Nm For Materials Research Articles

OrbiSIMS metrology Part I: Optimisation of the target potential and collision cell pressure

In 2017, we introduced the OrbiSIMS instrument that features a dual analyser configuration with a time‐of‐flight (ToF) mass spectrometer (MS) and an Orbitrap MS, which confer advantages of speed and high‐performance mass spectrometry, respectively. The ability to combine the MS performance usually found in a state‐of‐the‐art proteomics and metabolomics MS with 3D imaging at the microscale and from nanolayers of <10 nm of material has proved popular in a broad field of application from organic electronics to drug discovery. There are now several instruments in operation around the world, and metrology is needed to help ensure repeatability and reproducibility of the intensity scale. We conduct a systematic study of two key parameters, the target potential, VT, and the collision cell pressure, P, in the transfer optics on the transmitted secondary ion intensities. We measure VT–P maps of the ions across the mass range for Ag as a representative of inorganic materials and two different organic materials, Irganox 1010 and NPB (N,N′‐Di[1‐naphthyl]‐N,N′‐diphenyl‐4,4′‐biphenyldiamine). The manufacturer's defaults for these values ensure very good transmission for a broad range of analyte classes. However, the maps reveal a sometimes complex behaviour and indicate the possibility for additional separation of ions based on their shape, labile nature and kinetics of formation. Guidance is provided on how to optimise these parameters for sensitivity for different material classes and also the need for optimisation to improve spectral repeatability and reproducibility.

Directional Dark Matter Search with Nuclear Emulsion

Dark matter search by using nuclear emulsion, NEWSdm project is described here. The expected Dark matter signal is the recoiled nucleus track stopping within a few 100 nm in material. The short track length makes it difficult to detect. We have developed a fine grained nuclear emulsion to record the trajectory of the recoiled nucleus as several aligned silver grains by the nuclear emulsion. Track length is shorter than visible light wave length (400nm to 700nm). Therefore, dedicated techniques to overcome this limitation have been developed by the Collaboration. Carbon ion beams of relevant energies have been used to demonstrate the directional sensitivity with these techniques. Moreover, the scanning speed with automated microscopes is continuously increasing. WIMPs detection sensitivity with 10 kg nuclear emulsion target with a year exposure at LNGS will cover a large part of the DAMA signal region.

Luminescence characteristics of Na21(SO4)7F6Cl:Cu+ phosphor

ABSTRACTCopper-doped Na21(SO4)7F6Cl phosphor was synthesized via the conventional wet chemical method. The synthesis was carried using CuCl2 and Cu (NO3)2·3H2O as dopants in two different steps successively. The formation and phase purity of the compound were revealed by the X-ray diffraction pattern. Functional groups of the prepared phosphor were observed in the FT–IR spectrum. The emission along with excitation spectra were followed to explore the luminescence attributes. Photoluminescence (PL) emission spectrum of the material synthesized using CuCl2 as the dopant was observed at 358 nm due to 3dl0↔3d94s transitions when excited around 247 nm for various copper concentrations. Efficient blue emissions were obtained at peaks 423 and 469 nm for materials synthesized using Cu (NO3)2·3H2O as the dopant, when monitored at 357 nm excitation. The Commission Internationale de I’Eclairage chromaticity coordinates for different copper concentrations were calculated for the emission around 423 nm. TL glow curves of Na21(SO4)7F6Cl:Cu phosphor for different dopant concentrations, irradiated with 100 Gy gamma dose, were studied and hence the trap parameters, namely order of kinetics (b), activation energy (E) and frequency factor (s) associated with the most intensive glow peak of Na21(SO4)7F6Cl:Cu phosphor were determined by using Chen’s Peak shape method. The results indicate that Na21(SO4)7F6Cl:Cu+ is a potential novel blue-emitting lamp phosphor and may be quite suitable for use in dosimetry of ionizing radiations.

Raman Mapping and Delineation of III-V Semiconductor Band Bending Characteristics

Raman spectroscopy has previously been used to ascertain the nature and extent of band bending at III-V semiconductor interfaces. Many previous investigations have focused on samples cleaved in vacuum and subsequently exposed to oxygen gas or a thin metal deposition; generally, specific polarizations of incident and scattered light are applied and collected in order to specify and isolate the desired phonon peaks under investigation1,2. We demonstrate herein a more general and easily applied method to investigate III-V interfaces by Raman spectroscopy using analysis of the longitudinal optical (LO) and transverse optical (TO) phonons; furthermore, we provide a means by which the ratios resulting from such experimentation may be normalized. Interfaces can be deeply buried under tens or hundreds of nm of material and still be investigated, unlike the case with photoelectron spectroscopy; furthermore, this technique can provide electronic mapping of such interfaces with micron sensitivity.

