For my second attempt at gluten-free bread, I tried to modify the technique rather than the contents of the recipe. There were some minor deviations from GFB#01 ingredients, but I tried not to make any radical changes. When trying to do “good science”, it’s always best to change only 1 variable at a time so the experimenter can clearly identify the effect of that variable. In some experiments, this can be VERY difficult, but with something like a bread recipe it should be pretty easy. The reason I tweaked multiple variables here is that I wanted to make a smaller batch of bread. If I’m making disastrously hideous bricks of pseudo-bread in my quest for a good gluten-free loaf, I really don’t want to be making multiple loaves of nasty with each batch. Once I get a good recipe, I’ll work on scaling up to a “2 full-sized loaves” version of the recipe. By the way, “scaling up” is one of the big tasks that chemical engineers work on; a chemist figures out how to make a gram of material then hands the procedure off to a chemical engineer who figures out how to make 3 truckloads. But back to bread…

For GFB#02, I’m going to try and do a single mix and rise rather than letting the yeast develop as a starter before the rest of the flour is mixed in. There are a couple other little tweaks that I’ll mention when they come up in the experimental procedure, uh, I mean, recipe.


3/4 cup brown rice flour

3/4 cup tapioca starch

1/3 cup ground flaxseed meal

1 teaspoon xanthan gum

1 teaspoon dry yeast

1/2 tablespoon sugar

1/2 teaspoon salt

1 cup 110°F water (a “heavy” 1 cup…)

All the dry ingredients were combined in the bowl of a stand mixer with a dough hook. {This is one of the minor tweaks over GFB#01 in which I mixed by hand…} With the mixer running, the warm water was slowly poured in. After mixing for ~1 minute, the sides were scraped down and the dough was mixed for an additional 1-2 minutes. I’m calling this a “dough”, but it is very loose and soft, so it’s really more of a thick batter than a dough, but I’ll probably randomly swap those two terms. The dough was transferred to a glass bowl and allowed to rise. My oven hadn’t completely cooled to room temperature from its last use, so I put the dough in the slightly-above-room-temp oven to give the yeast a good kickstart. After ~1 hour, the dough had doubled in volume {Good sign!} and was transferred to a loaf pan. The dough was allowed to rise in the loaf pan for another ~1 hour and the oven was preheated to 435°F. The loaf was put in the oven, after ~15 minutes the temperature was lowered to 400°F, and the loaf was baked for ~25 additional minutes. The internal temperature of the loaf was check and was only at ~165°F, so it was returned to the oven for another 25 minutes. Internal loaf temperature was ~200°F, so the loaf was removed and allowed to cool.


Not much different than GFB#01… the loaf is a little lighter, but it’s still pretty dense. The crumb is a little sticky/gummy, but the air bubble structure is OK. Similar to GFB#01, this loaf did not have any noticeable rise/bloom when it was initially put in the oven, in fact, it seems to have shrunk a little. I though that starting with a warmer over would get the trapped gases to expand quickly enough to plump up the loaf before the starches and other gluten substitutes had a chance to set, but that doesn’t seem to be the case. Flavor is comparable to GFB#01 which is expected since the ingredients didn’t really change. There’s still a slight aftertaste, but it’s not as obvious with the slightly less dense loaf. I’m pretty pleased with the crust formation, the loaf has a nice crunchy/chewy crust that’s developed a bit of nutty flavor. The top crust is still kind of white and nasty looking, but if I close my eyes it’s good.Image

Revisions in the next iteration:

I’m not sure what I learned from GFB#02. From the processing side, it looks like doing a 1-step mix/rise is a little better than messing with a starter, but the poor bloom and gummy texture are not great. I’ve read a number of sources that say rice-flour-based gluten-free breads tend to have a grainy texture, and I’m clearly avoiding that problem, but dense and gummy aren’t really that much better than grainy. I think I’m going to try modifying the starch source next. Tapioca starch is supposed to be good for the crust, but a number of sources say that potato starch gives better internal structure. I think for GFB#03, I’ll try swapping potato starch for the tapioca starch. Ultimately, I’d imagine that some combination of the two will be best, but I’ll go all-potato for now, just to see what happens.

A couple of other modifications I’m going to look at in future attempts are:

1. Sorghum flour – I’ve seen a few very positive descriptions of sorghum flour for gluten-free breads, and it’s available at one of my local grocery stores. I might also try some of the bean flours… many people note a significant “beany” flavor in these flours so they might not be the best choice for a lighter flavored plain sandwich bread.

2. Other leavening agents – In the gluten-free pizza crust recipe I tried, baking powder was used as well as yeast, maybe I’m not getting good bloom because I need a little extra “oomph” in the oven. This might also call for some dried buttermilk powder to provide some acid, and maybe help with browning the crust.

3. Speaking of crust… The dusty white looking crust is a real turn-off. I’m tempted to brush it with a little oil to intensify the heat transfer and maybe smooth it out a bit. If I can get a good, strong bloom when the loaves go into the oven, the top crust issue may solve itself, but a little oil or egg or milk on that top crust might help its appearance and texture.

I do realize that at this point I’m putting some real effort into re-inventing the wheel here. There are a LOT of gluten-free bread recipes on the internet, and many of them make very good bread. The reason I’m doing this is because there is a lot of variability in the “good” recipes, and I have no feel for what effect different ingredients have on the final product. This makes it very hard to evaluate different recipes without making them all. By systematically varying the ingredients, I can develop that “feel” for different flours and starches and other adjuncts… When in doubt, I fall back on my training as a scientist; change 1 variable, repeat the experiment, evaluate the results, repeat, repeat, repeat.

Time to bake some more…