Degradation effects in polymer light emitting devices due to heat treatment

The characteristics of polymer light emitting diodes (PLEDs) (ITO/PPV/Ca) depend strongly on the conditions during preparation and operation. We studied the effects of heat treatment (during and after preparation) of PLEDs with OC₁C₁₀-PPV as active layer. PLEDs showed a reduction of both the current and the light output to 40 % after annealing for only 30 min at 65 °C. Effects on I-V characteristics were studied by measuring single carrier devices (hole- and electron-dominated devices). The current reduction after heat treatment can be ascribed to degradation of the ITO/PPV and the Ca/PPV interfaces.     

Methods for correlating T1ρ and FID components in wideline H NMR studies of motionally heterogeneous polymer systems

We address the problem of correlating the observed FID and T_1ρ components in wideline ¹H relaxation measurements of motionally heterogeneous polymers, and show that different methods of data treatment can highlight different aspects of the correlations present. For a sample of polypropylene we find that the T_1ρ relaxation behaviour is driven by relaxation associated with the intermediate FID component, which strongly suggests a motionally inhomogeneous amorphous region in the sample.     

Serial MR imaging of intracranial metastases after radiosurgery

Purpose: To evaluate the spatiotemporal evolution of radiosurgical induced changes both in metastases and in normal brain tissue adjacent to the lesions by serial magnetic resonance (MR) imaging. Methods and Materials: Thirty-five intracranial metastases of different primaries were treated in 25 patients by single high-dose radiosurgery. MR images acquired before radiosurgery were available in all patients. Sixty-three follow-up MR studies were performed in these patients including T₂- and contrast-enhanced T₁-weighted MR images. The average follow-up time was 9 ± 5 months (mean ± standard deviation [SD]). Based on contrast-enhanced T₁-weighted MR images, tumor response was radiologically classified in the following four groups: stable disease was assumed if the average tumor diameter after treatment did not show a tumor shrinkage of more than 50% and an increase of more than 25%, partial remission as a shrinkage of tumor size of more than 50%, a disappearance of contrast-enhancing tumor as a complete remission, and an increase of tumor diameter of more than 25% as tumor progress. Moreover, we analysed signal changes on T₂-weighted images in brain parenchyma adjacent to the enhancing metastases. Results: The overall mean survival time was 10.5 ± 7 months, with a 1-year actuarial survival rate of 40%. Stable disease, partial or complete remission of the metastatic tumor was observed in 22 patients (88%). Central or homogeneous loss of contrast enhancement appeared to be a good prognostic sign for stable disease or partial remission. This association was statistically significant (p < 0.05). Three patients (12%) suffered from tumor progression. In eight patients (32%) with stable disease or partial remission, signal changes on T₂-weighted images were observed in tissue adjacent to the contrast enhancing lesions. A progression of the high signal on T₂-weighted images was seen in seven of the eight patients between 3 and 6 months after therapy, followed by a signal regression 6–18 months after irradiation. Conclusion: MR imaging is a sensitive imaging tool to evaluate tumor response as well as the presence or absence of adjacent parenchymal changes following radiosurgery. Loss of homogeneous or central contrast enhancement on Gd-enhanced MR images appeared to be a good prognostic sign for tumor response. Tumor shrinkage seems not to be dependent on time. In addition, most cases of radiation induced changes in normal brain parenchyma observed on T₂-weighted images seem to be self limited.     

Efficient computation of the exponential operator for large,sparse, symmetric matrices

In this paper we compare Krylov subspace methods with Chebyshev series expansion for approximating the matrix exponential operator on large, sparse, symmetric matrices. Experimental results upon negative‐definite matrices with very large size, arising from (2D and 3D) FE and FD spatial discretization of linear parabolic PDEs, demonstrate that the Chebyshev method can be an effective alternative to Krylov techniques, especially when memory bounds do not allow the storage of all Ritz vectors. We also discuss the sensitivity of Chebyshev convergence to extreme eigenvalue approximation, as well as the reliability of various a priori and a posteriori error estimates for both methods. Copyright © 2000 John Wiley & Sons, Ltd.