Toward Optimized Surface δ-Profiles of Nitrogen-Vacancy Centers Activated by Helium Irradiation in Diamond.

The negatively charged nitrogen-vacancy (NV) center in diamond has been shown recently as an excellent sensor for external spins. Nevertheless, their optimum engineering in the near-surface region still requires quantitative knowledge in regard to their activation by vacancy capture during thermal annealing. To this aim, we report on the depth profiles of near-surface helium-induced NV centers (and related helium defects) by step-etching with nanometer resolution. This provides insights into the efficiency of vacancy diffusion and recombination paths concurrent to the formation of NV centers. It was found that the range of efficient formation of NV centers is limited only to approximately 10 to 15 nm (radius) around the initial ion track of irradiating helium atoms. Using this information we demonstrate the fabrication of nanometric-thin (δ) profiles of NV centers for sensing external spins at the diamond surface based on a three-step approach, which comprises (i) nitrogen-doped epitaxial CVD diamond overgrowth, (ii) activation of NV centers by low-energy helium irradiation and thermal annealing, and (iii) controlled layer thinning by low-damage plasma etching. Spin coherence times (Hahn echo) ranging up to 50 μs are demonstrated at depths of less than 5 nm in material with 1.1% of (13)C (depth estimated by spin relaxation (T1) measurements). At the end, the limits of the helium irradiation technique at high ion fluences are also experimentally investigated.

Single crystal growth, x-ray structure analysis, optical band gap, raman spectra, strain tensor and photoluminscence properties in [HgCl<sub>4</sub>]<sup>-</sup> [R]<sup>+</sup> and [ZnCl<sub>4</sub>]- [R]+ (R= 2-amino-5- chloropyridine) hybrid materials

The single crystal growth of tetrachloromercurate (II) [HM-1] and tetrachlorozincate (II) [HM-2] with 2-amino-5-chloropyridine has been performed by slow cooling (SC) crystal growth technique of solution growth methodin which needle shaped transparent single crystals (0.5 x 0.2 x 0.2)mm were obtained. The crystal structures of these hybrid materials have been studied by X-ray diffraction, experimental and computational methods. [HgCl 4 ] 2- anions have a distorted tetrahedral geometry and the tetrahedra hybrid structure exhibit interwoven inorganic-organic layers mingled through N-H δ+ ...Cl δ- hydrogen bonding interactions. The mercurophilic interactions [Hg...Hg = 3.984(5)A] and halogen interactions [Cl...Cl = 3.406(2)A] form 2D parallelogram pattern of secondary interactions in [HM-1] whereas for [HM-2] crystal structure is stabilized by Cl...Cl = 3.357(2)A interactions. UV-vis absorption spectra depict the change in optical band gap from 3.01 eV to 3.42 eV on replacing the metal halide group, could be due to increase in optical absorption as a function of wavelength. The Raman and Hyper-Raman tensors calculations were performed based on single crystal X-ray data and the Lagrangian strain tensor calculations show the degree of lattice distortion = 1.794 between [HM-1] and [HM-2] which are useful tools for the optical response properties of inorganic-organic hybrid derivatives. The photoluminescence emission spectra peaks were observed in the wavelength range of 371 to 598 nm for material [HM-1] and in the wavelength range of 384 to 600 nm for material [HM-2] and lie in the visible range for both materials.

Tunable stoichiometry of SiO x - BaTiO y - BO z fabricated by multitarget pulsed laser deposition

Oxide materials of desired stoichiometry are challenging to make in small quantities. Nanostructured thin films of multiple oxide materials were obtained by using pulsed laser deposition and multiple independent targets consisting of Si, BaTiO3, and B. Programmable stoichiometry of nanostructured thin films was achieved by synchronizing a 248-nm krypton fluoride excimer laser at an energy of 300 mJ/pulse, a galvanometer mirror system, and the three independent target materials with a background pressure of oxygen. Island growth occurred on a per pulse basis; some 500 pulses are required to deposit 1 nm of material. The number of pulses on each target was programmed with a high degree of precision. Trends in material properties were systematically identified by varying the stoichiometry of multiple nanostructured thin films and comparing the resulting properties measured using in situ spectroscopic ellipsometry, capacitance measurements including relative permittivity and loss, and energy dispersive spectroscopy (EDS). Films were deposited ∼150 to 907 nm thickness, and in situ ellipsometry data were modeled to calculate thickness n and k. A representative atomic force microscopy measurement was also collected. EDS, ellipsometry, and capacitance measurements were all performed on each of the samples, with one sample having a calculated permittivity greater than 20,000 at 1 kHz.

