Sourdough Fermentation Duration and Its Effect on Pizza Dough Extensibility

1. Introduction

Among the many variables governing pizza quality, fermentation duration is simultaneously one of the most influential and one of the most practically constrained. Professional pizzaioli are well acquainted with the trade-off between the extended fermentation times that maximize dough flavor and extensibility and the operational reality of needing to produce pizza on a schedule. The internet, for its part, has developed a robust amateur literature on "72-hour dough," "96-hour cold ferment," and related approaches, generating considerable enthusiasm and, our laboratory colleagues report, a fair amount of overconfident bread-making. The peer-reviewed literature has largely failed to keep pace with either the professionals or the internet.

Sourdough fermentation involves two primary microbial processes: alcoholic fermentation by Saccharomyces cerevisiae and related wild yeast species, and lactic acid fermentation by heterofermentative Lactobacillus species. Both contribute to flavor compound production and to the enzymatic modification of the gluten network that governs dough extensibility. Proteolytic enzymes released during fermentation partially cleave gluten peptide chains, reducing elasticity and increasing extensibility — qualities that practitioners assess by the ease with which dough can be stretched without tearing.

The relationship between fermentation duration and gluten extensibility has been studied in bread dough (Schmelz et al., 2017; Romano & Saucington, 2019), but pizza dough presents distinct conditions: higher hydration, no added fat or sugar in the Neapolitan tradition, and the expectation of hand-stretching to a thin sheet rather than shaping in a mold. Whether the fermentation-extensibility relationship characterized for bread dough generalizes to pizza dough was, until this study, unknown.

2. Materials & Methods

Sourdough Starter. A PRI-maintained sourdough starter (designated PRI-SD-2017-A, a rye-wheat composite culture maintained since 2017) was used for all experiments. Microbial profiling via 16S rRNA sequencing revealed a stable community dominated by *Lactobacillus sanfranciscensis* (now reclassified as *Fructilactobacillus sanfranciscensis*, a name that the author refuses to use on the grounds that it is considerably less charming) at 62% relative abundance, with secondary populations of *L. plantarum* and *Saccharomyces cerevisiae*. The starter was refreshed at 1:5:5 (starter:flour:water) every 12 hours for seven days before experimental use to ensure microbial stability.

Dough Preparation. A Neapolitan-style pizza dough formulation was used (Type 00 flour, 65% hydration, 2% salt, 20% sourdough starter inoculation relative to flour weight, no added yeast). All ingredients were combined by hand in a standardized sequence and kneaded for exactly 10 minutes using the PRI Planetary Mixer (Model PM-6) fitted with a dough hook at Speed 2. Dough was divided into 280g balls immediately after mixing, individually bagged, and placed in refrigerated proofing chambers at 4°C ± 0.5°C.

Fermentation Duration Conditions. Six fermentation durations were evaluated: 24, 36, 48, 60, 72, and 96 hours, each representing a distinct cohort of 24 dough balls. Following refrigerated fermentation, dough balls were tempered at 20°C ± 1°C for 2 hours before extensibility testing.

Extensibility Measurement. Dough extensibility was assessed using two methods: (1) the Kieffer Dough and Gluten Extensibility Rig (Stable Micro Systems), which measures maximum extension length (cm) and force at rupture (N) using a standardized geometry, and (2) hand-stretch evaluation by a panel of five trained pizzaioli (recruited from New Haven pizzerias under a PRI consultant agreement and paid at a rate they described as "acceptable, considering the commute"). The pizzaioli rated each dough on a 10-point extensibility scale and a 5-point "hand feel" scale assessing dough responsiveness, springback, and what one evaluator called "forgiving nature."

Flavor Compound Analysis. Following baking (485°C, 90 seconds), crust volatiles were captured by Headspace Solid Phase Microextraction (HS-SPME) and analyzed by GC-MS (Agilent 7890B/5977B). Target compounds included lactic acid, acetic acid, ethanol, and 12 fermentation-derived aldehydes and esters known to contribute to sourdough flavor complexity.

This study was approved under PRI IRB #IRB-2020-PZZ-002. No human subjects consumed experimental dough in its uncooked state, despite repeated requests from two of the five pizzaioli evaluators.

3. Results

Dough extensibility (maximum extension before rupture, Kieffer rig) showed a statistically significant non-linear relationship with fermentation duration (F(5, 138) = 412.3, p < 0.0001), with a peak at 72 hours (mean: 34.2 ± 1.8cm) that was significantly greater than all other durations (post-hoc Tukey HSD, all p < 0.001). Extensibility at 24 hours (18.4 ± 2.1cm) was significantly lower than at 48 hours (26.7 ± 1.6cm), which was significantly lower than 72 hours. Extensibility at 96 hours (28.9 ± 2.3cm) was intermediate — significantly lower than 72 hours (p = 0.002) but significantly higher than 24 hours (p < 0.0001) — indicating the onset of over-fermentation gluten degradation.

