risen (due to the higher density of salt water):

I then repeated this experiment placing the container in a water bath at 160F. As expected, the diffusion rate increased with increasing temperature:

(A bit of math is needed to convert from specific gravity to tsps salt removed. The gravy itself was salty, and that salt will also diffuse out of the membrane. About 1000 ccs of water, 250 cc of gravy, 20 gms added salt. Its important to stir the bath frequently. I tasted the water bath at the end of the experiment -after boiling for safety- and it was very salty, while the gravy was perfectly seasoned).

So this is actually a practical technique- in a few hours, you could easily halve the amount of added salt without changing the original flavor profile. If you were really ambitious, you could build a kitchen dialysis machine- running a tube of sheep casing though the stew and flushing water through the casing, continuously sweeping out salt.

Or you could just start over.

--------------------------------------------------------------------------------------------------------

¹ In additional (and tedious) experiments, I confirmed that instead of preferentially absorbing more than its "fair share" of salt, the potato simply comes into equilibrium with the brine salt concentration. So the geometric volume of salted potato limits the total salt removed, as shown above.

² Rice is also ineffective. For example, I added 200 gms of white rice to 1000 gms of salt water (6%). Along with a bit of red food coloring, as a surrogate for flavor molecules. And cooked until the rice was soft. The rice absorbed around 400 gms of water and was red to the center. Which means removing the rice removed flavor from the broth- you might as well have poured out 400 gms of broth to begin with. More importantly, the salt level in the remaining broth was unchanged, as measured with a refractometer. Plus the rice starch caused the broth to thicken and turn cloudy. Altogether a waste of time.

If you repeat this experiment (salt water aquarium enthusiasts or home brewers may own a refractometer), note the starch will interfere with the refractometer's operation. So you have to deactivate the starch. The easiest method is to burn the starchy, salty broth to a crisp, then rehydrate the char with the same amount of water. Filter out the blackened bits and measure the salt level with the refractometer.

³ While sausage casings are often available in a kitchen, they usually contain small tears which leak. Not critical in sausage making, but disastrous in dialysis. So I turned to a high quality, guaranteed pinhole free casing- natural lambskin condoms. These condoms are made from a lamb intestine cul-de-sac called the caecum, rather like the appendix. While impermeable to large flavor molecules (and ginormous sperm), smaller HIV viruses can make it through the caecum. So natural condoms are suitable only as a contraceptive, but not to prevent disease.

Salt ions are 3000 times smaller than HIV, so its possible a latex condom could block the virus but let ions pass freely. However, in addition to the off-flavor of latex, these condoms are nearly pore-free. In fact, when I tested a latex condom filled with salt brine, not only was the rate of ion exchange much slower than with lamb intestine, the ions stopped diffusing after 2 days, long before reaching equilibrium. As salt is used to coagulate liquid latex into a solid, I assume the charged ions are electrostatically trapped in the latex, plugging the pores. Good news for its intended use, though....