What the mesh size is
A hydrogel is specified by its mesh size more often than by its modulus. The mesh size ξ is the average distance between consecutive crosslinks along the swollen network, and it is the length that decides transport: a solute much smaller than ξ diffuses almost as it would in water, one comparable to ξ is hindered, and one larger than ξ does not get through at all. Drug release rate, protein exclusion, nutrient supply to encapsulated cells and the cut-off of a separation membrane are all the same number read four ways.
It is not measured directly. It is inferred, in two steps, from a swelling experiment you can do with a balance. First the swelling gives the molar mass between crosslinks; then the chain statistics turn that into an end-to-end distance, which the swelling stretches:
The second factor is the unperturbed end-to-end distance of a strand of N backbone bonds, and the first is the isotropic expansion that swelling applies to it. Both halves of that need real chain data, which is why this page reads the bond length, the bonds per repeat unit and the characteristic ratio out of the site's chain dimensions dataset rather than asking you to type them.
Mesh size from a swelling measurement
Sets the correction for chain ends, which are network defects. Leave it blank or very large if the chains were long enough that end effects do not matter.
Blot the surface, do not squeeze. Extract the sol fraction before drying, or the dry mass includes polymer that was never part of the network and every number below reads low.
1 for a gel crosslinked in bulk and swollen afterwards. For one crosslinked in solution, which is how most hydrogels are made, this is the polymer volume fraction in the reaction mixture. The page opens on 0.10, a gel formed at 10% polymer. Getting this wrong is the commonest error here, and it is silent: the answer stays plausible.
Belongs to the polymer and water together, not to the polymer. See Flory–Huggins. Above 0.5 water is a poor solvent and the gel will not swell much whatever its crosslink density.
A hydrodynamic radius, from Stokes–Einstein on a measured free diffusion coefficient. Leave it at 0 to skip the transport estimate.
Why the relaxed state matters
Flory and Rehner wrote their swelling relation for a network formed in bulk and swollen afterwards, so the dry, unswollen network is the reference state where the strands are unperturbed. A hydrogel is usually not made that way. It is crosslinked in water, already swollen, and the strands are stretched at the moment they are tied together. Treating that as the reference state counts the stretch twice.
Peppas and Merrill's form carries v2,r, the polymer volume fraction in the relaxed state, and reduces to Flory–Rehner exactly when v2,r = 1. The 2⁄Mₙ term is the chain-end correction: every chain has two ends that hang off the network and carry no load, so a network built from short precursor chains is looser than its crosslink count suggests. Set v2,r to 1 on a gel that was made at 10% polymer and the mesh size comes out wrong in a direction you will not notice, because it is still a plausible number.
What this cannot tell you
ξ is an average, and transport is decided by the small end. A real network has a distribution of strand lengths, and a solute finds the largest connected path rather than the mean one. Two gels with the same ξ and different heterogeneity do not release at the same rate, and the more heterogeneous one leaks first.
The chain-end correction assumes you know Mₙ. It is the molar mass of the precursor before crosslinking, not of the network, which has none. If the gel was made by copolymerising monomer with a crosslinker rather than by tying together preformed chains, there is no precursor Mₙ to use and the term should be dropped.
χ carries the same weight here as on the crosslink density page. It sits against a logarithm, and moving it from 0.45 to 0.55 can change Mc by a factor of two with nothing about the sample changing. If χ is a guess, ξ is an order of magnitude.
Ionic gels are outside this model. A polyelectrolyte network swells because counterions are trapped inside it and pull water in osmotically, and that term is not in the expression above at all. For a charged hydrogel this page gives a lower bound on the swelling and therefore the wrong mesh size; the ionic contribution has to be added separately, and it moves with pH and ionic strength.
Nothing here counts loops. A strand that starts and ends at the same junction consumed two crosslink sites and holds no network, exactly as on the network formation page, and hydrogels crosslinked dilute are where loops are worst.
Related tools
- Crosslink density. The same swelling measurement read as Mc for an elastomer in an organic solvent, plus the modulus route.
- Chain dimensions. Where the characteristic ratio and bond geometry on this page come from, and what they mean.
- Flory–Huggins. What χ is, and why 0.5 is the line between a solvent and a non-solvent.
- Network formation. Whether the network forms at all, and how much of it is elastically effective.