All grains contain peptides that mimic morphine or endogenous opioid substances. This is where I deal with my latest loaf craving. Get your bread-based exorphin fix here.

Tuesday, March 1, 2011

Straight Dough Method Whole Wheat Focaccia with Garlic

A Quick Whole Wheat Focaccia with Garlic

Fast on the heels of the last Focaccia, the one that requires a 16-hour Biga or preferment to make, I thought I would try a straight-dough version, based on the one given to us by Chef Stephanie at our recent bread-making class.  Chef runs a production kitchen: no time for preferments here!  So the recipe that she gave us for a Rosemary Focaccia is made with a straight dough method, with one bulk fermentation and one proof before baking.

Her recipe was made with bread dough, however, and I wanted a whole wheat version.  The one I made this morning was elaborated with the preferment that I put together yesterday.  My wife likes the taste of it a lot, and so do I, but I wondered: could I make a whole wheat focaccia without that preferment?  Using Chef's recipe, I just swapped out the bread flour with whole wheat flour, and added 1 tsp of honey.  Otherwise it is the same recipe.

Oh, yeah: again I added garlic, but this time I roasted it first and added it when I punched it down after the bulk fermentation.  And I painted on the oil and vegetable seasoning mixture prior to the final proof.

Roasting the Garlic
To roast my garlic, I took 2 small heads of garlic and sliced them horizontally.  I placed these in some foil and baked them for 30 minutes at 350 degrees F.   While still quite warm, I squeezed the half-cloves out of their papery shells and let them cool.

  • 343g tepid water (1 1/2 c)
  • 11g yeast (1 TBspn)
  • 27g olive oil (2 TBspn)
  • 532g whole wheat flour (3 1/4 c)
  • 9g kosher salt (2 tsp)
  • afterthought 1 tsp honey

  • 2 heads of garlic, sliced horizontally, roasted at 350 degrees F in foil for 30 minutes, then peeled

I stuck the water in the microwave for 35 seconds and afterward added the yeast.
I waited until the water cooled to almost room temperature, and it didn't appear to be doing anything, that is why I added the honey -- to see if that would wake it up.
When I began to see a little action, I added the oil to the water.  Then I added the salt to the flour, and added the wet to the dry.
I mixed it up, then kneaded by hand for about 6 minutes or so.

The dough was placed in a covered bowl and set in my Excalibur Dehydrator on the bread rising setting to bulk ferment.

At thirty minutes, I took it out and added the garlic by kneading for a couple of minutes.  (I first tried folding it in, but that didn't work).

Then I put it on a parchment-lined baking tray and gently spread it out with oiled fingers.
After about 5 minutes of rest, I gently spread it out some more, dimpling it as I went.  Then I spread some olive oil and vegetable seasoning on top, and let it proof for 30 minutes.
There was a substantial rise in volume in that time frame, and I put it into the preheated oven.

Chef Stephanie's bread flour focaccia gets baked at 370 degrees for about 30 minutes (well, that is what we determined in the college's convection ovens, from experience; the actual recipe says just to keep it in "until golden in colour").  I expected to see this whole wheat version brown in that time frame, at that temperature.  Unfortunately, I forgot to set the timer.  I think it was in the oven about 15 minutes before I peeked, and the bread was nowhere near done.  I set the time for another 15 minutes at that point.  It still looked a little blonde to me, so I kept it in another 10 minutes.  Even then, it didn't sound hollow when I tapped the bottom, so I stuck it back in for yet another 10 minutes.

still not quite done
Probably baking it for 20 minutes at a higher temperature (like the earlier focaccia, which was in for 20 minutes at 500 degrees F) would have been better.  The long time in the oven just dried it out before it was baked.

This dough rose nicely in the proofing stage; there was little oven spring, though.  The quick rise has to be due to the amount of yeast, which is substantial.  I calculate the amount of yeast to be 3.2% of the amount of flour.  That is about high-normal for a straight dough recipe.  But does it taste yeasty?

