49

I read many recipes of pretzels and they required to dip the raw dough in a bath of lye. As anyone should know for their own safety, lye is caustic and shouldn't be ingested.

What is the process involved that make them edible?

Edit: I am aware of the action of the lye. I wonder how the non-edible lye on the dough is transformed into something that is safe to eat.

A.D.
  • 1,081
  • 2
  • 10
  • 19
  • 26
    The difference between medicine and poison is the dosage. Neither Sodium (present in salt) or OH- (present in water) are inherently toxic. So lye in low concentration causes no harm. This question seems to assume any quantity of lye is hazardous. – Mindwin Remember Monica Mar 28 '17 at 17:43
  • 2
    [Related question on Chemistry.SE](http://chemistry.stackexchange.com/a/43913/38761) – Mindwin Remember Monica Mar 29 '17 at 13:51
  • 1
    @J... Part of the lutefisk preparation from link you give is "To make the fish edible, a final treatment of yet another four to six days of soaking in cold water (also changed daily) is needed". Pretzels don't need to be washed for days before eating. So while the fundamental answer is the same ("lye is used in preparation, but you don't eat significant amounts"), the processes aren't that similar. – armb Mar 31 '17 at 13:36
  • 1
    YouTube: [Using Potassium Hydroxide as a Condiment](https://www.youtube.com/watch?v=Nj46HrNmy2w). I actually use potassium hydroxide (KOH) for my pretzels because I have a bottle of it for hydroponics, but not NaOH. – Nick T Apr 02 '17 at 13:42
  • 1
    We natively have a safe dose of a potent corrosive (hydrochloric acid) in our stomachs... – rackandboneman May 10 '18 at 22:46

5 Answers5

53

Basically, the lye reacts with the CO₂ and moisture present during baking to form a non-toxic carbonate. This makes it safe to eat.

The reaction:

CO₂ (g) + H₂O (l) ­⇄ H₂CO₃ (aq)

H₂CO₃ (aq) + 2 NaOH (aq) → Na₂CO₃ (aq) + 2 H₂O (l)

From here (MS doc)

[EDIT]

Spurred by the comments, I have searched further.

tl;dr There is much going on wrt the lye dip. As far as safety goes, the lye is consumed in many reactions, including the above.

  • (Firstly: The equation source was not the basis of my answer; rather it was to refresh my memory of the reaction about which I was told/read several years ago was the reason why lye is safe to use on leavened breads, which was its combining with carbonic acid. (I apologize for not checking the balance adequately.)
  • My recent search only found one reference at The Kitchn to the reaction of lye with carbonic acid as the reason for its safe use. It is also unsourced.
  • Simultaneously, I found a research paper and a Food Chem Blog entry which referenced it, both of which discussed the behavior of the lye bath on pretzels. There is a lot there, so I shall only quote the paper abstract:

The effects of alkali dipping on starch, protein, and color changes in hard pretzel products have never been researched. Experiments were conducted to mimic reactions occurring on the pretzel dough surface. Dough was dipped in water or 1% sodium hydroxide solution at different temperatures between 50°C and 80°C. Protein and starch profile after dipping were analyzed. Color development on pretzel surface following the extraction of pigments from flour was investigated. Whole dough and pretzel samples were also made at pilot plant and the properties were analyzed. Only starch granules on the dough surface were gelatinized following dipping. Amylose-lipid complex dissociated at a lower temperature with alkali treatment but were not dissociated, even at high-temperature dipping in water. Treating the dough at 80°C in alkali solution resulted in the hydrolysis of proteins into smaller peptides that could be not precipitated by trichloroacetic acid (TCA). Dough surface color was different following pigment extraction from flour but not significantly different following baking. The results suggest that the color that developed on pretzel surface was not due to pigments present in the flour but was contributed by the reaction within or between the starch and protein hydrolysis derivatives during baking.

and what I think is the pertinent quote from the blog:

The protein results (2 in the list above [reproduced following]) indicate that the lye dip provides the smaller proteins needed for Maillard reactions, whereas the water dip does not. This seemed like perhaps the most important point to me.

  1. The dip resulted in the hydrolysis of protein into smaller peptides. This happened a little bit in 25°C water or lye dip, more in 80°C water, and a lot more in 80°C lye dip. Also, the smaller peptides in the hot lye dip had the smallest molecular weights; most of them “walked off” the electrophoresis gel, leaving no bands. The authors explain that the alkaline conditions of the lye dip result in like charges along the proteins, which repel and cause the proteins to unfold; this makes them more susceptible to hydrolysis.