Spatiotemporal temperature and density characterization of high-power atmospheric flashover discharges over inert poly(methyl methacrylate) and energetic pentaerythritol tetranitrate dielectric surfaces

A flashover arc source that delivered up to 200 mJ on the 100s-of-ns time-scale to the arc and a user-selected dielectric surface was characterized for studying high-explosive kinetics under plasma conditions. The flashover was driven over thin pentaerythritol tetranitrate (PETN) and poly(methyl methacrylate) (PMMA) dielectric films and the resultant plasma was characterized in detail. Time- and space-resolved temperatures and electron densities of the plasma were obtained using atomic emission spectroscopy. The hydrodynamics of the plasma was captured through fast, visible imaging. Fourier transform infrared spectroscopy (FTIR) was used to characterize the films pre- and post-shot for any chemical alterations. Time-resolved infrared spectroscopy (TRIR) provided PETN depletion data during the plasma discharge. For both types of films, temperatures of 1.6–1.7 eV and electron densities of ∼7–8 × 1017/cm3 ∼570 ns after the start of the discharge were observed with temperatures of 0.6–0.7 eV persisting out to 15 μs. At 1.2 μs, spatial characterization showed flat temperature and density profiles of 1.1–1.3 eV and 2–2.8 × 1017/cm3 for PETN and PMMA films, respectively. Images of the plasma showed an expanding hot kernel starting from radii of ∼0.2 mm at ∼50 ns and reaching ∼1.1 mm at ∼600 ns. The thin films ablated or reacted several hundred nm of material in response to the discharge. First TRIR data showing the in situ reaction or depletion of PETN in response to the flashover arc were successfully obtained, and a 2-μs, 1/e decay constant was measured. Preliminary 1 D simulations compared reasonably well with the experimentally determined plasma radii and temperatures. These results complete the first steps to resolving arc-driven PETN reaction pathways and their associated kinetic rates using in situ spectroscopy techniques.

Phase evolution during CuInSe 2 electrodeposition on polycrystalline Mo

Using structural analyses means of ex-situ Raman spectroscopy and X-ray diffraction combined with electrical measurements, we study the phase evolution in the growth by electrodeposition technique of CuInSe 2 on polycrystalline Mo. For this purpose the growth was stopped at different stages, and then the different layers were analysed. First growth steps seem to be controlled by the deposition of secondary phases, like elemental Se and Cu 2Se binary. After the deposition of approximately 300 nm of material, CuInSe 2 ternary and ordered vacancy compounds start to adequately form. At a thickness close to 2000 nm, the formation of binary Cu x Se is observed, remaining up to the final growth process (4350 nm). All these results are compared with the kinetic model of the system under the consideration of the experimental composition evolution.

Reducing thermal conductivity of thermoelectric materials by using a narrow wire geometry

The dependence of the thermal conductivity of narrow wires made from bismuth and covalently bonded materials on wire diameter was numerically calculated by considering contributions of mean free paths of carriers and phonons. The results suggest that a reduction in the thermal conductivity should be observable in a bismuth wire having a diameter of less than 1 μm sample. A reduction of nearly 20% in the temperature range of 150–300 K is expected due to the use of a narrow wire geometry. Such a geometry reduces the mobility and the thermal conductivity of the carriers, which is the dominant component, while the thermal conductivity due to phonons was dramatically reduced by using narrow wires at temperatures under 50 K due to the longer mean free path phonons. The thermal conductivity of materials with covalent bonding such as silicon was also estimated, and it is expected that the thermal conductivity of a silicon wire could be reduced due to the mean free path of phonons being longer than that of the carriers. The results suggest that it should be possible to enhance the figure of merit by reducing the thermal conductivity through using wire geometries having diameters of less than 100 nm in materials having low mobilities, high thermal conductivities, and high Debye temperatures.

