One of the primary factors influencing the research and development of rigid-rod polymers in the Materials Laboratory has been the excellent environmental resistance of this class of aromatic heterocyclic benzobisazole polymers. Thermal stability of these materials has been determined utilizing thermal analytical, Isothermal Aging and combined Thermal Gravimetric-Mass Spectrometry (TG-MS) techniques. Correlations between the thermal stability and chemical structure, varying substituent groups and processing conditions for these polymers have been observed. An overview of the thermal stability of these rigid-rod polymer systems is presented.

The rigid-rod polymer system poly(phenylenebenzobisthiazole), designated as PBZT, has considerable potential in materials engineering. The vacuum thermal behavior of this system and various substituted analogues, including methyl-, phenyl-, and benzothiazolyl-substituted members, has been studied using the combined techniques of thermogravimetry and mass spectroscopy (TG-MS).

Primary volatile products identified from the PBZT series include H2S, HCN, CS2, NH3, C6H5CN, C6H4(CN)2, and HSCN. The presence of pendant groups such as methyl, phenyl, or benzothiazolyl results in the evolution of methane, benzene, or benzothiazole in low-temperature degradation processes. The relative abundance and temporal sequence of these gases which are evolved during slow heating in vacuum provide valuable information concerning thermal stability.

The implications of the identification and study of evolved gases toward the understanding of the mechanisms of heat treating of rigid-rod polymer systems will be discussed.

The current state of understanding of the morphology of the rigid-rod polymers PBZT and PBZO is reviewed. The morphology is sensitive to both the structure of the rigid polymer molecules and the nature of processing. Structural models of fiber morphology consistent with experiment are presented. An extensive bibliography is included.

The structures of poly(p-phenylenebenzobisthiazole) (PBZT) and poly(p-phenylenebenzobisoxazole) (PBO) fibers have been investigated by fiber diffraction techniques. d-spacings were obtained from equatorial and meridional scans recorded on a four-circle diffractometer. Intensity data were derived from x-ray rotation patterns taken on Weissenberg and vacuum cylindrical cameras. Unit cells were found to be monoclinic and non-primitive, each containing two chains per cell of dimensions a = 11.79(2), b = 3.539(5), c = 12.514(9) A, γ = 94.0(2)o for PBZT; and = 11.20(1), b = 3.540(2), c = 12.050(3) Å, γ = 101.3(1) for PBO. The fiber axes correspond to c. The conformational torsion angle between the bisthiazole and phenylene units and the orientation of chains within the unit cells were obtained from a ‘linked-atom least-squares’ (LALS)refinement procedure. A packing model is proposed for each polymer in which two independent molecular chains are displaced longitudinally by discrete rather than random increments. Primitive unit cells (Z = 1), besides requiring perfect axial registry of molecular chains, suffer from the occurrence of short intermolecular contacts and are rejected from further consideration.

Semiempirical energy calculations have been used to investigate the conformational properties of poly(p-phenylene benzobisthiazole) (PBZT) and poly(p-phenylene benzobisoxazole) (PBO) having various substituents on the phenyl ring. Substituents considered included methyl, ethyl, t-butyl, hydroxyl, phenyl, and benzthiazole groups. The torsional potential for rotation about the backbone bond joining the phenylene and heterocyclic rings has a significant influence on the conformational energies. X-ray structure data and results of semiempirical molecular orbital calculations were used to determine this critical parameter.

The results show that most substituents have surprisingly little influence on the conformations of the molecules. In most cases, the substituent causes a rotation of the backbone rings away from coplanarity, but once the steric interactions are relieved, there is considerable latitude for further rotation. The substituents most effective in locking the molecules into specific conformations were hydroxyl and t-butyl, due, respectively, to the prospect for hydrogen bonding, and non-planarity. In most cases, the conformational freedom is such that packing considerations could significantly influence the conformations of the materials in the solid state.

Row-nucleated structures in PPTA and PBZT thin films are compared and their dependence upon quench rate and molecular weight illustrated. It is proposed that the lamellar appearance of the morphology is due to a coherent periodic twisting of fibrillar ribbons. The role of crystallosolvate phases is discussed.

Fibers of rigid-rod polymers are comparatively weak in compression with deformation occurring by strain localization into kink bands. Here, we present an ultrastructural study of the deformation processes accompanying kink formation in fibers of poly(paraphenylene benzobisthiazole) (PBZT) and poly(paraphenylene benzobisoxazole) (PBZO). Lateral packing between chains is maintained after shear deformation. The local strain is heterogeneous and has been analyzed in detail. Observation of sharp, high angle tilt boundaries confirms a deformation mechanism which involves kinking of individual polymer chains. A model of the micromechanisms involved in kinking is proposed.

