Hi Sonic,
Your question is a great one and you are correct that there is a difference between aging and mechanical durability. Latex ages by oxidation through exposure to oxygen, ozone, ultraviolet light, and through exposure to various metal ions (especially copper) and various chemical compounds (see here for a list).
Latex also breaks down through mechanical processes which involve shear forces, elongation and tensile strength, tear strength, abrasion, compression set, and others. Some of these are internal processes (inside the rubber slab itself) and some are through interaction with the rubber with whatever it is in contact with.
The answer to how each process affects each different formulation of rubber depends on the composition of the rubber itself and the many things that are added to it (such as vulcanizing agents, accelerators, antioxidants, fillers etc), and even to the source of latex itself (particle size, dispersion, and agglomeration, method of concentration, ratio of styrene to butadiene and many other factors). Latex chemistry and compounding methods is still an evolving science and there is a huge amount of different “formulae” that are used. It is an incredibly technical and complex question as indicated in just a few of the links I have included at the end of this post.
While I am also not a foam chemist … I have read a great deal of detailed information on the subject and spent a great deal of time talking with experts in the industry and I agree with you that most of the online information provided by many sources is slanted towards what they themselves sell or simply a repeat of incorrect information.
My personal belief based on a great deal of research and the “preponderance of the evidence” is as follows …
In general … NR (Natural Rubber) latex is superior to SBR rubber in most areas with a few exceptions. To some degree this depends on the type of SBR that it is compared to and on the additives that are used in compounding both. The exceptions are heat aging and wear resistance where SBR can have advantages. NR also tends to soften more with oxidative aging while SBR tends to harden with oxidative aging because the oxidative processes affect their polymer structure differently. Both will end up hard and brittle when fully oxidized (which if the latex is protected will take a very long time).
With Dunlop synthetic latex may lose height (develop impressions) more than 100% natural while natural rubber may become softer over time more than synthetic rubber.
If you include the effect of a cover or ticking (which all latex should have) which protects the latex from much of the oxidation and chemical breakdown (aging) and other substances that can break down the latex (see here and here for examples) … then in practical “mattress” terms it boils down to mechanical breakdown and either loss of height or loss of firmness (durability).
This is where the additional difference between the two different types of processing methods comes into play as they create two different cell structures. Dunlop is less consistent in its cell structure (has a wider variety of cell size, shape, and structure) which along with its particle dispersion is what accounts for its greater compression modulus. It also uses more latex in the foam (a higher ratio of rubber to air) and is denser.
Talalay has a more consistent and even and stronger cell structure because even though there is less latex overall … the cell walls or “struts” are thicker and more consistent.
What this means is that with the Dunlop method … NR would generally be more durable overall than a blend or synthetic (SBR) latex although there may be value in using a blend because it is less expensive and still a very durable material compared to other foams and in some cases a blend can improve the properties of the foam. Unless budget considerations don’t allow it I would tend to choose a 50/50 or higher blend in molded Dunlop and would lean towards higher percentages of NR because of the greater elasticity and higher performance but again I would also keep in mind that all latex is a good quality material compared to other types of foam.
For Talalay … the higher ILD’s use proportionally more latex (are denser) so the cell walls are thicker and more elastic. This in combination with the fact that higher ILD’s are usually used in the core of a mattress where there is less mechanical stress on the material means that it is probably a tossup and because of its better performance I would say the NR has better overall properties than a blend in higher ILD’s but of course also more expensive and the practical differences between them may not justify the increased cost.
In the lower talalay ILD’s … partly because of the difficulty of creating a homogenous mixture and partly because of the thinner cell walls … and because the lower ILD’s are usually used in the comfort layers which are subject to greater mechanical stress … then the shear forces inside the mattress and the higher elasticity of the NR (allowing the “thinner” material to stretch beyond its tensile limit) may work against it and it could break down faster than a blend.
The testing of the producers of Talalay and the experience of many manufacturers seems to confirm this although Radium uses a different curing past in their all natural talalay and they have told me that their testing indicates that their 100% natural Talalay even in lower ILD’s has a similar durability to their blend.
I should also mention that latex compounding is a continuously evolving science and there are newer formulas of latex that may change some of the information here as their durability is confirmed over time but I am one of the ones that takes more of a wait and see approach because historical experience can sometimes be different than theory or lab testing. This is particularly true of some of the continuous pour synthetic or blended Dunlop products which appear to be quite promising and may be closely comparable to latex with higher natural rubber content in terms of durability.
There are variables that are not included in this simple “translation” of many hours of reading and of course my thoughts and opinions can only take into account my less than complete understanding of some very complex theories and information but I believe in essence and in real life terms they are at least very close to “correct”.
A few of the references I have “consulted” (among many more) are …
Link #1
Link #2
Link #3
Link #4
Link #5
Link #6
Link #7
Phoenix