Fermentation: a science-based look suggests RT is better for flavor

[TXCraig1]

There are two main categories of extended fermentation 1) cold fermentation ("CF") which is defined here as temperatures common found in a refrigerator, i.e. 34-39F (1-4C), and 2) room temperature fermentation ("RT") which broadly encompasses everything else. For purposes of this post, RT is loosely defined as 60-77F (16-25C) +/- as it is fairly common for RT to fall into this range though it could be warmer or cooler.

On more than one occasion, I have commented unequivocally that with respect to pizza quality, flavor in particular, RT is superior to CF in every way and that there is nothing added as the direct result of CF.  CF may indirectly help you make better pizza because it allows you to do an extended fermentation that would otherwise not be possible for any number of logistical reasons; notwithstanding, it does not inherently add anything to pizza quality that RT fermentation doesn't do as well or better. I stand by these comments and believe the science laid out in this post supports the position that CF is suboptimal at best.

This post will look at two broad areas of aroma and flavor development in dough 1) direct byproducts of yeast and bacterial fermentation and growth (acids in particular), and 2) enzymatic activity of both enzymes native to the flour and those produced the microorganisms in the dough. The information generally applies to all pizza dough, however I believe it is particularly relevant to Neapolitan pizza and the use of unmalted flour as will be discussed below.

Direct byproducts of fermentation and growth
In past discussions, it has been generally assumed that CF was, for the most part, a different means to the same end as RT: more flavor. I've argued that RT is superior, and others have argued that CF can get you to the same place though perhaps taking a longer time to do so. However, up until a recent thread that derailed into this subject, I don't remember anyone ever citing a specific reason as to why CF might have flavor-creating advantages over RT. In the discussion, a quote from an article by well-respected cookbook author and James Beard Award winner Shirley Corriher (also a non-practicing biochemist) was offered as proof that CF added something missed at room temperature:

"There are bacteria in the dough from the beginning, but as long as the yeast is very active, it consumes sugars as quickly as they're produced, leaving no food for the bacteria, which also like sugar. But when bakers chill a dough and slow down its rise, the cold dramatically reduces yeast activity. The bacteria, on the other hand, function well even in cold temperatures, so they now have an opportunity to thrive, producing many more marvelously flavorful acids." http://www.finecooking.com/articles/yeast-role-bread-baking.aspx?pg=2

Her theory intuitively did not make sense to me, so I thought it would be worth spending some time to look into it deeper. After doing some research, I believe there are at least five possible problems with her statement: 1) the statement that yeast consume all the sugars as quickly as they are produced, 2) the assumption that yeast and lactic acid bacteria (a.k.a. "LAB" or "lactics") necessarily compete for the same resources, 3) the statement that the relevant LAB function well, let alone thrive, in cold temperatures, 4) the assumption that the relevant LAB produce meaningful quantities of acids at low temperatures, and 5) the statement that the relevant LAB are meaningfully less affected by cold than yeast.

1) Does yeast consume all the sugars as quickly as they are produced?
A study titled Interactions between Saccharomyces cerevisiae and lactic acid bacteria in sourdough (1)  found that the maximum growth rate of all but one of the LAB species studied was unaffected by competition for sugar with the baker's yeast (S. cerevisiae). The authors wrote, "antagonism for carbohydrates between yeast and the rest lactic acid bacteria studied, all of them being able to ferment glucose, fructose, maltose and sucrose (unpublished data), seemed much less important, as their maximum specific growth rates did not significantly change in the presence of the yeast." The authors also noted, "Growth of lactic acid bacteria is believed to be promoted when co-cultured with yeast... regardless the antagonism for the main carbon source [sugar]."

In another study, The sourdough microflora. Interactions between lactic acid bacteria and yeasts: metabolism of carbohydrates (2) , the researchers found that when grown in a wheat flour hydrolysate medium (wheat starch converted to sugars), both of the species of LAB studied grew at a faster rate when cultured along with yeast than when cultured alone (see Figure 1 below). These studies would seem to suggest that at least certain LAB can compete successfully with yeast for sugar, and while I found examples of LAB growth being suppressed when co-cultured with yeast, I didn't find a single study that suggested yeast could outcompete LAB 100% and consume ALL the sugars produced.

2) Do yeast and LAB necessarily compete for the same resources?
Adding to the studies cited above, in The fate of various sugars in fermenting sponges and doughs, Lee et al. (1959) found that when no sugar was added to the formulation, glucose and fructose were absent in bread probably because they were utilized by yeast. Maltose however was still present in the dough (3).  Many of the LAB most frequently identified in sourdoughs are able to ferment maltose (4, 5).   