Uniaxially oriented, crystalline fibers of poly(2,2'-(p,p'-biphenyl)- 6,6'-bis(4-phenylquinoline)) [PBPQ], were subjected to structure analysis via x-ray diffraction and molecular packing analysis. Packing analysis was required due to the sparse and diffuse nature of the fiber diffraction photographs. The combination of the two techniques allowed us to resolve the crystal structure. PBPQ packs in a syn conformation, in a modified herringbone structure. It has a unit cell of a = 16.2Å, b = 5.31Å, c = 21.4Å. There are two independent chains in the unit cell. An overview of the refinement of the crystal structure, the packing analysis and the crystallography are presented.

An exact pole inversion method is used to calculate the complete chain-axis orientation distribution from a single inter-chain reflection. A brief outline of the formulation and computation method is given. This method is demonstrated by comparing the measured chain-axis distribution from a (002) reflection of oriented polyethylene samples, to the calculated distribution from (110) and (200) reflections of the same sample. The error in the Hermans Orientation function is less than 0.5% for the two samples tested.

To determine the compatibility between the rigid rod and the flexible chain polyquinolines, both small angle x-ray and neutron scattering measurements were conducted on a sample containing deuterated flexible components. The scattering intensities from both x-ray and neutron were reduced to absolute scale in order to remove the scattering contribution from microvoids which tended to overshadow the signal of molecular origin. Quantitative information regarding the extent of molecular segregation in a polyquinoline multi-block copolymer was obtained, the result indicated that the material was strongly associated but not segregated into single component phases.

We report the room and low temperature crystal structures of the unsubstituted pquinquephenyl (PQP) and p-sexiphenyl (PSP), and substituted 22,45 -diphenyl-pquinquephenyl (DPQP), 22,65-diphenyl-p-septiphenyl (DPSP), and 1,2,4-triphenylbenzene (TPB) polyparaphenylene oligomers. The unsubstituted oligomers exhibit a solid state transition when cooled from room temperature to 110K, as indicated by a change in crystallographic space group. No transition is observed for the substituted oligomers other than the usual thermal contraction of the unit cell. The transition observed for the unsubstituted oligomers is interpreted in terms of a conformational change from an “averaged” planar structure to a static non-planar one. Comparisons of room temperature and low temperature crystal data are presented.

Recently, the concept of rigid rod molecular composites, originally demonstrated in the Air Force Materials Laboratory, has attracted wide spread attention in the global research community in the hope of creating a new generation of materials for structural and electronic applications in extreme environments. As originally conceived, a molecular composite is a homogeneous, synergistic composite of molecularly dispersed rigid rod polymer molecules (single molecules or small bundles of molecules with a lateral dimension of less than 50 Å) with high aspect ratios in a matrix material. The enhanced and desirable properties, such as superior chemical and environmental resistance, enhanced thermal and thermoxidative stability, toughness, and dimensional stability, resulting from the synergism between the reinforcing rigid rods and the matrix can only be realized in a true molecular composite. Without the development of true rigid rod molecular composite technology, the fundamental detrimental interfacial problems (adhesion, different thermal expansion coefficients, etc.) encountered in conventional fiber composites can not be averted. A truly revolutionary material technology can be developed only if one adheres to the original concept.

Rigid-rod molecular composites are a new class of high performance structural polymers which have high specific strength and modulus and also high thermal and environmental resistance. A rigid-rod, extended chain polymer component is used to reinforce a matrix of a ductile polymer with the intent of achieving a “composite” on the molecular level. After synthesis, the key to producing a molecular composite is to control morphology to disperse the reinforcing rod molecules as finely as possible in the matrix polymer. Individual rod molecules or bundles of molecular rods must have dimensions which result in a high ratio of length to width (aspect ratio) for efficient reinforcement. To achieve this, the reinforcing rod component must not phase separate at any stage of processing. Morphological characterization techniques, which can measure the orientation and dispersion (or, conversely, the degree of phase separation) of rod molecules provide the tools for correlating theoretically predicted and experimentally observed mechanical properties. Various morphological techniques which have been applied to molecular composite systems will be reviewed, including wide angle x-ray scattering and scanning and transmission electron microscopy. Structure-property correlations for molecular composite systems will be discussed with regard to models for mechanical properties. Application of new morphological techniques will also be discussed.

The thermally induced phase separation of poly(p-phenylene benzobisthiazole)/ Nylon 66 molecular composites was followed by small-angle light scattering which showed the development of a scattering ring. The intensity increased and the ring moved towards the main beam as a function of time.

The scattering vector qm and intensity maxima Im scaled as qm~t−∝ and Im~tβ with ∝~ 0.33 and β = 0.91-0.95, in close agreement with the cluster dynamics prediction of Binder. However, at longer times, the increase in both qm and Im slowed down dramatically indicating a different mechanism. The structure function S(x) was used to test the validity of universal scaling for a rigid-rod/flexible coil polymer blend. At large x, S(x) scaled to the power of −2.4 in x, which is in between values predicted for systems with one and two-dimensional diffusion.

Kinetics of phase separation of PBO in methane sulfonic acid are examined by time-resolved small angle light scattering. Thin films of polymer solutions are spincoated on microscope slides and observed in a controlled environment light scattering cell. These solutions appear to undergo spinodal decomposition, following Cahn- Hilliard type phase separation kinetics upon exposure to water, leading to an exponential increase in the scattered light intensity at short times.

