PDMS Network Structure-Property Relationships: Influence of Molecular Architecture on Mechanical and Wetting Propertie

Poly(dimethylsiloxane) (PDMS) is one of the most common elastomers, with applications ranging from sealants and marine-antifouling coatings to medical devices and absorbents for water treatment. Fundamental understanding of how liquids spread on the surface of and absorb into and leach out of PDMS networks is of critical importance for the design and use in another application - microfluidic devices. The growing use of PDMS in microfluidic devices raises the concern that some researchers may use this material without fully understanding all of its advantages, drawbacks, and intricacies. The primary goal of this Ph.D. dissertation is to elucidate PDMS network molecular structure to macroscopic property relationships and to demonstrate how molecular architecture can alter dynamic mechanical and wetting characteristics.;We prepare PDMS materials by using vinyl/ tetrakis(dimethylsiloxy)silane (TDSS) and silanol/ tetraethylorthosilicate (TEOS) combinations of PDMS end-groups and crosslinkers as two model systems. Under constant curing conditions, we systematically study the effects of polymer molecular weight, loading of crosslinker, and end-group chemical functionality on the extent of gelation and the dynamic mechanical and water wetting properties of end-linked PDMS networks. The extent of the gelation reaction is determined using the Soxhlet extraction to quantify the amount of material that did and did not participate in the crosslinking reactions, termed the gel and sol fractions, respectively. We use the Miller-Macosko model in conjunction with the gel fraction and precise chemical composition (i.e., stoichiometric ratio and molecular weight) to determine the fractions of elastic and pendant material, the molecular weight between chemical crosslinks, and the average effective functionality of the crosslinker molecule.;Based on dynamic mechanical testing, we find that the maximum storage moduli are achieved at optimal stoichiometric conditions in the vinyl/TDSS and commercial PDMS-based Sylgard 184 composite, but only keep improving with additional crosslinker in the silanol/TEOS systems due to in situ TEOS aggregation. We relate molecular network topology to mechanical properties using outputs from the Miller-Macosko model in the vinyl/TDSS system. The elastic fraction and storage modulus correlate well, as do the pendant fraction and the loss tangent, demonstrating the importance of each fraction in bulk mechanical properties. By studying the dynamic behavior of water droplets wetting PDMS substrates, we observe non-linear wetting behaviors that are markedly different from linear behaviors seen on glassy polymer substrates. The non-linear behavior is only observed prior to extraction, while after extraction, both systems demonstrate behavior similar to glassy polymers. This reveals the dramatic role small amounts of uncrosslinked materials present in the sol fraction play in the surface wetting dynamics of PDMS materials. We further demonstrate the role of uncrosslinked material by adding silicone oils into otherwise fully crosslinked PDMS networks and study their wetting properties.;Through careful formulation and preparation of PDMS materials, compared to simply mixing two formulations present in Sylgard 184, one can apply polymer network models to glean useful information about network topology. The benefits of doing so outweigh the costs. We stress the importance of performing Soxhlet extraction to remove unreacted components from PDMS materials, even when using optimal stoichiometry. These mobile molecules that remain after crosslinking can alter significantly wetting behavior and readily leach into liquid environments. However, it is equally important to stress that Soxhlet extraction will not remove all unreacted material. Some will always remain in PDMS, which is often the practice in preparing microfluidic devices. While Sylgard 184 is very well suited for some applications, the results presented in this dissertation demonstrate to researchers that the material does have its limitations and that other options are available. These findings will aid in the design and implementation of reliable microfluidic devices and other PDMS-based materials that encounter liquid interfaces.