3) Do the relevant LAB function well in low temperatures?
I could find no evidence in the scientific literature that the LAB species typically found in wheat doughs function well at all at refrigeration temps. In fact, it would appear likely that the opposite is the case; a study, Environmental stress responses in Lactobacillus: A review (6), identified six species of LAB, several of which are common in sourdough, that won't grow in wheat flour hydrolysate at 7C or below, and a study titled Glutathione Protects Lactobacillus sanfranciscensis against Freeze-Thawing, Freeze-Drying, and Cold Treatment (7) shows two strains of L. sanfranciscensis, one of the most prevalent and competitive LAB species found in bread (8) start to die off almost immediately after being exposed to refrigeration temperatures and that the mortality rate is as high as 99% after only a week at 4C (9).  (Figure 2)

Of the dozens of scientific studies I looked at, I didn't find a single reference that discussed the occurrence of psychrotrophic/philic (cold tolerant) bacteria of any species in a bread dough. The closest I came up with was in The Lactic Acid Bacteria: A Literature Survey which reported that "L. sake, L. curvatus, and L. plantarum are capable of growing at temperatures of 2°C to 8°C and so should be considered psychrophilic, although they grow only slowly at those temperatures " (10) (emphasis added is mine).  L. plantarum is sometimes found in sourdough. Also, Rodgers et al. (1978) notes the presence of psychrotrophic bacteria in most wheat flour samples though it is in the context of food safety (ever notice that every bag of GM flour says not to eat raw dough?), and there is no indication that those identified were desirable LAB (11).

The information I did find about LAB activity at cold temperatures was generally related food spoilage, meats and sauerkraut/kimchi in particular. For example, a study titled Behavior of Psychrotrophic Lactic Acid Bacteria Isolated from Spoiling Cooked Meat Products (12)  found several species of LAB that would grow at 4C, however these are not species found in breads, and at 4C it took 1-2 days for the bacteria to simply double (13) . Under ideal conditions bacteria may double as fast as every 15 minutes – almost 100X faster (14).  The bottom line is that the LAB that survive at low temperatures still grow very slowly and are generally not found in bread dough.

In fresh (cake) yeast, the yeast outnumber the LAB by ~10,000:1 (15,16)  with a dry yeast, it's 100,000:1 (17)  or more if IDY. Even when added to the flour and any LAB there, the yeast greatly outnumber the LAB (18, 11) 11. By comparison, in a typical sourdough, LAB outnumbers yeast by 100:1 (19).  To get to a meaningful level of LAB in a baker's yeast dough, one has to wonder how the slow growth rate at CF temps could result in a lactic dough in any reasonable fermentation time?

4) Do LAB produce meaningful quantities of acids at low temperatures?
Aasen et al. found lactate production rate declined significantly with decreasing temperature in Influence of complex nutrients, temperature and pH on bacteriocin production by Lactobacillus sakei though the study did look at temperatures below 20C (20).  I noted the study because this species has been identified in pizza in Naples (21).

A study titled The effect of temperature and pH on the growth of lactic acid bacteria: a pH-auxostat study (Adamberg et al. 2003) shows virtually no growth and zero lactate production below 10C for the two species of LAB subjected to low temperatures, though it should be noted that neither is commonly associated with sourdough (9). The charts for the other three species studied all appear to suggest there would also be no growth or lactate production at refrigeration temperatures. (Figure 3)

There is very little research I could find related to acid production by LAB at low temperatures. I suspect this has a lot to do with the apparent fact that LAB don't perform well at low temperatures, so it isn't very interesting and probably not worth the work necessary to do so. I found only two studies that looked at the relationship between temperature and growth of LAB across a wide swath of temperatures including those below 10C.  Adamberg et al. (9) referenced above comments "the fact that even widely used LAB are relatively poorly characterized in respect of response to changing pH and temperature can be explained by the absence of convenient and effective methods for quantitative characterization of bacteria."

5) Are relevant LAB species meaningfully less affected by cold than yeast?
Of course there are bacteria that can survive, if not thrive, in very cold environments, however these are not the bacteria found in a typical wheat dough, and their metabolism says little or nothing about the LAB that are important to baking.  There is not a lot of information I could find describing the activity of relevant LAB species across low temperatures. The vast majority of studies involving LAB in wheat doughs were focused near the optimum growth temperatures (where the yeast and LAB grow best) and don't go below 20-30C.

As noted above, there is not a lot of research on LAB growth at low temperatures; possibly the most on point is a study titled Modeling of Growth of Lactobacillus sanfranciscensis and Candida milleri in Response to Process Parameters of Sourdough Fermentation (22) . This study clearly shows that like yeast, the two strains of LAB studied, both found in bread, are dramatically slowed to near the point of dormancy at CF temperatures. It also shows that the optimum growth temperature for the LAB is higher than the yeast (Figure 4). The other study that looked at LAB across a wide swath of cooler temperatures is the above mentioned Adamberg et al. (9) paper which also showed minimal or no growth at CF temps.

Everything I found appears to cast significant doubt on Ms. Corriher's statement. I looked at over 50 scientific papers and I didn't see anything that made me think maybe there was something to it. Hoping for some clarification on exactly what she meant and the science as she understands it, I unsuccessfully tried to contact her through a blog where she is a member and also through her publisher. Unable to contact Ms. Corriher, I reached out to Professor Michael Gänzle (lead author of the above mentioned growth modeling study) for his thoughts on the question. He is unquestionably qualified to give a meaningful opinion on this matter:

"Dr. Michael Gänzle is Professor and Canada Research Chair in Food Microbiology and Probiotics in the Department of Agricultural, Food and Nutritional Science. Current research projects focus on the functional characterization of lactic acid bacteria for use as starter cultures or probiotic cultures in food with a focus on cereal-associated lactic acid bacteria; production of oligosaccharides from sucrose or lactose by lactic acid bacteria and biological activities of oligosaccharides; and intestinal microbial ecology with focus on the use of prebiotic carbohydrates and dietary fibre to improve host health."(23) 