Previous work investigated the processing/structure/property relationships for solution spun composite fibers of rodlike poly(p-phenylene benzobisthiazole) [PPBT] with nylon 6,6 [N66] and poly(ether ether ketone) [PEEK] [1]. Evidence that these composite fibers were effectively reinforced by a network-like structure of PPBT was given. In the present investigation, the morphology and properties of wet, as-spun, and heattreated PPBT/nylon 6,6 and PPBT/PEEK composite fibers are contrasted to gain insight into the structure/property relations which result from these post-spinning processes. It is concluded that heat-treatment is simply a means of perfecting the network structure, thereby enhancing the tensile properties of the composite fibers.

A novel method of preparing PBZT/nylon composite fibers by infiltrating nylon into pure PBZT fiber is described. The pure PBZT fiber formed a microfibrillar network structure during coagulation, which is effective in reinforcing the matrix in the “in-situ network” composite fibers (designated IC). These new composite fibers exhibit nearly indistinguishable mechanical properties as those of “molecular” composite fibers (MC) prepared from isotropic solutions before and after tension heat-treatment (E = 44 GPa, σ = 430 MPa, ε = 1.2 %, σc = 250 MPa, G = 1.75 GPa) for PBZT/nylon weight ratios equal to unity.

The fine structure of pure PBZT and its composite fibers spun from isotropic solutions was characterized using techniques based on nitrogen adsorption, small-angle X-ray scattering, scanning and transmission electron microscopies. The structure of both type of composites was found to be a microfibrillar network of PBZT in a matrix of amorphous nylon. The average diameters of the PBZT microfibrils were in the range of 10 to 20 nm for the IC and 4 nm for the MC.

The goal of this program was to form and demonstrate improved mechanical properties of biaxially oriented films of 30/70 rod-coil polymer blends of PBZT/ABPBT. These polymeric blends were prepared at three different solids contents C<CCR, C≳CCR and C>CCR.

Our approach examined PBZT and ABPBT in poly phosphoric acid (PPA) as the common solvent for the first time. Also, we set out to go one step beyond the current state of the art and produce biaxially oriented PBZT reinforced ABPBT molecular composites. Our investigation marked the first time such a study has been attempted. Additionally, fibers of C≳CCR were spun and characterized.

The wet-state-consolidation process was used to fabricate bulk rigid-rod/thermoplastic molecular composites of poly(p-phenylene benzobisthiazole) and poly(ether ether ketone). The composites were molded into bar-shaped specimens. The morphology of those specimens was studied throughout the entire process using X-ray scattering and scanning electron microscopy to correlate processing conditions and properties. The mechanical properties of the bulk composites were comparable to those of other rigid-rod/thermoplastic molecular composite systems.

Proton spin diffusion experiments are conducted on two blends of nylon and poly(phenylenebenzobisthiazole) (PBZT). The blends are prepared different ways and utilize either a semicrystalline or fully amorphous nylon. The spin diffusion experiment, which was conducted using three different schemes, yields information about the shortest distances typifying the individual domains. Results suggest that in a 40/60 nylon 6,6/PBZT blend, the nylon and PBZT molecules do not mix on the molecular level, however, each domain is only about 4 nm in its minimum dimension. At the same time, some crystalline nylon regions also exist having somewhat larger dimensions. In contrast, a 50/50 amorphous nylon/PBZT sample, made via a different processing route, showed evidence for similar mixing in only a portion of the material; there was also strong evidence for compositional heterogeneity on a scale larger than 10 nm. The exact stoichiometries describing this compositional heterogeneity could not be uniquely identified on the basis of the spin diffusion studies alone; however the possibilities range from 55% of pure nylon to 35% of pure PBZT being sequestered from the more intimately mixed phase.

In a second set of experiments, the 250°C thermal polymerization chemistry of a benzocyclobutene (BCB), a bismaleimide (BMI) and an equimolar mixture of BCB and BMI was studied. These materials are candidates for thermally polymerizable matrix materials in blends with rigid-rod polymers like PBZT. Polymerization of the BCB homopolymer is found to involve multiple pathways and there is reasonably strong evidence that the reactive carbons generated by thermal ring opening of the cyclobutene moiety not only attack one another; they also attack aromatic carbons adjacent to carbonyl carbons at sites far from the ring opening reaction. Spectra of the BCB homopolymer aged at 300°C in vacuum indicate that significant chemical changes occur. This result casts a shadow on the interpretation of the good apparent thermogravimetric stability of the BCB homopolymer in air at this temperature. The equimolar mixture of BCB and BMI ideally can polymerize via the Diels-Alder mechanism to produce a linear chain of alternating BCB and BMI repeat units. The spectrum of this thermally polymerized mixture suggests a maximum of 50% of the reaction following the Diels-Alder addition; the other 50% follows homopolymerization pathways.