Both the blog and the paper are worth reading.

My conclusion: the lye is consumed by the various reactions and therefore poses no safety concerns.

wumpus D'00m
  • 3,991
  • 14
  • 24
  • 15
    I don't think this reaction is correct or relevant. Your quoting from an experiment that puts a solution of sodium hydroxide in a PET bottle filled with pure carbon dioxide (CO2), which is nothing like the environment pretzels are made in. The final equation also looks wrong as its not balanced. It was probably meant to be NaOH + H2CO3 -> NaHCO3 + H2O. That product, sodium bicarbonate (NaHCO3) wouldn't be desirable on pretzels, as it without an acid present it would breakdown to sodium carbonate when baked (2 NaHCO3 -> Na2CO3 + H2O + CO2), and give the pretzels a bitter soapy taste. – Ross Ridge Mar 28 '17 at 23:18
  • 5
    @RossRidge I'll go a little farther. The claimed reaction is just wrong precisely because it doesn't balance: one atom of hydrogen just disappears. Chemistry department fail. On the other hand, there's not much lye so it's plausible that the relatively high temperature and the fairly CO2-rich atmosphere from the fermenting yeast is enough to make the reactions you describe happen. Note that lye is already turning the fats in the dough literally to soap, so maybe a soapy taste isn't significant. – David Richerby Mar 29 '17 at 08:44
  • 1
    However, note that the first reaction is given as producing carbonic acid in the aqueous phase (i.e., dissolved in water). Carbonic acid essentially only exists in the aqueous phase -- according to Wikipedia, NASA managed to make some solid H2CO3 in 1993 but, until then, it had never been seen before. If you don't have liquid water, you don't have an aqueous phase, so I'm sceptical that this reaction is what's happening between steam and CO2. – David Richerby Mar 29 '17 at 08:48
  • 1
    @RossRidge You're correct about the unbalanced H. I have corrected the equation. I also changed steam to moisture as the former is a bit sloppy. Thanks. – wumpus D'00m Mar 29 '17 at 12:07
  • 4
    Balancing the equations is good, but there's still the bit where this doesn't seem a likely reaction - I know there's a bit of water on the surface of the pretzels initially, but they dry pretty fast, so there's really not a lot of chance for the first reaction to take place. Do you have a source specifically saying this happens in pretzels, not just the more artificial conditions described in your link? – Cascabel Mar 29 '17 at 15:03
  • It’s still unbalanced. Presumably, it should read Na₂CO₃. – Emil Jeřábek Mar 29 '17 at 15:31
  • @EmilJeřábek Rebalanced. I forgot the extra Na in the previous edit. – wumpus D'00m Mar 29 '17 at 16:24
  • 3
    And another problem. Even if this reaction is possible, what would be the point? You're proposing that the lye gets neutralized by being converted to sodium carbonate. OK, fine. Why do you want sodium carbonate on your pretzels? Because, the way you write it, the only purpose of the sodium carbonate is a neutralization product of the lye. But if that's the only reason, there's a much easier way to deal with the lye -- just don't dip the pretzels in it in the first place! – David Richerby Mar 29 '17 at 23:03
  • 2
    It's not the most solid, but On Food and Cooking does have this: "The starch gel hardens to a shiny finish and thanks to the alkaline conditions created by the lye, browning-reaction pigments and flavor compounds rapidly accumulate. (The lye reacts with carbon dioxide in the oven to form a harmless edible carbonate.)" @DavidRicherby Well, conceivably the effects on the proteins are the desired effect and carbonate is formed as a byproduct. – Cascabel Mar 30 '17 at 15:00
  • @Jefromi So I guess the implicit claim there is that the lye (and, probably more specifically, the hydroxide ions) catalyze some sort of reaction since, otherwise, they'd be consumed by that reaction, rather than forming a carbonate. – David Richerby Mar 30 '17 at 15:04
  • @DavidRicherby Yes, the point of the lye is to affect the dough, I believe by hydrolysis of the proteins, not sure what effect it has on starch. There are other answers here suggesting that the lye is at least partially consumed by those reactions (if it's only partial, turning the rest into carbonate would still be useful), but it looks like protein hydrolysis is indeed catalyzed by alkalinity. – Cascabel Mar 30 '17 at 15:27
  • I don't think it's strictly accurate to say that the lye is "consumed", in the sense that _all of it_ participates in various chemical reactions. For that to happen, the quantity of lye relative to the quantity of dough would need to be very precisely controlled. – aroth Mar 31 '17 at 12:18
  • @aroth Well, it is pretty well controlled: you use a lye bath that's not insanely strong, and then you remove the pretzels from it, so the lye is only what can stay on the surface of the pretzel. From all the answers here, It's not entirely clear what happens to that lye, but pretzels don't end up with significant lye, so it's being consumed/converted by some reactions one way or another. – Cascabel Mar 31 '17 at 16:25
38