The pizzaioli hand-stretch ratings correlated strongly with Kieffer rig extensibility (r = 0.91, p < 0.0001), validating the instrumental measure against practitioner judgment. One pizzaiolo, who preferred to be identified only as "Carmine," provided extended verbal commentary during each evaluation that was not requested but proved analytically useful; his remarks are excerpted in Supplementary Materials Table S3.

GC-MS flavor profiling revealed monotonically increasing concentrations of lactic acid, acetic acid, and key fermentation esters (ethyl acetate, isoamyl acetate) with fermentation duration, without the non-linearity observed in extensibility. Total volatile flavor compound concentration at 96 hours was 2.4-fold higher than at 24 hours (p < 0.0001). This creates a practical trade-off: 72-hour fermentation optimizes extensibility, while 96-hour fermentation maximizes flavor compound production at the cost of slightly reduced extensibility. A consumer preference study (n=200) conducted as a secondary outcome confirmed that 72-hour dough received the highest overall pizza ratings (mean CSI: 82.4), while 96-hour dough received marginally lower ratings despite higher flavor compound concentrations, attributed by the research team to the extensibility-related difficulty of achieving uniform thickness during hand-stretching.

4. Discussion

The 72-hour optimum for dough extensibility confirms the practical wisdom of the pizza-making community and provides a mechanistic explanation for why this duration has emerged as the informal industry benchmark: it marks the point at which proteolytic gluten weakening has proceeded far enough to maximize extensibility while not yet causing the structural degradation that reduces extensibility at 96 hours. This is gratifying confirmation that bakers who arrived at 72-hour fermentation empirically were, in fact, right — a finding that our laboratory colleagues receive with mixed feelings, given that we spent 14 months proving it.

The monotonic increase in flavor compound production with fermentation time, uncoupled from the non-linear extensibility trend, presents formulators with a genuine optimization problem. A 72-hour dough represents the best single-point solution, but a practitioner willing to manage the reduced extensibility of a 96-hour dough would obtain substantially greater flavor complexity. The data support a "sweet spot" interpretation at 72 hours for operations prioritizing consistency, and a "specialist" position at 96 hours for operations prioritizing flavor maxima at the cost of slightly increased stretching difficulty.

The role of our proprietary PRI-SD-2017-A starter in these results should be acknowledged as a potential limitation. The specific microbial community of a sourdough starter determines the balance of proteolytic and acidifying activities and therefore the shape of the extensibility-time curve. Replication with different starters — particularly those dominated by homofermentative rather than heterofermentative Lactobacillus species — may yield different optima. We plan to address this in a follow-up study, assuming the starter cooperates. It has shown signs of temperament.

5. Conclusion

Sourdough fermentation duration has a statistically significant, non-linear effect on pizza dough extensibility, peaking at 72 hours under standard refrigerated conditions. The 72-hour optimum is consistent with practitioner experience and is now supported by instrumental and sensory data. Flavor compound production increases monotonically with fermentation duration, creating a trade-off between extensibility and flavor that practitioners should navigate according to their operational priorities. The 72-hour fermentation is recommended as a standard reference condition for pizza dough research.

Acknowledgments

The author thanks the five pizzaiolo evaluators for their time, their expertise, and their frankness. Carmine, in particular, is thanked for his extended commentary on what he termed the '24-hour abomination,' which, while unprompted and somewhat difficult to transcribe, was analytically valuable. The PRI Microbiology Core Facility is thanked for sourdough starter maintenance and for 16S rRNA sequencing support. Funding was provided by the PRI General Research Fund (PRI-GRF-2019-22). The author discloses that she has been maintaining a personal sourdough starter since 2014 and considers herself appropriately expert in its care.

References

1. [1]

   Cheeseberg, M. (2021). Moisture migration in frozen pizza (PRI Technical Report PRI-TR-2021-004). Pizza Research Institute.

2. [2]

   Marinara, T., & Saucington, A. (2021). The grease gradient: A study of lipid migration in commercial pizza substrates. Journal of Applied Pizzology, 20(2), 88–103. https://doi.org/10.1883/jap.2021.029

3. [3]

   Romano, G., & Saucington, A. (2019). Gluten network relaxation during extended fermentation: A rheometric study in pizza versus bread dough. Food Structure Quarterly, 7(4), 211–228.

4. [4]

   Romano, G., Saucington, A., & Crustworthy, F. (2024). Gluten network architecture in long-fermentation pizza doughs: A Cryo-EM study. Journal of Applied Pizzology, 23(1), 35–52. https://doi.org/10.1883/jap.2024.003

5. [5]

   Saucington, A., & Napolitano, F. (2023). Thermodynamic properties of mozzarella phase transitions during pizza baking. Journal of Applied Pizzology, 22(1), 33–51. https://doi.org/10.1883/jap.2023.008

6. [6]

   Schmelz, N., Forno, L., & de la Salsa, R. (2017). Proteolytic modification of gluten networks during sourdough fermentation in bread dough: A comparative species study. Annals of Culinary Chemistry, 11(2), 88–103.