Phthalonitrile polymers offer promise as matrix materials for advanced composite applications. The phthalonitrile monomer is readily converted to a highly crosslinked thermosetting polymer in the presence of thermally stable organic amine catalysts. Rheometric studies were conducted to elucidate the optimum amine concentration for composite formulations. High quality composite panels were processed in an autoclave using unsized IM7 carbon fibers. Mechanical properties of the phthalonitrile/carbon composite are either better than or comparable to the state‐of‐the‐art PMR‐15 composites. Dynamic mechanical analysis reveal that samples postcured at elevated temperatures (375°C) do not exhibit a glass transition temperature up to 450°C and also retain °90% of their initial modulus at 450°C. Flame resistance of phthalonitrile/carbon composites, evaluated by cone calorimetric studies, excels over that of other polymeric composites for marine applications. The composites also show low water uptake, <1% after exposure to water for 16 months.

Committee Chairman: Garth L. Wilkes Chemical Engineering (ABSTRACT)A series of four water-blown flexible polyurethane foams was produced in which the water content was varied from 2 to 5 pph at a constant isocyanate index of 110.A portion of each foam was thermally compression molded into a plaque. The morphology of the foams and plaques was investigated using OMS, DSC, FTI R, TEM, SEM, swelling, WAXS, and SAXS. A high degree of phase separation occurs in these foams and the degree of phase separation is independent of water (ha rd segment) content. In the foam with the lowest water content the morphology is similar to that of typical segmented urethane elastomers. Small hard segment domains are present with a correlation distance of roughly 7. 0 nanometers. When the water content is increased a binodal distribution of hard segments appears.There are the smal I ha rd segment domains typical of segmented urethane elastomers as well as large hard segment aggregates greater than 100 nanometers 1n diameter. The large domains are thought to be aggregates of polyurea that precipitated during the manufacture of the foam. The foam making process successfully incorporated the trifunctional polyols into a network indicating a high degree of polymerization for the hydroxyl-isocyanate reaction. Unreacted isocyanate is present in the foams a month after curing. It is believed to be trapped in the large urea aggregates.WAXS patterns of the foams suggest hard segment ordering that may be of a paracrystalline nature but certainly lacking in true crystallinity. ACKNOWLEDGEMENTSI would like to thank my major professor Dr. Wilkes for his patience, concern, and support, without which I probably would not have remained in Blacksburg to complete this work.I wou Id also Ii ke to thank Dr. Glasser and Dr. Sebba for the interest they showed in my work and the time and effort they spent reviewing my thesis.I would like to thank Rick Allen and Dr. Lin for their assistance in using the ORNL facilities and Dinesh Tyagi and Brandt Carter for many helpful discussions.Finally, I would like to thank all of the group members for their everyday greetings, smiles, and conversion. In particular I would like to thank Dinesh and Ruby, Brandt, Marty and Martha, andBruce.iv One reason for this is that in the production of polyurethane foams, the blowing and gelling reactions occur simultaneously. Any change in chemistry or processing conditions alters both the cell structure and the morphology of the material comprising the cell structure. As a result, there is no way to evaluate the material independent of cell structure or cell structure independent of material. Changes in chemical and processing variables have been extensively studied as to their influence on bulk properties of the foam and polyurethane foams with a wide range of physical properties can be produced ( 1-5).This 'technological' approach to the development of polyurethane foams has been very successful and presently, the yearly production of polyurethane foams far outweighs other types...

Phthalonitrile polymers, under development at the Naval Research Laboratory, offer promise as high temperature, high performance composite matrix materials. A fully cured resin shows outstanding thermal stability with no evidence of a glass transition temperature or Tg up to 450°C, good mechanical properties, and is easily processed into void‐free components. Phthalonitrile/glass fabric composite panels have been successfully fabricated by conventional consolidation of prepregged glass and by a more recently developed simplified process, resin infusion molding. Both processes can be used to produce panels with comparable mechanical properties. More important, flammability performance of these composites, evaluated in terms of specific optical density, combustion gases, heat release, and ignitability, excels over other state‐of‐the‐art polymer/glass composites. This finding is significant given that overcoming flammability obstacles has been the main limiting factor for use of composites in marine applications.

No abstract

Isothermal spherulite growth rates were measured over a sufficient range of undercoolings, ∆T, for a narrow linear polyethylene fraction M ) 70 300 (70.3K), polydispersity 1.12, such that the fraction exhibited all three growth regimes as crystallized from the subcooled melt. The I-II transition occurred at ∆T I-II ) 15.8 °C and the II-III transition at ∆TII-III ) 23.8 °C. (Neither transition was fully abrupt.) The nucleation constants Kg and preexponential factors G0 that described the absolute growth rates for each regime were determined, thus quantifying key parameters for all three regimes for a single specimen measured in the same apparatus. The Kg's for 70.3K conformed to the predicted relationship Kg(III) = Kg(I) ) 2Kg(II). Theoretical relationships for the preexponential factors were employed using the observed G0's to investigate the nature of the transport of chain segments to the growth front. It was reconfirmed that this process was forced "near-ideal" reptation for an M = 30K fraction. For M ) 70.3K, it was found that the reptational transport mechanism in regimes II and III was perturbed and thereby slowed beyond that attributable to "near-ideal" forced reptation; the additional retardation was taken to be the result of labile chain attachments on a surface some distance from the site where the dangling chain was being drawn onto the substrate. In another test, the expression S k/ao for the stem separation between primary surface nuclei in regime II was employed to calculate ∆TI-II and ∆TII-III. This was successful for both M = 30K (near-ideal reptation) and M ) 70.3K (perturbed reptation). In this test, earlier estimates of quantities of importance to nucleation theory, such as C0, nIII, and the substrate length L, were found to be either identical or only slightly modified. The treatment leads to satisfactory numerical estimates of the absolute substrate completion rate g and the nucleation rate i, and is consistent with the crystal morphology present in melt-crystallized PE, including the lenticular crystal f truncated lozenge transformation associated with the I f II regime transition. In general, this work provides significant additional support for the "three regime" concept in narrow PE fractions crystallized from the melt through a consideration of nucleation, regime, and reptation concepts.