I asked Dr. Gänzle (omitting the pleasantries):

"This specifically relates to dough leavened with baker's yeast.  In my opinion, a long, slow ferment at room temperature (~16-24C) is superior for developing flavor and texture as compared to cold fermenting (<4C). In the course of discussion, a quote was put forward from  Shirley Corriher, a very well respected author:

[Corriher quote from above]

Two parts of this quote don't make sense to me:

1)   In light of Gänzle et al. (1998), the underlined portion doesn't make sense to me. Granted the LAB is probably not L. sanfranciscensis, notwithstanding, I would think other LAB would react similarly to changes in temperature? Is there enough difference in activity between the relevant yeast and bacteria below 4C to make a difference over a 2 day +/- period?
2)   Is there even enough bacteria relative to the yeast in a typical baker's yeast leavened dough to make a difference?"
 
To which he kindly replied (emphasis added is mine):

the question relates to yeast leavened dough, not sourdough?

Let's stay with sourdough for a moment, specifically sourdough used for leavening (and containing C. milleri and L. sanfranciscensis): here, cold (refrigerated) fermentation favours yeast over lactics; baker's that want to reduce the acidity in the product typically ferment cold (<20°C down to 4 -8°C)

Yeast dough without added lactics will turn into a sourdough after 8 - 16 h of fermentation at room temperature. How long that takes depends on the yeast and flour, fresh yeast has more lactics than dried yeast, i.e. the dough acidifies faster; whole grain flours have more lactics than white flours, i.e. the dough acidifies also faster. I never read anything about how long the lactics take to grow to reasonable numbers in the fridge, I assume it takes at least twice as long compared to room temperature.

Refrigeration does not inhibit enzyme activity (amylase, protease, a few others). I assume that much of the flavour impact of long fermentation results from more time given to cereal enzymes to break down starch and protein.

Extended fermentation of yeast dough for flavour formation seems a bit pedestrian. The use of a proper sourdough starter will produce superior results in shorter time. 

I followed up with:

"One point I'd like to clarify, when you write that refrigeration does not inhibit amylase activity, is that to say that activity at refrigeration temps isn't meaningfully less than activity at room temperature? I would have expected that enzyme activity would follow the Arrhenius equation and slow as the temperature drops."

To which he replied:

with "not inhibited" I meant slower but not completely inactive...

Professor Gänzle's comments, this like the research cited above, would also seem to put to rest the idea that the lactic acid bacteria in typical yeast dough are somehow benefited by refrigeration temperatures and thus have the enhanced opportunity to feed and produce acids while under refrigeration.  In fact, giving consideration to everything noted above, it appears likely that the opposite is true – that refrigeration meaningfully increases the time it takes for lactic activity to rise to a meaningful level if it doesn't actually prevent in from reaching a meaningful level altogether.   

Enzymatic activity
In a dough, enzymes not only break down starch in the flour into the sugar consumed by yeast and bacteria during fermentation, they are also indirectly responsible for creating much of the aroma and flavor in the dough. The action of enzymes, both those present in the flour and those produced by the yeast and bacteria, creates many of the molecular precursors of aroma and flavor in the baked product. Protease enzymes (largely a product of the LAB) can also contribute directly to flavor by creating peptides. Three of the most important ways enzymes in the dough indirectly contribute to flavor and aroma is by breaking down starch molecules into sugars, by breaking amino acids off of proteins, and by the action of lipoxygenase on fatty acids.(24, 25, 26)   

Chemical reactions are the result of molecules slamming into each other. The more energy the molecules have, the harder they hit and the more likelihood that chemical bonds will break and reactions will take place. The minimum energy required for a particular reaction to take place is known as the activation energy.  Higher temperature = more energy = more reactions. We experience this all the time whether we realize it or not; a familiar example is that pizza bakes faster at higher temperatures. 

Through a variety of mechanisms, different enzymes lower the activation energy of specific reactions to levels where they can take place at functional temperatures. Their importance cannot by understated. For example, without enzymes there would be all but zero starch broken down into sugars at any fermentation temperature. 

Enzyme activity is temperature dependent. Enzyme reactions follow the same principles of other chemical reactions. They have an activation energy requirement, and as such, all other things being equal, there are more enzyme-catalyzed reactions at higher temperatures - up until the point where the heat starts to break down the enzyme itself. A 10C increase in temperature will increase the activity of most enzymes by 50-100% (27) ; likewise a 10C reduction can cut enzyme activity by half – or more!  Enzymes encountered in baking tend to be at the top of this range (30).

The enzyme responsible for breaking down grain starch into sugar is amylase. There are two forms of amylase: α-amylase and β-amylase. α-amylase randomly breaks starch in the middle of the molecule while β-amylase breaks maltose molecules off the end of the starch chain. The two work together; because β-amylase can only act on one end of damaged starch molecules, the action of α-amylase breaking apart starch into smaller pieces creates many more locations for the β-amylase to act (28).

