The Fresh Loaf

News & Information for Amateur Bakers and Artisan Bread Enthusiasts

Experiment on incorporating air into the dough / aerobic yeast metabolism effects

glasgowjames's picture

Experiment on incorporating air into the dough / aerobic yeast metabolism effects

Alright guys, just thought I'd post this up here. Basically, I was just trying to see if it actually made a difference how one kneads. Turns out, yes. Air = good.



The aim of this experiment is to determine whether or not incorporating air whilst kneading dough is on beneficence to the baked product.


Bread baking techniques have long been based on experience and tradition. Arguably, however, this has since led to misconceptions about the processes involved. Many books will tell you that you knead bread to develop the gluten in the flour, whilst others will still claim that kneading is to incorporate air.[i]

The formation of gluten from gliadin and glutelins on kneading has been confirmed. This can then provides a stable structure which can hold the CO2 produced as a by-product of yeast metabolism.[ii]

The incorporation of air then, as the cause of the dough’s increased stretchiness, is disregarded. However, air may still be important. Baker’s yeast, or saccharomyces cerevisiae, can produce CO2 by both anaerobic fermentation and by aerobic respiration (Citric Acid Cycle and Oxidative Phosphorylation).[iii] Aerobic respiration yields CO2 at a ratio of 6:1 for the same mass of substrate. As a result, the deliberate incorporation of air into the microstructure of bread dough, which we know has the capacity to store CO2, may have an effect on the final product.[iv]


  1. The rise of the dough with air incorporated will be faster and greater.
  2. The flavour difference of the breads will be insignificant in those loaves proved at room temperature. Flavours will be contrasting in retarded loaves, due to different levels of yeast stress.



  • One simple white bread dough was made from the following:

500g strong white flour

375g tepid water

10g salt

10g instant yeast

  • This dough was mixed thoroughly and split into two smaller doughs of equal mass (to the nearest 1g)
  • One dough was kneaded in a stand mixer (Kenwood Chef KM001) on slow setting with the dough hook. This was intended to incorporate as little air as possible into the mixture.
  • At the same time, the other dough was kneaded by hand by slapping and folding, deliberately incorporating as much air as possible.
  • When both doughs were deemed finished, they were tested for gluten development with the windowpane test. Small samples were pinched off and stretched. The length at which there was breakage was equal.
  • Both doughs were then split in two equal balls, making four equal sized doughs in total; two oxygenated, two non-oxygenated.
  • The doughs were left to prove for 1.5 hours.
  • Each dough was shaped until of maximum tightness available by the skill of the baker and placed in floured proving baskets to prove.
  • One oxy and one non-oxy dough were placed in the fridge to prove for 14 hours, whilst the others proved for just one hour before baking.
  • The loaves were baked at 220 degrees Celsius for 30 minutes, next to each other in the oven (turned and switched halfway through)
  • The loaves were then judged for crust and crumb, before submitted for a blind taste test by a third party.



  • After kneading, although the water content was the same and the windowpane test identical, the non-oxygenated dough had a marginally looser consistency, as if it had a higher hydration



  • After the first prove, the oxygenated dough had splayed out less than the non-oxy, again consistent with a lower hydration dough
  • When slapped, the oxygenated dough was more resonant than the non-oxy.



  • When shaped, it took less effort to get the oxygenated dough tighter than the non-oxygenated.




  • After the bake of the SHORT PROVED BREAD, the size of the oxygenated bread was significantly greater.



  • The crumb of the oxygenated bread was marginally more consistent, with slightly more uniform, smaller bubbles
  • Due to its increased size, the oxygenated bread had a less substantial crust.
  • On blind tasting, no difference could be discerned or preference made in flavour or texture




  • The LONG PROVED DOUGH showed an amplification of the differences of the short proved. The oxygenated dough significantly larger in the basket.



  • After the bake, it could be seen that the oxygenated bread was much larger.




  • The crumb of the oxygenated bread was much superior; a more even distribution of consistently sized bubbles.
  • Oven spring of the oxygenated bread was significantly greater
  • Consistent with their size, the oxygenated bread had a thinner crust than the non-oxygenated.
  • On blind tasting, there was an absolute and definitive preference for the oxygenated bread. There was deemed greater complexity of flavour.




