Abstract
Yes-and the direction of that difference may not be what you expect. Proofing time is not a fixed background parameter in emulsifier evaluation; it is a variable that can rewrite the conclusion. The performance gap between SSL and DATEM under shifting proofing time cannot be reduced to "SSL shortens time while DATEM lengthens it," nor can it be captured by a single label of "promoting" or "retarding." A more accurate framing is this: both alter the proofing process, but they do so through different pathways, at different stages, and with different logics of tolerance to over-proofing. Proofing time must therefore enter the experimental design as an active variable, not as a single time point at which one emulsifier is compared against the other.
A Contradiction Worth Facing First
Two seemingly opposed observations coexist in the literature on emulsifiers and proofing time.
In a classic 1981 study, Tsen and Weber found that both DATEM (reported as DATA) and SSL shortened the proof time of yeasted doughs, and that both significantly promoted yeast gas production in flour-water systems. This finding was echoed in later work on CSL and DATEM: nonfrozen doughs containing these surface-active agents showed significantly higher gassing power than the control.
A different line of evidence paints another picture. Gómez et al. (2004), comparing a range of emulsifiers, reported that the presence of emulsifier retarded dough proofing, so that longer proofing times were required when emulsifiers were used. More importantly, the positive effect of emulsifiers on bread volume was only observed under long proofing times.
These conclusions do not necessarily cancel each other out. Together they point to a methodological fact: whether proofing time appears "shortened" or "lengthened" depends on which stage is being measured and what endpoint defines completion. An increase in gassing power can accelerate early dough expansion, showing up as a shorter time to reach a target height. But gluten network reorganization and stabilization may be proceeding simultaneously, so that the volume and texture benefits only emerge on a longer time scale. If SSL and DATEM are compared at a single fixed proofing time, what is observed may be only one phase of their action rather than the whole.
SSL and DATEM Follow Different Pathways, and Shifting Proofing Time Magnifies That Difference
Both are dough-strengthening emulsifiers, but their binding preferences in dough are measurably different. Lipid-binding work by Collar et al. showed that SSL preferentially binds to the interior of starch granules and binds loosely to gluten, while increasing the pool of free lipids; DATEM tends to associate with the outer regions of starch granules and with gluten proteins. This difference has consequences when proofing time is stretched or compressed.
DATEM's strength lies in tolerance. Because its interaction with gluten proteins is more direct, DATEM reinforces the gluten network so that the dough is less likely to collapse when proofing is moderately extended, exhibiting greater tolerance to over-proofing. It stabilizes the foam structure at the gas–liquid interface and prevents bubble coalescence, maintaining structural integrity while fermentation continues and gas pressure accumulates. In other words, DATEM's value may not lie in "making proofing faster" but in "making proofing safer"-providing a margin of error when the process window is stretched by equipment, environment, or handling variability.
SSL's distinctive position is on the starch side. Its preference for the starch granule interior gives SSL a functional role that DATEM alone does not provide: delaying starch retrogradation and maintaining crumb softness over time. In a scenario where proofing time shifts, this means SSL's effects may manifest more clearly after baking: altered proofing changes the state of starch gelatinization and the subsequent staling pathway, and SSL intervenes in that pathway to modulate final texture. SSL's contribution to dough stability is well established, but its binding to gluten is comparatively "shallower," so its route to protecting against over-proofing may not be the same as DATEM's.
One detail is worth noting: in a test of recovery after mechanical abuse, the 0.5% DATEM treatment performed better than the SSL treatment, suggesting that under non-ideal "process shock" conditions, DATEM's structural protection may be more pronounced. This does not put SSL at a disadvantage in proofing-time management-if the endpoint includes shelf-life softness, SSL's starch-side action enters the scoring.
An Evaluation Framework: Turn Proofing Time from a Fixed Condition into a Test Factor
Given these differences, the evaluation of SSL versus DATEM should not stop at a single comparison at one proofing time. A more effective approach is a layered framework with proofing time on the horizontal axis and endpoint indicators on the vertical axis.
Layer one: identify which kind of "proofing time change" you are dealing with. Is the change process-driven (equipment delays, ambient temperature fluctuation, proofing after frozen dough thawing) or formulation-driven (deliberately extending proofing to pursue greater volume)? The former is about tolerance; the latter is about efficiency and volume gain. The weighting of SSL versus DATEM is entirely different depending on which question is being asked.
Layer two: sample at least three proofing-time levels. Do not compare only at a "standard time" and one increment above it. Cover a short, standard, and extended gradient, and at each point measure gassing power, dough expansion, and post-bake specific volume and crumb hardness. The findings of Gómez et al. already suggest that an emulsifier's positive volume effect may be invisible at short times and only emerge at long times; using a single time point is equivalent to discarding information by design. A rheofermentometer, which records dough development and gas-release curves simultaneously, is better suited than a single endpoint to distinguishing the stages at which SSL and DATEM act.
Layer three: treat the post-over-proofing state as an endpoint in its own right. If there is a risk that proofing time cannot be precisely controlled in production, then observing dough collapse, volume loss after baking, and internal structural non-uniformity under extended-time treatments is more decision-relevant than observing volume at the "optimal" proofing time. DATEM is expected to be stronger on this dimension, but that expectation should be verified in your own formulation and flour system rather than adopted directly from the literature.
Layer four: do not omit the starch-side endpoint from independent scoring. If the product has a shelf-life softness requirement, SSL's value will not be fully captured by proofing curves or gas-production data. Running crumb hardness over time (a staling curve) as a parallel evaluation line, crossed with the proofing-time gradient, is necessary to judge whether SSL contributes something in your system that DATEM cannot replace. In some formulations the two are not mutually exclusive options; they can separately cover "structural tolerance during fermentation" and "textural retention during storage."
In summary: when proofing time shifts, SSL and DATEM do behave differently-but that difference is not a ranking of "which is faster." It is a divergence in stage of action and center of functional gravity. The evaluation task is not to determine "which emulsifier is better under changing proofing time," but to answer: where in your process does proofing-time variability arise, how large a time window are you willing to trade for structural stability or volume gain, and does shelf life count in the final assessment. Once proofing time is treated as a variable to be actively scanned rather than a condition to be held fixed, the SSL–DATEM difference stops being a line in the literature and becomes information your formulation decision can actually use.