α-amylase is only found in grain that has begun to sprout, and it accumulates rapidly after sprouting starts. β-amylase is found un-sprouted grains and does not increase much after sprouting. Both are activated by water. Enzyme (amylase) activity in flour is typically reported as the "falling number." Typical malted bread flour might have a falling number of 220-250. Unmalted flours such as '00' used for Neapolitan pizza have much less enzymatic activity and are much higher. Caputo Pizzeria, for example, is 340-360. The higher the number, the lower the enzyme activity. Low enzyme activity is typically adjusted by the addition of malted barley which is high in α-amylase.(29)   This is what we call "malted flour."

Using the activation energy figure for wheat α or β-amylase and the Arrhenius equation, it is a simple matter to calculate the difference in activity between CF (at 3C/37F) and RT (at 19C/66F) to be ~3-4X meaning that dough would need to be held at 3C for up to four times as long as at 19C to convert a similar amount of starch to sugars (30, 31).   The relationship for other enzymes in the dough is similar.

Enzyme activity is also a function of concentration: all other things being equal, more enzymes = more activity. Perhaps when using malted flour with its added α-amylase, it doesn't make as much difference if you CF or RT, however, with the low enzyme activity of the unmalted flour used in Neapolitan dough, the importance of optimizing temperature is certainly heightened. 

With enzymes accounting for a very large percentage of the aroma and flavor in the baked product (27), reducing the enzymatic activity by upwards of 75% during CF, as opposed to 19C RT, almost certainly has a significant negative impact on flavor development to say nothing for the corresponding decrease in the yeast and LAB growth and acid and other flavor compound/precursor production -  likely taking as long as 4 days at CF temps to achieve what is done by the enzymes in 24 hours at RT. Also, this is to say nothing for the protease enzymes and acids produced by the LAB which may never realize levels produced at RT in any reasonable amount of time during CF.

I also found it interesting that amylase is relatively heat stable up to about 70C for very short periods of time (4), however it can lose activity at much lower temperatures if held there for as little as an hour. In Figure 5, you can see that this particular example of barley α-amylase is stable indefinitely at temperatures up to about 20C/68F. Above that temperature, it begins to lose activity fairly quickly over extended fermentation time (31). I'm wondering if this is at least part of the reason why more than one member of this forum has had their best results near this temperature? If wheat amylase is similar, perhaps this is approximately the temperature where yeast activity, LAB activity, and enzyme activity is maximized for extended periods of time.

For more background information on enzymes, particularly as related to baking, this is a good read: http://blogs.scientificamerican.com/guest-blog/enzymes-the-little-molecules-that-bake-bread/

It should be noted that CF is not without certain advantages. The key benefits of CF come in the form of equipment requirement, logistics, and margin of error. These are not immaterial.

Equipment requirements
Extended RT fermentation generally requires precise temperature control. While pretty much everyone has access to a refrigerator, few people have the equipment necessary to maintain temperatures in the RT range with sufficient precision to assure consistent, predictable results nor are many people interested in buying a wine refrigerator just to make dough or going to the trouble of implementing an ad-hoc solutions such as this: http://www.pizzamaking.com/forum/index.php/topic,18509.msg179991.html#msg179991

Logistics
The window of usability of a CF dough may be several days to more than a week without significant loss of quality. With RT, the window of usability is measured in hours. The extended usability made possible by CF allows for considerable flexibility in the timing of when dough is made and used - something significantly more difficult or in many cases not possible with RT. CF also allows for big changes in plans such as pushing back the time you were planning to bake by several days – something else that probably can't be done with at all with RT and almost certainly not without significant loss of quality.

Margin of error
In order to have predictable and consistent results with extended RT fermentation, both accurate and precise yeast measurement is necessary – often beyond the measurement precision most people have readily at their disposal. Most people who make dough don't have a scale let alone a scale with 0.01g or better precision. With RT, small measurement error may cause a dough to be ready many hours before or after the targeted time.  CF is much more forgiving – quantities measured are generally much bigger and accuracy and precision are much less critical. Even relatively large yeast measurement errors may have little or no effect on the quality of the final product.

Given these advantages over RT, it's hardly surprising that many chefs and cookbook authors often recommend (and use) CF as a method to develop flavor in dough. The alternative is simply not workable for, or sufficiently interesting to, a large majority of people (and restaurants), dramatically limiting the appeal of RT and thereby the appeal of a chef or cookbook author who would recommend such. It is these reasons of convenience, simplicity, and practicality, but not flavor development, why CF is often recommended; that and perhaps a lack of knowing any better.

Conclusion
The available scientific research appears to confirm the belief that the only reasons to use CF are related to simplicity, ease, and logistics. Nothing was found in the published literature that suggests any reason why CF might have any flavor development advantage over RT. On the contrary, the published research appears to suggest significant limitations of CF when it comes to flavor development. At a minimum, it would appear that CF needs as much as 4X the amount of time to achieve similar results as RT, and it's possible that certain aspects of flavor development related to LAB growth and lactate production do not happen at a significant level during CF in a period of time short enough for the dough not to degrade significantly in other quality aspects.

If you disagree with anything I wrote, please quote the exact words you disagree with, state specifically where you disagree, and cite a reference to support your argument. 