Summary and Conclusion:

  • Both hypotheses were correct. Therefore when making bread, hand-kneading with the aim of incorporating as much air as possible into the dough is recommended.
  • The common assumption that shaping incorporates air into the bread has little support. Here, shaping was consistent and did not lead to oxygenation of the non-oxy dough during the long prove.
  • If proving in the short term, oxygenation increases size of the bread (greater yield) and the consistency of the crumb (even bubbles with decreased range of diameter)
  • If proving in the long term, the differences in oxygenation are amplified significantly. The yield is much greater, the crumb is much more consistent, oven spring is easier to achieve and the flavour is more complex and palatable.
  • However, it must be taken into account the larger size of the bread means an inferior crust for the same baking time. Therefore, with hand kneading the baking time should be increased slightly.



Hypotheses to explain results requiring further research:

  • The surface area of the bread is not increased when shaping to a level that can provide all the yeast with oxygen. Therefore
  • The dough of the oxygenated bread seemed tighter and less hydrated because the increased aerobic respiration by the yeast caused increased pockets of CO2 to form, therefore increasing tension
  • Flavour was increased on retardation due to increased yeast stress. When yeast is stressed it produces by-products, giving flavour. The stress of temperature change was amplified with the stress of switching from pure aerobic to anaerobic respiration once the incorporated oxygen was depleted.


  • The blind taste testing was single blind, and the subjects’ conclusions could have been affected by what they saw.
  • This was a small scale experiment. Therefore, stress testing equipment to determine the tensile strength and so gluten formation was not available. There may, therefore, have been small differences in gluten formation between the two kneading methods.

[i] Bertinet R; Dough; Kyle Cathie; UK 2008

[ii] Abonyi T et al; Gluten formation from flour of kernels in developing wheat grain; Cereal Research Communications; Mar 2010;38,1

[iii] Nissen TL et al; Anaerobic and aerobic batch cultivations of Saccharomyces cerevisiae mutants impaired in glycerol synthesis; Yeast; Mar 2000;30;16(5):463-74.

[iv] Gust LK; Experimental Investigation of Yeast Activity and Carbon Dioxide Production in Bread Raising; 2.671 Measurement and Instrumentation; 2010;9-12

GAPOMA's picture

Well done James!  The photos show a definite size difference in the loaves, and it is clearly maximized with a longer rise.  The variable that is difficult to control is the difference between machine and hand kneading.  

In the past I generally have just used a machine to knead my doughs, mostly because I'm lazy.  Because I usually use a machine, I have a couple of questions about hand kneading:

   1)  Why does hand kneading infuse more air into the dough?  
   2)  What specifically do you do when hand kneading to try to maximize the air incorporated?

Looks like I'm going to have to start hand kneading more often (and work on my technique)!!

- Greg

glasgowjames's picture

Hi Greg!

Basically, it is all about the surface area of dough you are exposing to the air, and then what proportion of that you are potentially incorporating into the dough. With a mixer and a dough hook, it is designed to twirl the dough around itself, using leverage and twisting motions to develop the gluten quickly and effectively. However, it almost does this 'within itaelf', ie much of the same proportion of the dough is exposed to the air at a time and it is never more than a solid round lump - the worst Surface Area to volume ratio for incorporating air.

As for which technique - you want to stretch and fold. The traditional French method, involving grasping the dough and flipping it over itself, slapping down on your surface. In my opinion, Richard Bertinet has the best of the guides on how to do this effectively, so check out Dough or Crust.


Cheers and happy bread making!




GAPOMA's picture


Well stretch & fold is certainly easy enough!  And I like the concept of surface area to volume ratio.  I'm definitely trying this soon!  Thanks James!!

- Greg

daveazar531's picture

This was very well documented, you have reduced my electrolux envy 


I will be sticking with the good old Stretch and Fold for a little while longer

All at Sea's picture
All at Sea

... what a delightfully lucid account, James. Bertinet bangs on considerably about how his dough-working method whips air into the dough - and I use his technique all the time, it's so easy and reliable no matter how high the hydration. And your experiment demonstrates the benefits of incorporating air beautifully. Bravo and thank you!

All at Sea

mwilson's picture

Nice experiment. However it is highly flawed from an understanding perspective.