The purpose of dipping in lye (or other basic solution, like baking soda...or even baked, baking soda) is that it promotes coloring, as the solution reacts with the surface of the dough. It also promotes the Maillard reactions when the dough cooks. The result is even browning and that typical alkali flavor. If choosing lye, food grade is important, as commercial grades may include other, heavy metal, impurities. Lye is extremely caustic. So it must be used carefully! In pretzel and bagel making the solution is generally quite dilute...maybe around 3% lye in water. In both pretzel and bagel making, the product is typically given a brief bath in boiling water, after a dip in the lye solution. The boil and/or subsequent baking neutralized the alkali rendering it safe to eat.

moscafj
  • 72,382
  • 3
  • 117
  • 207
  • 2
    I am aware of what you're describing. I found many questions on the subject here. My question was on the process that transform the non-edible lye on the dough in something that is safe to eat. – A.D. Mar 28 '17 at 12:08
  • The bath and baking make it safe. Unless that's not what your asking. – coteyr Mar 28 '17 at 14:49
  • 14
    @A.D. Caustic compounds are edible when diluted sufficiently. Now I haven't done the math to see if 3% lye is already OK, and there is certainly also a reaction between the lye and the dough, but if you were not aware that dilution matters (and you seem not to) then this answer adds an important detail. – rumtscho Mar 28 '17 at 16:55
  • 2
    I think this answer when combined with @wumpusD00m 's answer above would be the most complete. This answer tells why the lye process is important, and that food-grade must specifically be used (in a very dilute solution, too), while wumpus' answer goes into more detail on how the lye gets neutralized in the process. – Doktor J Mar 28 '17 at 18:02
  • 4
    This doesn't answer the question that was asked. – Ryan Cavanaugh Mar 28 '17 at 18:23
  • 3
    Boiling doesn't neutralize alkalis. – David Richerby Mar 28 '17 at 21:14
  • @davidricherby I think it washes the lye off a bit or something. It really depends on how deep the lye from the lye bath penetrates the dough. – htmlcoderexe Mar 28 '17 at 22:49
  • 2
    I think this is the substantially correct answer. During baking the lye absorbed by the pretzel dough is consumed in the reactions that produce the browning. Any other reactions to neutralize the lye would be unnecessary. – Ross Ridge Mar 29 '17 at 00:12
  • 5
    Here in Germany (pretzel country) the pretzels are never boiled or rinsed after their stint in the lye. – Stephie Mar 29 '17 at 07:41
  • 3
    Seems likely that most people think of caustic compounds in light of what they've seen in cinematic effects. I.e. as if every strong acid or base were either some kind of [universal solvent](https://en.wikipedia.org/wiki/Alkahest) or a catalyst. E.g. the Alien movies. – can-ned_food Mar 29 '17 at 11:20
  • 3
    If you happened to stick your finger in an unmarked container of food grade lye, and place in on your tongue in an effort to identify it, you will get one hell of a burn. It's use requires caution and attention. – moscafj Mar 29 '17 at 14:53
  • @rumtscho - if I've done the maths correctly, 3% NaOH solution is equivalent to a molar concentration of about 14mM, which according to [this calculator](http://www.endmemo.com/chem/phcal.php) results in a pH of 12.15, which I think is quite likely to cause problems if ingested. I can't find figures for safety in humans, but [this study on use in animal feed](https://www.efsa.europa.eu/en/efsajournal/pub/2882) suggests it is a skin irritant at concentrations > 0.5%, so that is the highest concentration that could realistically be safe to ingest. – Jules Mar 30 '17 at 01:19
  • 1
    @Jules OK, then in this case, the dilution is not sufficient. Thank you for producing the numbers! I find it good to keep in mind that there *is* a safe dilution with caustic substances, and that's why I upvoted this answer along with the "reacts with amino acids" one. – rumtscho Mar 30 '17 at 13:52
  • @Jules 3% NaOH is 30 grams per liter, molecular weight is 40 grams per mole, so 0.75 M or pH 13.9 – DavePhD Mar 31 '17 at 16:29
31

The reason it is safe is three-fold.