References:
(Several of these are hidden behind a paywall. If you would like a copy of any of them, PM me with your email, and I'll send it to you).
1.   Paramithiotis, Spiros, et al. "Interactions between Saccharomyces cerevisiae and lactic acid bacteria in sourdough." Process Biochemistry 41.12 (2006): 2429-2433.
2.   Gobbetti, M., A. Corsetti, and J. Rossi. "The sourdough microflora. Interactions between lactic acid bacteria and yeasts: metabolism of carbohydrates." Applied Microbiology and Biotechnology 41.4 (1994): 456-460.
3.   Maga, Joseph A., and Yeshajaha Pomeranz. "Bread flavor." Critical Reviews in Food Science & Nutrition 5.1 (1974): 55-142.
4.   Martínez-Anaya, M. Antonia. "Enzymes and bread flavor." Journal of Agricultural and Food Chemistry 44.9 (1996): 2469-2480.
5.   Okada, Sanae, et al. "Identification and characteristics of lactic acid bacteria isolated from sour dough sponges." Bioscience, biotechnology, and biochemistry 56.4 (1992): 572-575.
6.   De Angelis, Maria, and Marco Gobbetti. "Environmental stress responses in Lactobacillus: a review." Proteomics 4.1 (2004): 106-122.
7.   Zhang, Juan, et al. "Glutathione protects Lactobacillus sanfranciscensis against freeze-thawing, freeze-drying, and cold treatment." Applied and environmental microbiology 76.9 (2010): 2989-2996.
8.   Vogel, Rudi F., et al. "Genomic analysis reveals Lactobacillus sanfranciscensis as stable element in traditional sourdoughs." Microb Cell Fact 10.Suppl 1 (2011): S6. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3231932/
9.   Adamberg, Kaarel, et al. "The effect of temperature and pH on the growth of lactic acid bacteria: a pH-auxostat study." International journal of food microbiology 85.1 (2003): 171-183.
10.   Carr, Frank J., Don Chill, and Nino Maida. "The lactic acid bacteria: a literature survey." Critical reviews in microbiology 28.4 (2002): 281-370.
11.   Rogers, R. F., and C. W. Hesseltine. "Microflora of wheat and wheat flour from six areas of the United States." Cereal Chemistry 55 (1978): 889-898. http://naldc.nal.usda.gov/download/25386/PDF
12.   Hamasaki, Yoshikatsu, et al. "Behavior of psychrotrophic lactic acid bacteria isolated from spoiling cooked meat products." Applied and environmental microbiology 69.6 (2003): 3668-3671.
13.   http://www.ncbi.nlm.nih.gov/pmc/articles/PMC161497/
14.   http://aggie-horticulture.tamu.edu/food-technology/food-processing-entrepreneurs/microbiology-of-food/
15.   http://www.classofoods.com/page1_3.html
16.   http://www.fermex.com.au/img/File/CompressedYeast.pdf
17.   An example of a dry yeast spec sheet: http://www.pizzamaking.com/forum/index.php?topic=32478.msg324719#msg324719
18.   Rollán, G., et al. "Update in bread fermentation by lactic acid bacteria." Current Research, Technology and Education Topics in Applied Microbiology and Microbial Biotechnology (2010): 1168-1174. http://www.formatex.info/microbiology2/1168-1174.pdf
19.   Gobbetti, M. "The sourdough microflora: interactions of lactic acid bacteria and yeasts." Trends in Food Science & Technology 9.7 (1998): 267-274.
20.   Aasen, I. M., et al. "Influence of complex nutrients, temperature and pH on bacteriocin production by Lactobacillus sakei CCUG 42687." Applied Microbiology and Biotechnology 53.2 (2000): 159-166.
21.   De Vuyst, Luc, and Patricia Neysens. "The sourdough microflora: biodiversity and metabolic interactions." Trends in Food Science & Technology 16.1 (2005): 43-56.
22.   Gänzle, Michael G., Michaela Ehmann, and Walter P. Hammes. "Modeling of growth of Lactobacillus sanfranciscensis and Candida milleri in response to process parameters of sourdough fermentation." Applied and environmental microbiology 64.7 (1998): 2616-2623. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC106434/
23.   https://uofa.ualberta.ca/integrative-health-institute/scholars/michael-ganzle#sthash.iIa6JTwW.dpuf
24.   https://books.google.com/books?id=GHQBMyTzJ-oC&pg=PA127&lpg=PA127&dq=dough+peptidases&source=bl&ots=vgPphliujx&sig=7APye6hl8OYL6o_-9eKY4eqsCe0&hl=en&sa=X&ved=0ahUKEwi1rebA4cjJAhVCNj4KHRSPAAkQ6AEIJzAE#v=onepage&q=dough%20peptidases&f=false 
25.   Thiele, C., M. G. Gänzle, and R. F. Vogel. "Contribution of sourdough lactobacilli, yeast, and cereal enzymes to the generation of amino acids in dough relevant for bread flavor." Cereal Chemistry 79.1 (2002): 45-51.
26.   Pozo-Bayón, M. A., E. Guichard, and N. Cayot. "Flavor control in baked cereal products." Food Reviews International 22.4 (2006): 335-379.
27.   http://www.worthington-biochem.com/introbiochem/Enzymes.pdf, http://www.classofoods.com/page1_7.HTML
28.   http://www.classofoods.com/page1_7.HTML
29.   http://www.breadlink.co.uk/fallingNumber.pdf
30.   Greenwood, C. T., and E. A. Milne. "Studies on Starch‐Degrading Enzymes. Part VIII A Comparison of α‐Amylases from Different Sources: Their Properties and Action Patterns." Starch‐Stärke 20.5 (1968): 139-150.
31.   Rothfus, JOHN A., and STEPHEN J. Kennel. "Properties of wheat beta-amylase adsorbed on glutenin." Cereal chem 47 (1970): 140-146.

