Library

Notes: In this article, nitrogen-rich silicon nitride plasma deposited under conditions minimizing Si—H bonding is shown to possess extremely low bulk electron trapping rates which are as low or lower than plasma-deposited oxides produced using He dilution. The bulk trap density, measured by avalanche injection decreases as the rf power is decreased. The total charge trapped within these silicon nitrides reaches a saturation value determined by high field detrapping in thick nitride films.

Notes: Boron modulation-doped Si/SiGe heterojunctions have been grown by molecular beam epitaxy. The two-dimensional hole gas formed along the heterojunction, just inside the alloy, has a sheet density in the range 2–5×1011 cm−2 and a typical mobility at 5 K of 2000 cm2 V−1 s−1. An explanation for the magnitude of the mobility is sought by considering likely scattering mechanisms, namely those due to remote impurities, interface roughness, alloy disorder, and interface impurities. A self-consistent model is used to determine the sheet density in terms of structural and energy parameters and dopant concentrations in the heterostructure. It is shown that the presence of negatively charged impurities at the heterojunction provides the basis for a consistent interpretation of the experimental results.

Notes: A selected ion flow tube study has been carried out at 300 K of the reactions of some atomic and molecular positive and negative halogen ions with Cl2 and Br2 from which the rate coefficients k and ion product distributions have been determined. For the energetic F+ ion reactions, dissociative charge transfer is the dominant process, while for the Cl+ ions, only nondissociative charge transfer occurs. For the less energetic Br+ and I+ reactions, dihalogen molecular ions are important products. All these positive ion reactions proceed quite efficiently, i.e., the k are appreciable fractions of kc, their respective collisional rate coefficients, except for the reactions of Cl2 with the lower energy ions of the spin–orbit triplet of I+, i.e., I+(3P1,0), for which k∼0.07kc, this being due to the endothermicities of the reactions. The molecular ion Cl2+ undergoes rapid nondissociative charge transfer with Br2, a process which is, of course, endothermic for the reaction of Br2+ with Cl2 and so no reaction is observed. The less-energetic atomic negative ion reactions proceed—via atom exchange—in which the atomic negative ion of the reactant molecular species and a dihalogen molecule are produced. For those reactions that are exothermic, the k are, within error, equal to (2/3)kc, implying that they proceed via complexes which separate statistically back to reactants (1/3) and forward to products (2/3). Both the Br−+Cl2 and Cl−+Br2 reactions are somewhat less efficient (i.e., k〈2/3kc), a result of the slight endothermicities of the reactions. Of the molecular negative ion reactions, electron transfer is the major process in the Cl2− reaction with Br2, whereas the reaction of Br2− with Cl2 proceeds relatively slowly producing the triatomic ion BrCl2−.

Notes: Objectives 1. To compare the ultrasound biparietal diameter and crown-rump length of fetuses with and without Down's syndrome in the first half of pregnancy; 2. To investigate the effect of estimation of gestational age using either measure on the detection rate of serum screening for Down's syndrome.Design Matched case-control study. Cases were singleton Down's syndrome pregnancies with a biparietal diameter or a crown-rump length recorded. Five controls were matched to each case on: medical centre; the date of the ultrasound scan examination (within two years); gestational age measured as the number of days since the first day of the last menstrual period; and the ultrasound measure used (ie the biparietal diameter (the measure of choice), or the crown-rump length otherwise). If a woman had a serum screening test for Down's syndrome, the biparietal diameter or crown-rump length measurement had to be taken prior to the screening test so that the result of the test could not influence whether a scan was performed.Setting Ten antenatal screening centres in seven countries in Europe and North America.Subjects Two hundred and one women with singleton Down's syndrome pregnancies and 1005 women with unaffected singleton pregnancies.Results The median biparietal diameter of fetuses with Down's syndrome was identical to that among the controls (median difference 0.0mm, 95% confidence intervals (CI)–0.5 to 0.5mm). The estimates of gestational age based on biparietal diameter yielded a median gestational age less than that based on the women's last menstrual period: three days less for cases and two days less for controls; small but statistically significant differences probably reflected a minor systematic difference in the conversion of a biparietal diameter to a gestational age estimate. The median crown-rump length of fetuses with Down's syndrome was also identical to that among controls (median difference 0.0mm, 95% CI–1.5 to 2.0 mm). There was no significant difference between the median gestational age estimate based on crown-rump length and that based on the women's last menstrual period.Conclusion In antenatal screening for Down's syndrome the routine use of an ultrasound biparietal diameter or crown-rump length measurement to estimate gestational age will not adversely affect the detection rate. To avoid differences in gestational age estimates using the last menstrual period and the biparietal diameter influencing screening performance, separate medians should be derived for each serum marker using the two methods of estimating gestational age. The appropriate set of medians can then be used to calculate the multiple of the median value for each woman screened depending on the method used to estimate her gestational age.

Notes: The often-observed extreme asymmetry in the sidelobe spectra of integrated acousto-optic filters exacerbates cross talk between wavelength-multiplexed optical channels. We explain the source of this asymmetry in terms of a systematic even-order variation of the effective waveguide birefringence as a function of distance from the device center. Calculations show what degree of nonuniformity in Ti-stripe thickness, width, or diffusion temperature are required to account for typical asymmetries. We describe two experiments in which temperature gradients were applied to low-power acousto-optic filters in order to induce sidelobe suppression on either side of the optical transmission function, while greatly enhancing sidelobes on the opposite side.

Notes: Nanometer-scale wires cut into a Si/Si0.87Ge0.13 multiple quantum well structure were fabricated and characterized by using photoluminescence and photoreflectance at temperatures between 4 and 20 K. It was found that, in addition to a low-energy broadband emission at around 0.8 eV and other features normally observable in photoluminescence measurements, fabrication process induced strain relaxation and enhanced electron-hole droplets emission together with a new feature at 1.131 eV at 4 K were observed. The latter was further identified as a transition related to impurities located at the Si/Si0.87Ge0.13 heterointerfaces.