Foam-derived multiferroic BiFeO3 nanoparticles and integration into transparent polymer nanocomposites

Transparent BiFeO3/PMMA (PMMA = poly(methyl methacrylate)) nanocomposites with a thickness up to 40 µm are obtained via solution film casting of BiFeO3 nanoparticle dispersions. The multiferroic BiFeO3 nanoparticles are obtained in a space confined crystallisation process in a porous carbon matrix and subsequent removal of the matrix. The average crystallite size ranges from 46 nm to 106 nm for materials synthesised at 803 K and 903 K respectively, but, at higher temperature, significant amounts of side products are detected by means of X-ray powder diffraction. The nanoparticles are multiferroic with a magnetic coercivity of 1.88 A m−1 at 400 K and a polarisation of up to 1.04 µC/cm2 at 65 kV/cm (298 K). Surface functionalisation using oleic acid allows the integration into PMMA films with a high transmittance.

Superparamagnetic Cobalt Ferrite Nanocrystals Synthesized by Alkalide Reduction.

AbstractFor Abstract see ChemInform Abstract in Full Text.

Superparamagnetic Cobalt Ferrite Nanocrystals Synthesized by Alkalide Reduction

CoFe2O4 nanocrystallites have been synthesized by alkalide reduction of Co2+ and Fe3+ to form nanoscale CoFe2, followed by oxidation with aerated water at room-temperature resulting in the nanocrystalline ferrite. As produced, the material consists of 2−4 nm nanocrystals that are superparamagnetic with an average blocking temperature of ∼350 K. Annealing at 100 °C in air results in a decrease in the blocking temperature to ∼250 K, with no detectable nanocrystallite growth. The dramatic change in the magnetic properties upon annealing is probably due to removal of crystal defects, namely oxygen vacancies. Further annealing to temperatures as high as 400 °C results in little change in the nanocrystallite size or the magnetic properties. Annealing at 500 °C results in the onset of significant growth in the nanocrystallite size, reaching 30 nm for material annealed at 1000 °C. The saturation magnetization, remanence, and squareness ratio, measured at 300 K, increase smoothly with increasing annealing temperature above 500 °C reaching 30 nm, 75 emu/g (94% of bulk value), 28 emu/g, and 0.37 respectively, for material that had been annealed at 1000 °C. The unannealed material has the largest coercivity observed in this study, 5.13 kOe at 5 K falling to 116 Oe at 300 K. The coercivity at 300 K declines dramatically to 0.9 Oe upon annealing at 100 °C, rising sharply to 67 Oe for material annealed at 500 °C, falling to 44 Oe for material annealed at 600 °C, and then steadily growing with increasing annealing temperature to 59 Oe for material annealed at 1000 °C. The anomalous increase in coercivity observed for samples annealed at 500 °C appears to be due to an increase in the average crystallite aspect ratio, which declines upon annealing at higher temperature.

Gas-cluster ion-beam smoothing of chemo-mechanical-polish processed GaSb(100) substrates

Gas-cluster ion-beam (GCIB) processing of surfaces provides individual atoms within an accelerated gas cluster (∼1,500 atoms per cluster), an energy approximately equal to the individual bond energy of the target surface atoms. The gas-cluster beam is thus capable of providing smoothing and etching of the extreme surface of numerous semiconductors, metals, insulators, and magnetic materials. For semiconductor material systems, the gas-cluster processing effect on the surface and subsurface material is of critical interest for device and circuitry application integrity. In the case of III–V GaSb, chemo-mechanical or touch polishing is the final step in the semiconductor-wafer manufacturing process, often leaving scratches of various depths or damage on the polished surface. In this paper, we report the GCIB etching and smoothing of chemical-mechanical polished GaSb(100) wafers. Using a dual-energy, dual gas-cluster source process, ∼100 nm of material was removed from a GaSb(100) surface. Atomic-force microscopy (AFM) imaging and power spectral-density (PSD) analysis shows significant decrease in the post-GCIB root-mean-square (Rms) roughness and peak-to-valley measurements for the material systems. X-ray rocking-curve analysis has shown a 24-arcsec reduction in the full-width at half-maximum (FWHM) of the (111) x-ray diffraction peak of GaSb. High-resolution transmission-electron microscopy (HRTEM) shows the crystallinity of the subsurface of the pre- and post-GCIB surfaces to be consistent, following the 1 × 1016 ions/cm2 total-fluence processes, with dislocation density for both pre- and post-GCIB cases below the HRTEM resolution limit. X-ray photoelectron spectroscopy (XPS) indicates a strong Ga 3p electron binding-energy intensity for gallium-oxide formation on the GaSb surface with the use of an oxygen GCIB process. Analysis of the Ga 3p electron binding-energy peaks in the XPS data in conjunction with HRTEM indicates a higher Ga or GaSb content in the near-surface layer (less stoichiometric-oxide presence) with use of a CF4/O2 GCIB process. The same peak analysis indicates that the surface gallium-oxide state is nearly unchanged, except in thickness, with the use of an O2-GCIB second step. The material results suggest that GCIB provides a viable method of chemo-mechanical polish (CMP) damage removal on group III–V material for further device processing.