[kiwipete]

Thanks for that excellent work, Craig.

I've been doing RT fermentation since I came on this forum 10 years ago, but now I might just have to focus more precisely on that 65F temp..

Wine fridge here we come..

[jsaras]

I got lost in that middle part.   

[mitchjg]

I got lost in that middle part.   

Is this going to be on the test?

[jvp123]

Thanks for posting Craig.  I'm not very scientific, but intuitively what you say makes a lot of sense to me.   

As you mentioned, I think many people CF (and use a mixer) as opposed to stretch/fold because its just easier for many people's lives/schedules and if it gets them close to what a RT ferment would yield they are fine with it. 

Personally speaking, my 3 day CF is a little better than my 24hr RT (so far), but the techniques are very different.  (I'm working on refining my 24 hour dough when I find the time).

I know you tend to do 2-day RTs (at least on your NPs).  Do you think there is a "sweet spot" in the RT ferment i.e. two days?  Or will a 24 hour RT, lets say, still beat out a 3 day + CF in your opinion?

edit.  Maybe that last question was in those charts/graphs somewhere so sorry if I missed that. 

[Jackitup]

Thanks for posting Craig.  I'm not very scientific, but intuitively what you say makes a lot of sense to me.   

As you mentioned, I think many people CF (and use a mixer) as opposed to stretch/fold because its just easier for many people's lives/schedules and if it gets them close to what a RT ferment would yield they are fine with it. 

Personally speaking, my 3 day CF is a little better than my 24hr RT (so far), but the techniques are very different.  (I'm working on refining my 24 hour dough when I find the time).

I know you tend to do 2-day RTs (at least on your NPs).  Do you think there is a "sweet spot" in the RT ferment i.e. two days?  Or will a 24 hour RT, lets say, still beat out a 3 day + CF in your opinion?

edit.  Maybe that last question was in those charts/graphs somewhere so sorry if I missed that. 

I have used Craig's suggestion from a while back of using a plastic container of ice, in a cooler to keep temps in the 50s for a couple days, well above refrigeration temps. I don't know if he still does that or not but would fit into the model of closer to room temps I would think, and not at cold fridge temps.

Craig, nice job, term paper work there!! Do you still go sub room temp in a cooler towards the 50s?

jon

[TXCraig1]

I have used Craig's suggestion from a while back of using a plastic container of ice, in a cooler to keep temps in the 50s for a couple days, well above refrigeration temps. I don't know if he still does that or not but would fit into the model of closer to room temps I would think, and not at cold fridge temps.

Craig, nice job, term paper work there!! Do you still go sub room temp in a cooler towards the 50s?

jon

Generally around 63 +/- 1F.

[mitchjg]

Is this going to be on the test?

When I was a math teacher, this was usually the first question I was asked when going over material the students found to be difficult. 

On a more serious note, thanks for putting so much effort into this complicated subject that, from time to time, causes confusion or disagreement.  I will need to read through it well more than once or twice to absorb it.

I do have a question about your research.  That is, have you ever looked at the work of a French baker named Gosselin? 
Peter Reinhart raves about this guy in the Bread Baker's Apprentice (I know, but there is no Pizza Baker's Apprentice equivalent beyond this website,  ).  The approach that he takes, which he calls delayed fermentation, is to use cold flour and ice water in mixing the dough and leaving it in the cold fridge overnight.  The yeast goes to sleep very quickly.  But, the amylase enzymes go to work right away.  Then, the wakeup call for the yeast begins the next day and the rest of the dough fermentation and preparation is all done at room temperature.  I know a few of the pizza makers here use ice water but I do not think they do much at room temperature beyond tempering the dough.

Anyway,  I thought you may have interest in this since the fermentation is, pretty much, all at room temperature but there is an initial period of cold temperatures.

Thanks again.....

PS Your timing is perfect.  I have 3 dough balls in the fridge for the last 3 days.  It has been several months since I did a cold ferment for my WFO pizza and I wanted to try it out again.

[Pete-zza]

Mitch,

Gosselin's work is discussed here: http://www.thefreshloaf.com/node/8524/philippe-gosselin039s-pain-%C3%A0-l039ancienne-according-peter-reinhart-interpretted-dmsnyder-m.

Peter

[mitchjg]

Mitch,

Gosselin's work is discussed here: http://www.thefreshloaf.com/node/8524/philippe-gosselin039s-pain-%C3%A0-l039ancienne-according-peter-reinhart-interpretted-dmsnyder-m.

