Nanotube functionalization with 4-hydroxymethylaniline (HMA) via the diazonium salt method has been done on SWNT produced by the CoMoCAT process. Thermal analysis indicates 1 out of every 33 C atoms remains functionalized. Raman, FTIR, and optical absorption spectroscopy confirm the side-wall functionalization and show that it can be reversed by thermolysis. The OH group that is generated from the functionalization was used to start the ring-opening polymerization of ε-caprolactone. The polymer thus produced (PCL) remains grafted to the nanotube as demonstrated by FTIR and a dramatic increase in suspendibility in chloroform. By combining several techniques (TGA, TPD, and TPO) a quantification of the number of polymer chains attached to the nanotube and their average length has been attempted.

A commercially available aliphatic thermoplastic polyurethane formulated with a methylene bis(cyclohexyl) diisocyanate hard segment and a poly(tetramethylene oxide) soft segment and chain‐extended with 1,4‐butanediol was dissolved in dimethylformamide and mixed with dispersed single‐walled carbon nanotubes. The properties of composites made with unfunctionalized nanotubes were compared with the properties of composites made with nanotubes functionalized to contain hydroxyl groups. Functionalization almost eliminated the conductivity of the tubes according to the conductivity of the composites above the percolation threshold. In most cases, functionalized and unfunctionalized tubes yielded composites with statistically identical mechanical properties. However, composites made with functionalized tubes did have a slightly higher modulus in the rubbery plateau region at higher nanotube fractions. Small‐angle X‐ray scattering patterns indicated that the dispersion reached a plateau in the unfunctionalized composites that was consistent with the plateau in the rubbery plateau region. The room‐temperature modulus and tensile strength increase was proportionally higher than almost all increases seen previously in thermoplastic polyurethanes; however, the increase was still an order of magnitude below what has been reported for the best nanotube–polymer systems. Nanotube addition increased the hard‐segment glass transition temperature slightly, whereas the soft‐segment glass transition was so diffuse that no conclusions could be drawn. Unfunctionalized tubes suppressed the crystallization of the hard segment; whereas functionalized tubes had no effect. © 2007 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 45: 490–501, 2007