The improved volume can be attributed to increased dough strength via the effects of stretch and fold upon the gluten network rather than the incorporation of oxygen. Think of gluten as a chain of interlocked elastic bands. Take this chain and fold it back on itself - it will be stronger harder to stretch. Please read this article on dough strength.

A dough that is beaten or whipped really fast will certainly incorporate a lot of  air. A slow and gentle knead will incorporate less air but stretch and fold will barely incorporate any. Think of air like water. You fold the dough over itself and as it comes together, the air is all but pushed out...

An aerated dough will appear noticeably whiter in colour.

Yeast in the presence of oxygen will consume sugar and reproduce (bud). CO2 is produced during fermentation which is anaerobic only. 


fancy4baking's picture

That's why i tend to believe that the French Style Stretch & Fold is so effective in this area than any other methods, particularly when dealing with high hydration doughs. Personally i have tried this method (French S&F) with dough higher than 80% and the result of the bread was awesome to a great deal. Thanks for sharing James :)


dabrownman's picture

I routinely do high hydration doughs in the KA 3 with the dough hook for 4 minutes before doing (4) S&F's, every 15 minutes,  to finish it off over then next hour or so.  Maybe you get the best of both worlds this way?

I'm going to always do 95% hydration ciabatta for 10 minutes on KA 8 with the dough hook for 10 minutes before doing (6) S&F 's over 1 1/2 hours to get it right.  Do slap and folds with that and you will need to spray clean the ceiling, walls, and apprentice, not that she doesn't need it :-)

It is nice to know that hand mixing is stil  the way to go though.

suave's picture

I have a question for you - how do you know it is really oxygenated?

mcs's picture

...that the dough mixed with the mixer is more 'oxygenated' as the hook repeatedly dives in and out of the dough, than the dough mixed by hand and strengthened with the stretch and fold.   Therefore showing that the dough strengthened by hand (stretch and fold) is superior because less oxygen is incorporated into the dough.
I've got to disagree with Bertinet on his 'whipping air into the dough' explanation (not his method, mind you) and agree instead with Michael's (mwilson) explanation above (and below) instead.

A dough that is beaten or whipped really fast will certainly incorporate a lot of  air. A slow and gentle knead will incorporate less air but stretch and fold will barely incorporate any. Think of air like water. You fold the dough over itself and as it comes together, the air is all but pushed out...   -mwilson

There you go.


glasgowjames's picture

The dough hook twists the dough inside of itself. It is extremely effective at developing gluten because that is what it is designed to do (think about the leverage on the 'elastic bands' that it provides). Also, remember, as I outlined in the method, they were worked until their windowpane results were undetectably different.

We're talking about yeast here, not air pockets, you understand, so perhaps I have not made myself clear. We are looking for maximum surface area exposure for continued aerobic metabolism for as long as possible. Thinking about it as pockets of airbubbles is a little simplistic; oxygen dissolves. Anyone a brewer? Look at beer. If you transfer it from one fermenter to another in a reckless way, you risk oxidising the beer and causing huge harm.

Mini Oven's picture
Mini Oven

I'm coming from a different angle.  In ceramics, clay is kneaded to press the air (or any other gas) out of the mass.  Many potters also make bread.  I don't think you can convince any one of them that kneading works air into the dough forming pockets.

Mebake's picture

Hi, James. Thanks for sharing your really nice experiment!

In my opinion, when a dough is first mixed (flour, water, yeast and salt) , it will surely incorporate air ,regardless of the mixing technique. It is only when a dough starts to come together in a coherent matrix, that it will stop accepting any air... and will retain dissolved air, if any. Therefore, the idea of incorporating air, or even exposing a dough surface to more air during kneading is of no significance to the aerobic activity of yeast.

As to the results you have obtained, i believe that the dough hook is not nearly as efficient in developing a dough as hand mixing does, which is evident in the crumb pictures of both breads. A taller loaf means a well developed less slack dough, whereas a shorter slack one, means the opposite.  The action of kneading by hand stretches a dough and folds it upon itself, which results in maximum gluten development. The dough hook twists, and tears a dough into gluten development, and the dough matrix isn't developed to the maximum. In most cases a combination of the two mixing techniques is advisable.

Having said all that, i'm a Hand mixing die hard, if not for my bad back. I now mix a dough into a medium development, and continue development by hand.