First, the concentration is only 1% NaOH and the pretzels are only dipped for 10 seconds (see Snack Food Technology pages 180-182) which limits the amount of hydroxide per pretzel.

Second, the dough itself, for example protein of the dough, has acidic groups, such as amino acid side chains of lysine and tyrosine, which neutralize the hydroxide.

Finally, as explained in Effect of Alkali Dipping on Dough and Final Product Quality Journal of Food Science vol. 71, pages C209-C215, protein in the dough is partially hydrolyzed under the alkaline conditions. This exposes more terminal amino acid groups which also participate in neutralization.

The Snack Food Technology book cited above also explains:

If the caustic concentration becomes too high, there is not a complete conversion to sodium bicarbonate in the baking and drying cycles and the pretzels will be hot to the taste due to the residual sodium hydroxide

DavePhD
  • 906
  • 8
  • 8
  • Actual science at last! Thank you! (And the word "carbonate" doesn't occur anywhere in the linked paper.) – David Richerby Mar 30 '17 at 15:09
  • Though this answer may not have the votes because it lacks an unbalanced equation, I think it should be noted as an excellent and concise answer. – QueueHammer Mar 30 '17 at 21:19
  • 1
    @QueueHammer Late answers always suffer. The accepted answer already had 30+ votes by the time this one was posted and it will get more votes in the future, simply by being the first answer that people see. – David Richerby Mar 31 '17 at 07:54
  • Huh, sodium *bi*carbonate? The reaction in the top answer is definitely sodium carbonate, not bicarbonate. Do you happen to know what the actual reaction is here? – Cascabel Mar 31 '17 at 15:55
  • @Jefromi yes, I'm trying to add more, having trouble formatting the chemical equations. Usually I'm on chemistrySE. – DavePhD Mar 31 '17 at 15:58
  • @Jefromi is that clear now? – DavePhD Mar 31 '17 at 16:05
  • I think I'm missing something - I saw "conversion *to* sodium bicarbonate...residual sodium hydroxide" and thought that meant the lye was being converted to sodium bicarbonate, but your reactions and explanation show conversion of sodium bicarbonate to sodium carbonate, and treat the one getting rid of lye as undesirable? – Cascabel Mar 31 '17 at 16:15
  • @Jefromi If all the bicarbonate is consumed neutralizing the lye, then there will be no bicarbonate left for leavening purpose. I think that is the book's point. – DavePhD Mar 31 '17 at 16:18
  • What you're saying could make sense, but it doesn't seem consistent with what the book says. It's talking about there being undesirable residual lye, not about a lack of remaining bicarbonate, and it definitely says "conversion *to* sodium bicarbonate" and your explanation involves converting bicarbonate into other things, not converting anything *to* bicarbonate. – Cascabel Mar 31 '17 at 16:21
  • @Jefromi I think you have a good point, I'll just leave it at what the reference itself says. – DavePhD Mar 31 '17 at 16:36
  • Fair enough! I'm still curious about what I originally asked (it sounds like the reaction might not be what's in wumpus D'00m's answer) but the reference is a step at least. – Cascabel Mar 31 '17 at 16:37
  • 1
    @Jefromi supposing it is converted to bicarbonate initially, it would still eventually be converted to carbonate upon baking due to thermal decomposition. This 1948 source says the lye is changed to carbonate https://books.google.com/books?id=saPQAAAAMAAJ&q=%22lye+was+changed+to+sodium+carbonate%22&dq=%22lye+was+changed+to+sodium+carbonate%22&hl=en&sa=X&ved=0ahUKEwip6_bZmYHTAhVHeCYKHcuZChEQ6AEIGjAA – DavePhD Mar 31 '17 at 16:55
  • 1
    @Jefromi Principles of Cereal Science and Technology says "bicarbonate" and that it forms due to reaction with CO2 from air. https://books.google.com/books?id=s8hWAAAAYAAJ&q=%22sodium+hydroxide+reacts+with+the+carbon+dioxide+in+the+air+to+form+sodium+%22&dq=%22sodium+hydroxide+reacts+with+the+carbon+dioxide+in+the+air+to+form+sodium+%22&hl=en&sa=X&ved=0ahUKEwjcocSfnYHTAhVF4CYKHbjZB2IQ6AEIGjAA – DavePhD Mar 31 '17 at 17:07
  • 1
    Typical Lye used for German "Laugengebäck" ("Lye-Bakes") such as Pretzels is around 3%. Your answer is still correct, but the original concentration is three-times higher than stated in your answer. – user2705196 Apr 01 '17 at 16:16
  • @user2705196 The "Snack Food Technology" book gives examples of 0.5% and 1.25%, so it seems the concentration varies between 0.5% and 3% depending upon who is doing it. – DavePhD Apr 01 '17 at 23:20
18

Lye will readily react with either amino-acids (producing respective sodium salts) or with fats (producing soaps), both reactants being readily present in dough. You don't need CO2 to neutralize it.