Peter

Thanks, Peter.  dmsnyder is one of the best bakers participating in that forum.  His formula is consistent with the brief writeup in the BBA.  An important item is that the Gosselin approach that dmsnyder followed is more true to Gosselin's instructions.  Most prominent is that there is no yeast added until the flour/water is removed from the fridge.  Reinhart puts it in right away.

[TXCraig1]

When I was a math teacher...

I do have a question about your research.  That is, have you ever looked at the work of a French baker named Gosselin? 
Peter Reinhart raves about this guy in the Bread Baker's Apprentice (I know, but there is no Pizza Baker's Apprentice equivalent beyond this website,  ).  The approach that he takes, which he calls delayed fermentation, is to use cold flour and ice water in mixing the dough and leaving it in the cold fridge overnight.  The yeast goes to sleep very quickly.  But, the amylase enzymes go to work right away. 

No, I have not. My question would be: why even add the yeast in the first phase? Why not do a "pre-hydrolysis?" Like a preferment, but without yeast? You could do it at much higher temperatures and have an order of magnitude more amylase activity.

All of this is math. It's all very predictable. Certainly you can have amylase activity at a temperature where yeast activity is slowed to near stand still, however they amylyase activity is going to be very slow.

[The Dough Doctor]

The last paragraph in #3 should be noted as we have said many time that adding yeast to a sourdough formula can result in the loss of the desired flavor in the finished product. One other thing that was not mentioned is the fact that there are other acids formed during the fermentation process (primarily acetic and propionic) which also play an important part in determining what the finished flavor profile will look like. When fermenting doughs at different temperatures the balance between these acids can be changed resulting in different flavor profiles. This was somewhat alluded to in the discussion of the affect of temperature on LAB. When the LAB is taken out of the equation a different finished flavor profile is the result. What it all seems to boil down to is what flavor profile are you looking for, and if its that of a sourdough (especially a San Francisco sourdough) don't refrigerate the dough. It's also good to probably keep in mind that there are many different strains of LAB each one produces its own unique flavor (just look at what it does in cheese manufacturing and yogurt production) and exhibits different conditions for growth. This is why the study of fermentation is so fascinating.
By the way, bread flavor is the one flavor that flavor scientists have NEVER been able to manufacture, it's that complex.
Tom Lehmann/The Dough Doctor

[schold]

Craig, thank you very much for this highly interesting read. I have enjoyed it very much (with a shot or two or three of ice-cold limoncello). I have a few comments, which I intend to give in a later post, but first a question, if I may. The Swedish genius Svante (Arrhenius) who thankfully invented activation energy - life on earth was terrible before its introduction, with lots of big bangs and fires everywhere - once said that "at first sight, nothing seems more obvious than that everything has a beginning and an end". I wonder what he meant?

Have you contemplated exothermic chemical reactions which are equilibriums as far as flavour in dough is concerned, and what effect would cooling have according to Le Châtelier's principle?

[Icelandr]

A coincidence indeed. Thank you very much for the post!
Yesterday I posted and account of my first try with room temperature ferment with the question about effect on flavour ( http://www.pizzamaking.com/forum/index.php?topic=14506.2875#lastPost ) last night I read your post on the same topic. I love this resource and thank you for the research. I will re-read a few more times to wrap my aging mind around its content.

And at one point a few years ago I was thinking, I would like to try making pizza, - 4 ingredients, what can go wrong . . .... Obsession my wife suggests!

Thank you again

Greg

[TXCraig1]

The last paragraph in #3 should be noted as we have said many time that adding yeast to a sourdough formula can result in the loss of the desired flavor in the finished product. One other thing that was not mentioned is the fact that there are other acids formed during the fermentation process (primarily acetic and propionic) which also play an important part in determining what the finished flavor profile will look like. When fermenting doughs at different temperatures the balance between these acids can be changed resulting in different flavor profiles. This was somewhat alluded to in the discussion of the affect of temperature on LAB. When the LAB is taken out of the equation a different finished flavor profile is the result. What it all seems to boil down to is what flavor profile are you looking for, and if its that of a sourdough (especially a San Francisco sourdough) don't refrigerate the dough. It's also good to probably keep in mind that there are many different strains of LAB each one produces its own unique flavor (just look at what it does in cheese manufacturing and yogurt production) and exhibits different conditions for growth. This is why the study of fermentation is so fascinating.
By the way, bread flavor is the one flavor that flavor scientists have NEVER been able to manufacture, it's that complex.
Tom Lehmann/The Dough Doctor

Tom,

In general, the post is intended to apply to a baker's yeast dough and the LAB native to the flour and yeast. Much of the cited research in the first part is related to SD simply because there is not much published on the spontaneous LAB fermentation in a baker's yeast dough.

[TXCraig1]

Craig, thank you very much for this highly interesting read. I have enjoyed it very much (with a shot or two or three of ice-cold limoncello). I have a few comments, which I intend to give in a later post, but first a question, if I may. The Swedish genius Svante (Arrhenius) who thankfully invented activation energy - life on earth was terrible before its introduction, with lots of big bangs and fires everywhere - once said that "at first sight, nothing seems more obvious than that everything has a beginning and an end". I wonder what he meant?

