A facile method for permanent and functional surface modification of poly(dimethylsiloxane).

Poly(dimethylsiloxane) (PDMS) is the choice of material for a wide range of applications,1-3 because PDMS has many advantageous properties such as chemical inertness, nontoxicity, ease of handling, and commercial availability. It is impossible, however, to have one material that meets all the individual needs of microfluidic systems,1 micro-electromechanical systems (MEMS),2 and cellular study.3 New materials have been developed to replace PDMS. For example, a photocurable perfluoropolyether (PFPE) was synthesized to fabricate microfludic devices that were organic solvent compatible.4,5 In a consensus, it is costly to develop a new elastomer for each individual need. And we believe surface modification of PDMS will be a cost-effective and time-saving strategy, if a facile method for surface modification can be developed, since surface modification retains the desired bulk properties of PDMS and reveals the need for new material development. A number of strategies have been developed for PDMS surface modification, which can be divided into two categories, namely physisorption and chemical coupling. Physisorption of materials to PDMS surface, such as surfactants6 and polyelectrolytes7 are driven by hydrophobic force and electrostatic force, respectively. This simple method ensures PDMS based devices after modification perform well in situations that only require moderate density and thickness of coatings, and only to sustain low shear force. Chemical coupling is stable but is difficult to achieve because PDMS is chemically inert, which is ironically one of its merits. Common for this approach, the first step is to apply high-energy bombardment (i.e., plasma) to PDMS surface, which results in a silicate layer with functional groups on the surface, such as -OH and -NH2. Those functional groups not only render the surface hydrophilicity but also allow further modification via chemical coupling.8 Chemical coupling has two problems: (1) Plasma treatment is easy but not sustainable; recovery of hydrophobicity of treated PDMS is well documented.9 High-energy bombardment also has the tendency to damage PDMS. Furthermore, this strategy is only applicable to planar structure because of its limited penetration depth. (2) Concentration gradient in “grafting to” strategy prevents the preparation of thick and dense films.10 We reported herein a facile method for permanent and functional surface modification of PDMS based on a commercial material. First, a vinyl-terminated initiator (v-initiator, part C in Scheme 1) was mixed with the viscous base and curing agent of Sylgard 184, resulting in an initiator integrated PDMS (iPDMS). The base is a poly(dimethyl-methylvinylsiloxane) prepolymer with small amount of platinum (Pt) catalyst (part A in Scheme 1) and the curing agent is a mixture of vinyl-endcapped PDMS precursors and poly(dimethyl-methylhydrogenosiloxane) precursors as cross-linkers (part B in Scheme 1). Upon mixing together (the so-called curing process), the vinyl groups and the hydrosilane hydrogens undergo a hydrosilylation reaction in the presence of Pt catalyst, which results in highly cross-linked three-dimensional networks. It is a common practice to tune the mechanical property of PDMS by varying the ratio between A and B.11 We reasoned that although the attachment of initiator to cross-linkers would evidently decrease the degree of cross-linking, one could introduce enough v-initiators into the network but only cause limited property change (i.e., mechanical property) by carefully choosing the ratio of A/B/C. In fact, as a random cross-linking process, the network formation is not perfect even without component C and there is always a small amount (<5%, w/w)12 of unreacted functional groups left. We found that below a critical ratio of 10:1:0.5 (A/B/C) the mixture cured as regular PDMS (Young’s modulus E ∼2.12 MPa, contact angle θ ∼112°) and the resulting iPDMS (E ∼2.05 MPa, θ ∼114°) was successfully used in replica molding. The key for successful surface modification was whether v-initiators would be presented at the surface. X-ray photoelectron spectroscopy (XPS) was applied to characterize the surface composition of iPDMS. Fresh iPDMS was extracted thoroughly with organic solvents to remove unreacted oligosiloxanes and trapped v-initiators.12 Survey scans of iPDMS showed a v-initiator unique Br 3d peak at 71 eV (Figure 1A). Three-dimensional XPS scans provided more information on the distribution of initiators in iPDMS: the X-Y plane was characterized via sequential point-scan of a 3 × 3 square (9 points, step of 1 mm). The calculated and experimental atomic concentrations (atom %) agreed well for both PDMS and iPDMS for Si, C, and O (Table 1). The value of atom % for Br was lower than the † Department of Biomedical Engineering. ‡ Department of Advanced Materials and Nanotechnology. Scheme 1. Preparation of iPDMS and Permanent Surface Mondification of iPDMS via SI-ATRP Published on Web 05/17/2007