Gas cluster ion beam processing of gallium antimonide wafers for surface and sub-surface damage reduction

In order to bring low-power epitaxy-based gallium antimonide (GaSb) electronics and electro-optics to market, high-quality GaSb substrates with smooth surfaces and no surface damage are required. Here, a novel final polishing technique, gas cluster ion beam (GCIB) processing, is shown to improve the surface finish of chemical–mechanical polished (CMP) 50 mm (1 0 0) GaSb wafers by etching and smoothing CMP surface atoms through the sub-surface damage. For the first time, a fluorine-based gas cluster ion beam is reported for GCIB surface etching and smoothing of GaSb material. For the selected processing sequence, the surface roughness of a high-quality, 0.70 nm RMS GaSb wafer was reduced to 0.18 nm RMS without any observed changes in the full-widths at half-maximum (FWHM) of the (4 0 0) and (1 1 1) X-ray peaks of 14 and 20 arcsec, respectively. Results indicate that the GCIB process did not contribute to wafer surface or sub-surface polish damage. In a second case, a GCIB etch removed 200 nm of material from a non-optimal CMP (1 0 0) GaSb surface and reduced the full-width at half-maximum (1 1 1) X-ray peak from 76 to 52 arcsec in conjunction with a surface roughness decrease from 0.70 to 0.35 nm RMS. The data suggests that GCIB processing appears to be promising as a final GaSb wafer polish with an etch rate compatible for large scale manufacturing.

Microstructural study of iron nitride thin films deposited by ion beam sputtering

An amorphous iron nitride thin film was deposited using reactive ion beam sputtering of iron by a beam of argon and nitrogen ions. Nitrogen content in the film as determined from conversion electron Mössbauer spectroscopy (CEMS) and X-ray photoelectron spectroscopy (XPS) was FeN 0.7. The mass density of the film was calculated using energy-dispersive X-ray reflectivity (EDXRR) measurements and is found to be 6.0 gm/cm 3. CEMS shows that the film is nonmagnetic in nature. Morphology of the film is obtained from atomic force microscopy (AFM). The surface roughness of the film does not increase appreciably beyond that of the substrate even after a deposition of 131 nm of material with these qualities the film is a good candidate for the multilayer superstructure of a nuclear Bragg monochromator of the type 56 FeN 0.7/ 57 FeN 0.7 .

Positron spectroscopy of vacancy-type defects in Si created by 5 keV implantation

Structural damage resulting from the implantation of 5 keV ions into FZ-Si has been investigated by positron annihilation spectroscopy (PAS) using a tuneable monoenergetic beam. Four samples, exposed to ion fluences from to , were studied. The PAS results demonstrate the applicability of the technique to the study of vacancy-type defects in small-scale device structures created by very low-energy ion implantation. Ion depth profiles determined by SIMS exhibited tails extending well beyond the limit predicted by the code TRIM, attributed to ion channelling. PAS, when extended by repeated measurements after precise etching of 40 and 140 nm of material via anodic oxidation, showed that the vacancy-type defect depth profiles also extended far beyond the limit predicted by TRIM. The ratio of defects to ions increases with depth, suggesting that the defect tails are not simply correlated to the implanted ions but that there may also be a contribution from post-implantation defect diffusion.

