Adsorption of Poly(methyl methacrylate) Melts on Quartz

• Published in: Macromolecules Vol. 32; no. 20; pp. 6772 - 6781
• Main Authors: Durning, C. J, O'Shaughness, B, Sawhney, U, Nguyen, D, Majewski, J, Smith, G. S
• Format: Journal Article
• Language: English
• Published: Washington, DC American Chemical Society 05.10.1999
• Subjects: Applied sciences; Exact sciences and technology; Organic polymers; Physicochemistry of polymers; Properties and characterization; Surface properties; Plane surface; Liquid solid interface; Molten state; Liquid solid adsorption; Methyl methacrylate polymer; Experimental study; Quartz; Layer thickness; Molecular weight property relation; Interface structure; Monodispersed polymer; Adsorbed layers; Kinetics
• ISSN: 0024-9297; 1520-5835
• DOI: 10.1021/ma981785k

We studied by neutron reflection the architecture of poly(methyl methacrylate) (PMMA) layers adsorbed on hydroxylated quartz from the melt. The samples were prepared by spin-coating about 1 μm thick films of PMMA onto quartz plates, annealing at melt conditions for an extended period, and then leaching away unbound material in good solvent (benzene). Data on dry residual layers indicated a dense PMMA layer the thickness of which gradually increased with annealing time in the melt from an initial minimal value toward a final equilibrium thickness. Evidently, annealing in the melt gradually relaxes the rather flat nonequilibrium structure produced by spin-coating. The thicknesses, h, in a series of dry residual layers annealed for very long times in the melt obey h ∼ N 0.47±.05, where N is the degree of polymerization. This is close to the scaling expected for a reflected random walk (RRW) immobilized by the surface (h ∼ N 1/2). Data on residual adsorbed layers swollen in a good solvent (deuterated benzene) indicate a strongly stretched, brushlike structure with a diffuse segment density profile, φ(z). The segment density decays φ(z) ∼ z -0.77±.03, faster than predicted by RRW statistics and indicates fewer long loops per chain than the RRW model.