Am I to read between the lines correctly then that in a sourdough system, water bound in a starter is not available to the rest of the dough for microbial activity? So basically, that it’s not about total hydration but total hydration minus starter hydration?
Great question. Admittedly, I too was under the impression there exists two fractions of water in fabricated food systems: one that is "bound" (or "immobile" or "unavailable"), another that is "unbound" (or "mobile" or "free"). Food science and technology literature emptily clings onto this notion, despite criticism from scientists in other fields (physics, biophysics, and materials/polymer sciences). There are food-science symposia stretching back to the early 1980s where physicists warn about adopting such a view because it's both wrong and dangerous. (The latter especially so: Relying on water activity alone as a measure of microbial stability has led to many food recalls.)
Water molecules are small, labile and extremely mobile. They interact with other molecules and macromolecules via hydrogen bonding. No hydrogen bond (H-bond) is permanent, to the point that the "average" exchange lifetime of any given hydrogen bond lasts less than a millisecond. In pure water, H-bonds extend down to the pico- or femtosecond scale (one trillionth to one quadrillionth of a second).
This means H-bonds are in a continual process of forming and breaking. So, in a liquid every single water molecule is "free." That is, if you could trace the path of a single water molecule picked at random in a liquid system (and doughs and batters are technically classified in materials science as liquid systems), every single position a water molecule could occupy is equally probable during the lifetime of that system.
However, there is an emergent (one could say "statistical") property when the system is observed at larger scales: there appears to be concentration gradients of water, similar to gasses in our atmosphere. Atmospheric density increases the closer to earth you get, to the point that if you zoom out enough, distinct layers or "rings" form, each with their own internal, consistent properties (density, composition, etc.) that remain stable through our timescale of observation. No single gas molecule in our atmosphere is ever "bound" or "restricted" to a particular location, yet clearly-defined "layers" with different chemical compositions appear. So it is for water in liquid systems. I will explore this concept in much greater detail in future posts, as well as fully answering your question (a reply here wouldn't do it justice).
Last, I've long had a suspicion that a greater and non-negligible percentage of water in liquid starters (relative to stiffer starters) is "unavailable" for the hydration of the final dough. I came to this hypothesis purely through observation as a working baker. Turns out my hunch was indeed correct. Why? Because there is an infinite number of dynamic equilibria when partitioning water in poroelastic systems that are tuneable via process (e.g., mixing parameters) or formulation (e.g., ingredient quantity or quality). However, these distributions cannot be reliably predicted ahead of time.
Am I to read between the lines correctly then that in a sourdough system, water bound in a starter is not available to the rest of the dough for microbial activity? So basically, that it’s not about total hydration but total hydration minus starter hydration?
Great question. Admittedly, I too was under the impression there exists two fractions of water in fabricated food systems: one that is "bound" (or "immobile" or "unavailable"), another that is "unbound" (or "mobile" or "free"). Food science and technology literature emptily clings onto this notion, despite criticism from scientists in other fields (physics, biophysics, and materials/polymer sciences). There are food-science symposia stretching back to the early 1980s where physicists warn about adopting such a view because it's both wrong and dangerous. (The latter especially so: Relying on water activity alone as a measure of microbial stability has led to many food recalls.)
Water molecules are small, labile and extremely mobile. They interact with other molecules and macromolecules via hydrogen bonding. No hydrogen bond (H-bond) is permanent, to the point that the "average" exchange lifetime of any given hydrogen bond lasts less than a millisecond. In pure water, H-bonds extend down to the pico- or femtosecond scale (one trillionth to one quadrillionth of a second).
This means H-bonds are in a continual process of forming and breaking. So, in a liquid every single water molecule is "free." That is, if you could trace the path of a single water molecule picked at random in a liquid system (and doughs and batters are technically classified in materials science as liquid systems), every single position a water molecule could occupy is equally probable during the lifetime of that system.
However, there is an emergent (one could say "statistical") property when the system is observed at larger scales: there appears to be concentration gradients of water, similar to gasses in our atmosphere. Atmospheric density increases the closer to earth you get, to the point that if you zoom out enough, distinct layers or "rings" form, each with their own internal, consistent properties (density, composition, etc.) that remain stable through our timescale of observation. No single gas molecule in our atmosphere is ever "bound" or "restricted" to a particular location, yet clearly-defined "layers" with different chemical compositions appear. So it is for water in liquid systems. I will explore this concept in much greater detail in future posts, as well as fully answering your question (a reply here wouldn't do it justice).
Last, I've long had a suspicion that a greater and non-negligible percentage of water in liquid starters (relative to stiffer starters) is "unavailable" for the hydration of the final dough. I came to this hypothesis purely through observation as a working baker. Turns out my hunch was indeed correct. Why? Because there is an infinite number of dynamic equilibria when partitioning water in poroelastic systems that are tuneable via process (e.g., mixing parameters) or formulation (e.g., ingredient quantity or quality). However, these distributions cannot be reliably predicted ahead of time.
I have observed the same anecdotally as a home baker and am striving to understand it. Looking forward to your posts on the matter!