Progress in the Development of Table-Top Discharge-Pumped Soft X-Ray Lasers

The demonstration of large soft x-ray amplification in a dischargc-created plasma has opened a new patle to the development of compact and practical soft x-ray lasers. We review our progress in the development and study of these ultrashort wavelength lasers. The field has advanced from the first observation of large amplification in a discharge-created plasma in Ne-like Ar [J.J. Rocca. V.N. ShlyapLsev, F.G. Tomasel, O.D. Cortazar, D. Hartshorn, and J.L.A. Chilla, Phys. Rev. Lett. 73, 2192 (1994)]. to the denionstration of an extremely compact saturated laser at 46. 9 nm. In this paper we give an overview of these and other selected results. They include the observation of large amplification (gain-length product of 7.5) in Ne-like at 60.8 nm in material ablated from a solid target by a discharge, and preliminary results of the search for gain at a shorter wavelength in Ne-like Ca. A recent study of the spatial coherence of the capillary discharge 46.9 nm Ne-like Ar lasers which provides the first experimental measurement of a monotonic increase of the spatial coherence with length in a soft x-ray amplifier. is also summarized.

Phase separation in sputtered amorphous metal-germanium alloys.

Anomalous small-angle x-ray scattering has been used to observe and characterize phase separation in sputtered amorphous ${\mathrm{Fe}}_{\mathit{c}}$${\mathrm{Ge}}_{1\mathrm{\ensuremath{-}}\mathit{c}}$ and ${\mathrm{Mo}}_{\mathit{c}}$${\mathrm{Ge}}_{1\mathrm{\ensuremath{-}}\mathit{c}}$ thin films. Significant chemical inhomogeneity and anisotropy is observed on a very fine size scale. At low metal concentrations (c0.25), no films are homogeneous nor isotropic through the metal-insulator transition composition region (0.10\ensuremath{\le}c\ensuremath{\le}0.25). Measurements performed with the scattering vector both in and oblique to the film surface plane indicate that the chemical inhomogeneities are quite anisotropic. The cylindrical correlation functions, which have been calculated from the oblique scattering with a spherical harmonics approach, suggest a physical picture in which well-correlated inhomogeneities are formed at the surface during film growth, with little correlation between those formed in one layer and those formed after 2 nm of material have been deposited. These results suggest that fluctuations in the growth direction play a pivotal role in preventing simple columnar structure growth or structures that evolve systematically as the film grows. In addition, the anomalous scattering measurements confirm that the metal atoms (Fe or Mo) are the source of the inhomogeneity, with the Ge atoms distributed homogeneously. A method for using these measurements to determine the compositions of the phase-separating species has been developed. The results indicate phase separation into end-point compositions of amorphous Ge and an intermetallic phase of composition close to ${\mathrm{FeGe}}_{2}$ or ${\mathrm{MoGe}}_{3}$. Finally, by manipulating the deposition parameters, ${\mathrm{Fe}}_{\mathit{c}}$${\mathrm{Ge}}_{1\mathrm{\ensuremath{-}}\mathit{c}}$ films that have the same Fe composition can be grown to different states of phase separation. These results may help explain the difficulty workers have had in isolating the metal-insulator transition for these and other vapor-deposited amorphous alloys.

The phase structure of high-pressure-crystallized polyethylene

The phase structure of three linear polyethylene (PE) samples, crystallized from the melt at high pressure, has been studied by electron microscopy and high-resolution solid-state 13C n.m.r. spectroscopy. In general, three phases are required to account for the n.m.r. data: the lamellar crystalline phase, the crystalline-amorphous interphase and the amorphous phase. All three are present in high-molecular-weight samples but there is no amorphous phase for samples with lower molecular weights and large lamellar thicknesses. The amorphous phase appears when the ratio of the number-averaged extended molecular chain length ( X n) to the number-averaged crystalline stem length ( L n) exceeds two. High-pressure-crystallized materials differ from those crystallized at atmospheric pressure in that the mass fraction of the amorphous phase does not exceed 0.05; the thickness of the crystalline-amorphous interphase reaches 8.0 nm for material with the highest molecular weight, a value which is considerably larger than those reported for samples crystallized at atmospheric pressure or which have been estimated theoretically. Extraordinarily long 13C spin-lattice relaxation times have been found: a figure of T 1C = 7000 s, higher than any previously reported, for the highest-molecular-weight sample is still less than would be expected from the large lamellar thickness. In consequence, this relaxation is attributed to molecular motion in the vicinity of the crystal defects; this is in addition to 13C spin diffusion to the non-crystalline region, occurring with a shorter T 1C. The discrepancy between the observed and calculated values of T 1C increases as the molecular weight falls in those samples for which the crystal stem lengths exceed the extended molecular lengths. For these, the unexpectedly shorter T 1C is attributed to defects such as methyl end-groups within the crystalline regions.