You wonder what he meant? I suspect you have a very good idea what he meant as it is quite clear form the Nobel lecture he began with the quote. Perhaps you are implying that despite it seeming so obvious that RT is better than CF maybe it isn't? And the pizza philospohers of today who, for purely speculative reasons, believe CF to be the better method will eventually be proven right? I suppose it's possible. Notwithstanding, as tough as it is - being a Swede myself - I am going to have to side with the English philosopher William of Ockham this time.   

Have you contemplated exothermic chemical reactions which are equilibriums as far as flavour in dough is concerned, and what effect would cooling have according to Le Châtelier's principle?

What is the largest increase in temperature you have ever experienced to occur in a dough during fermentation. What is the largest you have ever heard of for that matter? Precisely define what you mean by "cooling."

[TXCraig1]

Have you contemplated exothermic chemical reactions which are equilibriums as far as flavour in dough is concerned, and what effect would cooling have according to Le Châtelier's principle?

If you are suggesting that per Le Châtelier's principle, lowering the temperature would favor the right side of the fermentation equation, do you have any reason to believe that any resulting increase in production would not be more than offset by the reduction in reaction rate?

[fagilia]

Thank you for the interpret of the studys.
I have Been Reading these for some Times and missed many points.
Have we ever discused textural effekts due to CF? How it affekts handling , spring, etc.
Good work bravo

[schold]

Perhaps you are implying that despite it seeming so obvious that RT is better than CF maybe it isn't?

No, I think RT fermentation is better, but CF is more practical. Personally, I do both - all depending.

Notwithstanding, as tough as it is - being a Swede myself - I am going to have to side with the English philosopher William of Ockham this time.   

Are you a Swedish-American or an expat? Ockham, that's the famous guy who cut himself during shaving, right?

What is the largest increase in temperature you have ever experienced to occur in a dough during fermentation. What is the largest you have ever heard of for that matter? Precisely define what you mean by "cooling."

Craig, I come here to be educated, I ask one question and get three back? I have never measured the change in dough temperature during fermentation. The largest I've heard of? I don't think I have ever heard it mentioned. By cooling I meant lowering the temperature.

If you are suggesting that per Le Châtelier's principle, lowering the temperature would favor the right side of the fermentation equation, do you have any reason to believe that any resulting increase in production would not be more than offset by the reduction in reaction rate?

I'm not sure I understand what you mean here, but no doubt we both agree that if a dough is placed in the refrigerator instead of on the kitchen counter, the fermentation process will be slower. What I asked was if you have contemplated exothermic chemical reactions which are equilibriums as far as flavour in dough is concerned, and what effect would cooling have according to Le Châtelier's principle?

Another question, if I may, since this is going so well. You mentioned aroma, and this got me thinking. Let's say you got a bottle of pure β-damascenone from me, and I told you that this stuff smelled very nice. If you then opened the bottle, spilling a drop in the process, and took a good sniff, I doubt very much that you would be impressed. The next day, however, the whole room would smell beautiful of roses. This is a common example of how olfactory perception can be a function of concentration in a somewhat counterintuitive way. Is it possible that some of the odorants could be judged more desirable when present in a smaller quantity in the dough, i.e. where less is more?

[TXCraig1]

No, I think RT fermentation is better, but CF is more practical. Personally, I do both - all depending.

This is exactly my point.

Are you a Swedish-American or an expat? Ockham, that's the famous guy who cut himself during shaving, right?

2nd generation Swedish-American.

Craig, I come here to be educated, I ask one question and get three back? I have never measured the change in dough temperature during fermentation. The largest I've heard of? I don't think I have ever heard it mentioned. By cooling I meant lowering the temperature.

I suspect the maximum change in temperature from exothermic reactions (outside of a large-scale manufacturing environments anyway) is perhaps a couple degrees K? At this point in the discussion, I'm painting with a slightly broader brush.

I asked about a precise definition of "cool" as it could refer to the couple degrees described above, or it could be intended to mean cool to refrigeration temperatures which of course would be a very different thing.

I'm not sure I understand what you mean here, but no doubt we both agree that if a dough is placed in the refrigerator instead of on the kitchen counter, the fermentation process will be slower. What I asked was if you have contemplated exothermic chemical reactions which are equilibriums as far as flavour in dough is concerned, and what effect would cooling have according to Le Châtelier's principle?

Are they? Is amylase cleaving sugar off starch molecules or the action of carboxypeptidase on wheat proteins exothermic? It also should be pointed out that flavor in dough and flavor in the baked product are not the same thing.

Another question, if I may, since this is going so well. You mentioned aroma, and this got me thinking. Let's say you got a bottle of pure β-damascenone from me, and I told you that this stuff smelled very nice. If you then opened the bottle, spilling a drop in the process, and took a good sniff, I doubt very much that you would be impressed. The next day, however, the whole room would smell beautiful of roses. This is a common example of how olfactory perception can be a function of concentration in a somewhat counterintuitive way. Is it possible that some of the odorants could be judged more desirable when present in a smaller quantity in the dough, i.e. where less is more?

I suppose it's possible, however I'm not sure what would suggest such is the case here? Lacking any other information to the contrary, it would seem equally more likely that in this case less is zero (or close enough to be zero). I think our shaving friend is in play in this question as well.