Ingesting small amounts of those end-products is indeed safe, and normally only a small quantity of lye is used in the process.

Dmitry Grigoryev
  • 531
  • 4
  • 10
  • +1. Consider what NaCl + H_2O forms. You can, in fact, consume pretty *high* concentrations of lye as long as the result isn't caustic when it touches us. Nobody whines that there is lye in Gatorade. – The Nate Apr 01 '17 at 16:44
  • @TheNate I don't understand what you mean by "consume pretty high concentrations of lye as long as the result isn't caustic when it touches us." Lye in high concentrations (actually, even in fairly low concentrations) is caustic; consuming requires touching. So how can I consume pretty high concentrations of lye without touching anything caustic? – David Richerby Apr 02 '17 at 14:54
  • @TheNate And Gatorade doesn't contain lye. At least, it's not on the [ingredients list](http://www.gatorade.co.uk/products) (go to the "nutrition" tab; no direct link available). – David Richerby Apr 02 '17 at 14:59
  • NaOH is lye. They would have added salt, which then disassociates in the water. NaCl + H_2O dissociates into NaOH and HCl. Hence, it's both present and not on the ingredients list in *lots* of places. If you're concerned about the lye, just make sure to eat it with mustard so the vitriol can neutralize it before it goes inside you. – The Nate Apr 02 '17 at 22:35
  • @TheNate NaOH + HCl is not the same as NaOH alone. Nothing will happen if you put your finger in table salt, but in lye your skin will melt pretty fast. Of course, Na is not toxic, so you can say "It's OK", but you'll still lose your finger. – Dmitry Grigoryev Apr 03 '17 at 07:37
  • Go ahead and explain to the folks at home what happens to your finger if you stick it in glacial acetic acid, Dmitry. (in contrast to eating mustard.) I know *you* understand my point, so why are you reinforcing the misconception that the question is based on? You know the one: conflation of the dangers of lye in a pure solution vs. a highly diluted and buffered form. – The Nate Apr 10 '17 at 17:15
  • @TheNate I'm not trying to reinforce the misconception, and it's true that a highly diluted NaOH solution is harmless. But your example is still flawed: table salt is safe at much higher concentrations than lye, because it will not affect pH in any way. Not sure what you mean by *glacial acetic acid*. – Dmitry Grigoryev Apr 11 '17 at 10:07
  • Glacial acetic acid is pure (ish) acetic acid. I know pure is the high end, but I can't recall how low the percentages have to get to disqualify it from "glacial". – The Nate May 08 '17 at 17:47
1

The references noted above have mostly looked at the specific chemical changes to the constituents of the dough and the species in solution. A few point to the Maillard reactions as a contributor to what is going on.

It is worth noting that the Maillard reactions are quite complex and involve a lot of intermediate products. However, in many cases the rate limiting factor is the pH of the constituents. It is possible to accelerate the reactions by raising the pH, and more Maillard products are produced if you let the process run for an extended period of time or raise the temperature which further increases the reaction rate. Some people don't believe that you can get the reaction to go at all at temperatures below 300°F, yet adding a little baking soda to a batch of onion soup and pressure cooking it (265°F) for 40 minutes will yield the same browning that much longer cooking of the onions produces in the classic technique.

So increasing the pH by using lye (pH 13) vs sodium carbonate (pH 10) vs sodium bicarbonate (pH 8) will facilitate a dramatic speed up the Maillard reaction rate, and subjecting the pretzel to high temperature in the oven will carry it out. What happens to the NaOH to detoxify it is most likely a combination of neutralization, dilution, and chemical conversion through interaction with other available species. I would not recommend eating dough dipped in lye without baking the dough first.

I am intrigued by the notion that the high pH breaks down the proteins into shorter amino acid sequences which facilitates the Maillard reactions but does not affect the rate constants.

Doc Dough
  • 31
  • 2