Economist Writing Every Day

Someone forwarded me this article by Superfoodly, Is Liquid Smoke Flavor Safe or Cancer in a Bottle? This article seems to have useful health information. I will unpack the physical basis for this below, but the key takeaway is:  smoked foods (i.e., have been exposed to actual smoke) like smoked turkey, and especially fatty meat/fish like salmon, have appreciably more carcinogens than food flavored by “liquid smoke” type flavorings.

Let’s look at the chemicals involved here. I won’t tax readers with molecular formulas, just the names of some key players. Polycyclic aromatic hydrocarbons (PAHs) are a broad grouping of large molecules, consisting mainly of so-called aromatic (benzene-type) rings that are fused together. Some classes of PAHs are known to be particularly potent carcinogens. For instance, the PAH known as benzo(a)pyrene is found in smoked meat and it gets the absolute most hazardous carcinogen rating by the World Health Organization’s IARC. A Group 1 rating (“Carcinogenetic to Humans”, not merely “Probably Carcinogenic”) for benzo(a)pyrene puts this substance in an elite group including hepatitis C, mustard gas, formaldehyde, and neutron radiation. Lovely.

Most of us savor the taste of “smoke” in at least some of our foods. This love affair with fire and smoke may hearken back to prehistoric times, when fire was one of the few advantages that we soft, weak, slow humans had over other species. We will pay a premium for authentically smoked turkey, ham, herring, salmon, etc.  Why would it be better to concentrate smoke flavor in some liquid form, and then apply this “artificial” elixir to our foods?

A 1993 study by Michigan State University found that genuinely smoked (or even grilled) meats and fish had benzo(a)pyrene equivalents one or two orders of magnitude higher than expected from foods flavored with liquid smoke. (I won’t reproduce the table in the Superfoodly article, since I am not sure of the meaning of the units, and the original 1993 article does not seem available on line).

More recent studies, comparing smoked foods to foods artificially flavored with liquid smoke, also show more carcinogens in the former, but differences tend to be more like a factor of two:

Source: Superfoodly

So, the verdict is in. It is well known that cooking meats at high temperatures even without smoke exposure generates carcinogens, but smoking them makes it worse. The article concludes:

“If you insist on eating smoked meats and fish, whether that be salmon, prime rib, applewood bacon, or summer sausage, your best bet will be to buy a bottle of Stubbs, Wright’s, or Colgin liquid smoke and use that, instead of making them the old-fashioned way over smoldering wood or charcoal.

The research suggests preparing them the fake way will be much safer than the real way.”

As a chemical engineer, I wondered what it is about concentrated liquid smoke that would make safer than plain smoked meat. After all smoke is smoke, isn’t it? Well, not exactly – – smoke is a complex mixture of chemicals, not a single compound, and it turns out that the more dangerous chemicals like PAHs preferentially absorb in fatty materials such as meats and fish like salmon and herring.

Liquid smoke, on the other hand, is made by running smoke through water to condense the gases and particulates which comprise the smoke. Most of the PAHs will end up in the tar or waxy fractions, and only the water-soluble compounds end up in the liquid smoke flavorings. Now, these water-soluble chemicals include some funky stuff that gives me pause, but net net there will be less of the known carcinogens in the water than in the smoke as a whole.