Baking Soda in Cookies and Spread - a Note on Biscottini
Posted: 12 December 2010 01:06 AM   [ Ignore ]
Total Posts:  3
Joined  2010-12-11

I am the son of a retail baker who retired in 1965. I believe that it was in his shop, so this is an old memory, that Dad told me that one needs to be very careful about the amount of baking soda in a cookie formula because it is a determining factor in the spread of a cookie.  I believe that at the time he was talking about ‘lunch box’ cookies, i.e., oatmeal, chocolate chip, ginger molasses, etc.

I tried a Google search to confirm this, but it is not easy to construct one that does not deliver many false positives or no answers at all. Has anyone seen this phenomenon explained by gurus like Harold McGee or Shirley Corriher?

I was recently baking biscottini and confirmed the phenomenon to my satisfaction. When I was done, I was mildly put off by the residual taste from the soda, (1/4 tsp, rounded, per 3/4 cup of flour.) It led me to ask the question, are there alternative ingredients that would promote spread. I use one basic dough recipe and vary the flavoring ingredients. I’ve always noticed that if I use melted chocolate the loaves spread more. Perhaps it is the fat content in the chocolate?

Does the soda function as a leavening agent here? Would ammonium bicarbonate work as well? It should leave no residue as it disproportionates into ammonia and carbon dioxide, both gases which would bake out in a hot oven. There remains the question of whether or not it would promote spread. It does not seem to do so in my springerli formula, which uses quite a bit.

For reference, here is an example of the biscottini formula (I am open to suggestions regarding formatting columns):

                                                                        as fxn     as oz
0.38   cup   sugar (2.75 oz)                                              3/8       2.75
0.25   stick   unsalted butter, room temperature                           1/4       1
0.5   egg   large eggs, beaten                                             1/2      
0.5   tsp   grated dried orange peel (softened in warm brandy)            1/2      
0.25   tsp   baking soda (use a rounded tsp full)                          1/4      
0.13   tsp   salt                                                           1/8      
0.75   cup   all purpose flour                                             3/4      
0.31   cup   hazelnuts, toasted, husked, coarsely chopped                 5/16     1.
0.22   cup   dried tart pie cherries, chopped and softened in brandy       7/32     1.06  
0.19   lb   bittersweet or semisweet chocolate                             3/16     3
0.13   tsp   vanilla
0.13   tsp   rum, and a few drops of rose water to taste

All batches were ? egg batches - use a ‘full’ 2 cl shot glass full per batch ~ 25 ml if your large eggs are ~50 ml. Beat the egg and add the vanilla, rum, and rose water to it. One can mix several eggs and add proportionate amounts of flavor to them to simplify the measuring. For the record, vanilla/rum/rose is the flavor that says ‘bakery’ when you smell it.
Slicing ? start with a ‘wavy’ serrated knife and finish with a razor sharp knife ? steel it regularly
For those biscottini which are chocolate, I used 3 oz of melted Guittard L’Etoile du Premiere 58% pastilles.
Every batch was flavored with 2 heads (dried) of English lavender, ground in a mortar with 0.125 t of salt, with 0.25 t of cinnamon and anise seed powder and 0.125 t of mace added at the end. Each was also flavored with 0.5 tsp of dried Trovita orange peel softened with hot brandy.

Pay attention to the protocol, cream sugar and butter, mix in eggs carefully, mix in the spices, mix in half of the flour, add the fruit and nuts (be careful to exclude any excess brandy used to soften the fruit) and mix then add the rest of the flour and knead lightly

Position 1 rack in center of oven and preheat to 325 F. Beat sugar and butter until well blended. Beat in egg just until blended. Mix in orange peel, baking soda and salt and spices. Add ? of the flour, and the hazelnuts and dried cherries; stir until well blended. Add remaining flour, stirring until well incorporated.  Knead until it can be rolled. Divide the dough into 3 equal portions. Start a rolled log by hand and then finish it by rolling it between a 4” wide piece of wood and the countertop until it is the length of a cookie sheet. Bake on parchment. Bake 15 - 20 minutes at 325 F, or until the top of the loaf is firm to the touch. Remove to a cooling rack for 10 minutes. Using long wide spatula, transfer the logs to a cutting board. Using a razor sharp knife, cut the warm logs crosswise into 1/2-inch-thick slices. Arrange the slices cut side down on 2 baking sheets. Bake them for 10 minutes. Turn them over and bake bake them again until light golden, about 10 minutes longer. Transfer to a rack and cool completely. They are especially nice if they are coated at one end with a wipe of molten Guittard L’Etoile du Premiere 58%.

Here are the additional varieties I made for Christmas 2010: Azteca (sliced almond, Zante currants, smoke dried Haba?ero and jalape?o chiles or chipotles - be careful) in the chocolate base, Britannica (pine nuts and quadruple English lavender), pine nut and chopped candied lemon peel, dried pear with pistachio nuts and candied ginger, Bing cherry with hazelnuts, and candied lemon peel and sliced almond with chopped chocolate pastilles.


Posted: 12 December 2010 01:49 AM   [ Ignore ]   [ # 1 ]
Sr. Member
Total Posts:  1440
Joined  2008-09-27
baumgrenze - 12 December 2010 05:06 AM

Has anyone seen this phenomenon explained by gurus like Harold McGee or Shirley Corriher?

Yes, this is common knowledge in the baking world.  Not baking soda, per se, but pH level.  More acidic doughs set more quickly in the oven, reducing spread.

Does the soda function as a leavening agent here?

Maybe somewhat with the chocolate, but it might be included as a browning agent;  alkaline doughs brown more easily.

ammonium bicarbonate work as well? It should leave no residue as it disproportionates into ammonia and carbon dioxide, both gases which would bake out in a hot oven.

Haven’t used it, but it’s reputed to be most suitable for very thin items like crackers; the ammonia might not evaporate from something thicker.


If error is corrected whenever it is recognized as such, the path of error is the path of truth.

—Hans Reichenbach

Posted: 12 December 2010 09:20 AM   [ Ignore ]   [ # 2 ]
Sr. Member
Total Posts:  310
Joined  2009-07-24

Thank you to Newbie and CharlesT. Intriguing questions with very good answers. I always appreciate being able to expand my baking knowledge.



Please visit me at for recipes and photos

Posted: 12 December 2010 10:30 AM   [ Ignore ]   [ # 3 ]
Total Posts:  4698
Joined  2008-04-16

Yes to what Charles said, and specifically, I think it is the acidity of the dough with respect to eggs- i.e., if the dough is eggless, like shortbread, then changing the acidity won’t do much to alter the spread.  In a more acidic dough (i.e., less baking soda, which is alkaline) the eggs will set earlier in the baking process, limiting spread. 

As Charles points out, there are other effects of the soda, namely leavening if there is an acidic ingredient for it to react with, and browning.  And as you point out, if the soda is not fully reacted with an acidic ingredient then there is a soapy taste.  So altering the soda may not be the best way to control spread due to unwanted side effects. 

The amount and type of fat also have a profound effect on the spread of the cookie, and that may be an easier way to control spread.  Increasing the butter/cocoa butter will promote spread. 

Your recipe looks very interesting, with lots of beautiful flavorings and variations. smile

Posted: 12 December 2010 10:33 AM   [ Ignore ]   [ # 4 ]
Total Posts:  4698
Joined  2008-04-16

You might also check out Harold McGee’s blog and articles for the NY Times, I thought I remembered that he uses baked baking soda as a substitution for stronger alkalines in baking.  I have it in the back of my mind to try it out, but haven’t yet.  There may be some use for the process in your biscotti.

  Back to top