Influence of a Polymer Solution on a Polymer Interface P. Auroy* ,† and L. Auvray ‡ Institut Curie, URA 448 du CNRS, Section de Recherche, 11 rue Pierre et Marie Curie, 75231 Paris Cedex 05, France, and Laboratoire Le ´ on-Brillouin, CNRS-CEA, Centre d’Etudes Nucle ´ aires de Saclay, 91191 Gif-Sur-Yvette Cedex, France Received April 27, 1994; Revised Manuscript Received September 26, 1995 X ABSTRACT: We have investigated the influence of a solution of mobile polymer chains on a polymer interfacial layer. The mobile chains were synthesized to be invisible during the scattering experiment. We show how the interfacial density profile is modified as a function of concentration of mobile chains, polymer grafting density, and compatibility of the two polymeric species. We use these results to discuss the penetration of mobile polymer chains into immobilized polymer chains. The question whether a polymer interface can be penetrated by free polymer segments is of great tech- nological importance. For example, the degree of pen- etration can have a strong effect on the adhesive properties of two polymer blocks. This effect is well- known by industrial researchers, and the phenomenon is well understood and characterized from a scientific point of view. 1 The issue of interpenetration becomes more delicate when the chains of one polymer are in solution. de Gennes 2 has provided a theoretical “phase” diagram for the case of a grafted layer (grafting density σ, chain length N) in the presence of a homopolymer solution (chemically identical, concentration φ b , chain length P). Various regimes have been described, which depend on all these parameters. For instance, it has been shown that at “high” grafting density, in the brush regime, the interface evolves as the concentration φ b increases. It goes from a strongly stretched configura- tion without penetration to a weakly stretched config- uration with penetration (if P > N 1/2 ). Observing polymer penetration has remained an experimental challenge for scientists. Most of the techniques 3 that have been used for studying the influence of a polymer matrix are not applicable to the study of the influence of a polymer solution because the solution flows or because the solvent evaporates or degrades easily or because the solvent would dominate the signal, etc. Thus, to the best of our knowledge, no experimental results have been obtained on the influence of a polymer solution on a polymer interface. We have solved the above difficulties by using the small-angle neutron scattering (SANS) techniques. 4 In the past, SANS has allowed us to determine the density profile of various polymer interfaces in a pure solvent or more generally in a simple liquid. 5,6 This was achieved by using different isotopic compositions (H/D) for the solvent. Modifying the isotopic composition varies the contrast of the samples without introducing significant perturbations to the system. For instance, a solvent made using 90% CD 2 Cl 2 and 10% CH 2 Cl 2 allowed us to show that polymer brushes have a parabolic density profile in a good solvent. 5a When a polymer soluion replaces the solvent, the experiment becomes far more difficult. Indeed, the polymer in solution gives a coherent scattering signal which domi- nates the contribution from the interfacial layer. Due to chain correlations, this problem persists even when H-polymer chains and D-polymer chains are mixed to have the same average scattering length density as the solvent (henceforth referred to as n s ). The only way to avoid this problem is to use a statistical polymer of H-monomer and D-monomer whose composition exactly matches n s at any concentration. Under this condition, the free polymer in solution (henceforth referred to as the “stealth” polymer) is invisible to the neutrons and the scattering signal is due solely to the polymer interface. This contrast-matching technique has allowed us to determine how the shape of the density profile of a polymer interface changes as the concentration of the free polymer solution, φ b , is increased. Three types of interface were studied: (i) a high grafting density polystyrene (PS) brush, (ii) a low grafting density polystyrene layer in the slightly overlapping mushroom regime, and (iii) a polydimethylsiloxane (PDMS) pseudo- brush, consisting of a layer of irreversibly adsorbed PDMS (cf. Figure 1). These samples will be described in detail in the first section. For all three layers, the structure of the interface in pure good solvent has been extensively discussed in previous reports. 5 In this paper, we will limit ourselves to a brief description of the experimental method. In the second section, we will * To whom all correspondence should be sent. † Institute Curie. ‡ Laboratoire Le ´on-Brillouin. X Abstract published in Advance ACS Abstracts, December 1, 1995. Figure 1. Schematic representation of the three types of interface we treat in this article, in good solvent. 337 Macromolecules 1996, 29, 337-342 0024-9297/96/2229-0337$12.00/0 © 1996 American Chemical Society