No, it doesn't taste yeasty.  But it does taste rather dry, and I suspect that it will stale much faster than the other focaccia.    That could be because of the longer oven time.  This one has more garlic, and that garlic is on the interior, so it hasn't carmelized as much.  The crumb is not quite as holey, but that might have more to do with the garlic and the extra kneading than it does with the fact that there is no vital wheat gluten.  There is not as much spice on the top, it could have used more.

I personally like the taste of the other one better.  But this one will certainly do, in a pinch.

Notes to Myself
  • The reason this dough rose so quickly, so well: (1) the amount of yeast is substantial (2) the yeast was awakened and ready to do its job, and (3) the bulk fermentation was done in a controlled warm environment, my Excalibur.  Can you achieve similar results with less yeast?  With several stretch and folds, like the last recipe?
  • Would this bread also work with roasted peppers?  With olives?  As a free-form loaf?
  • Try baking this at 500 degrees for 20 minutes like the last focaccia you made.
  • Use more oil and more vegetable seasoning.  Don't skimp on this stuff.
  • I was comparing the nutrients of whole wheat flour to all-purpose flour in a previous blog, and I was wondering why vital wheat gluten got added to MarieH's focaccia.  There is actually less protein in a processed flour, so theoretically there is more gluten in whole wheat.  But what there is less of, substantially less, is carbohydrate that yeast can use.  What we should be doing is adding some simple carbohydrate that the yeast can eat -- about 13.8g of carbs per 100g of whole wheat flour that you use.  So, if here we are using 532g of wwflour, you could add as much as 73g of carbohydrate -- in theory.  But what form should that carbohydrate take?  Surely not simple sugar.  And if you added honey, then you would have to adjust the hydration.  You could similarly add some orange juice, but cut back on the water by that volume.  Or -- is this idea even necessary, seeing as you had a fairly nice rise in this loaf without doing all that?


  1. What type of yeast are you using? Personally, I use instant yeast because I never have to deal with blooming / proofing, or occasionally fresh yeast. One less thing to worry about.

  2. oh, and, as far as some of your other questions go... you should be able to throw any topping in there comfortably as long as you saute or roast them so they won't scorch in the oven. You could coax more flavor out of your garlic by roasting it longer, too.

    You probably hurt yourself a bit on this one by kneading the garlic in after your bulk fermentation... I checked an olive levain recipe, and the olives get added in the last minute of initial mixing. over-mixing will over-oxidize the bread and you'll lose flavor, texture and keeping quality.

    Personally, when I make focaccia, I dimple the crap out of it before I throw it in the oven, basically pressing through the dough to the surface below (or at least 3/4 the way there)

  3. Right on all your advice.

    It is true what you say about the instant yeast not requiring proofing. Probably this comes from my using some older recipes, from a time before instant dried yeast was invented (or the 'production kitchen' recipes of my bread instructor chef Steph). But I have had good results by proofing IDyeast, nonetheless. If you get the temperature right, it may even wake up the dough faster, or cause quicker yeast replication, but that is just anecdotal from the odd time I've tried it and it worked beyond my wildest dreams. I guess what I'm saying is, that for a straight dough method, if time is an issue, perhaps it can speed things up, who knows?

    If I had had the garlic ready, I would have thrown it in prior to the bulk fermentation, but alas, no. It wasn't even out of the oven then, and even if it was, it would have been too hot. Adding it by kneading it in after the bulk fermentation definitely hurt the loaf: I was rough with it just when I should have been gentle.

    I would also agree the garlic could have used a bit longer roasting. I was hoping that it would roast a bit more during the bread baking, but that idea didn't work out too well. House smelled great though.

    And yeah, I lost my dimples on this loaf because I did not believe! I obviously didn't press hard enough, and I watched as in the oven, the loaf filled in my fingerprints.

    Thanks for the comments, I'm learning a lot from you.