Search

The CaO-ZnO-B2O3-P2O5 glass-ceramics with various amounts of TiO2 as crystallized agent were prepared by solid-state method. The effects of TiO2 addition on the crystallization kinetics, microstructure and dielectric properties were systematically evaluated. Adding moderate content of TiO2 can obviously promote crystallization and improve the microstructure of the glass-ceramics, due to the reduction of crystallization activation energy. It is also found that with the content of TiO2 increasing, the dielectric constant (er) and quality factor (Q×f) of the glass-ceramics firstly increase and then decrease, however, temperature coefficient of resonance frequency (τf) moves to positive value. Meanwhile, the sintering temperature also has a direct influence on the dielectric properties of the glass-ceramics. The glass-ceramic with 0.5 wt% TiO2 sintered at 760 °C for 2 h possesses good dielectric properties of er = 4.75, Q×f= 23500 GHz, τf = −17 ppm/°C, which indicates that the as-prepared glass-ceramic is a promising candidate for low temperature co-firing ceramics.

High quality multi-walled carbon nanotubes (CNTs) grown at high density using a low temperature growth method are used as an alternative material to polymer sizing and is utilised in a series of epoxy composites reinforced with carbon fibres to provide improved physical and electrical properties. We report improvements for sizing-sensitive mechanical and physical properties, such as the interfacial adhesion, shear properties and handling of the fibres, whilst retaining resin-infusion capability. Following fibre volume fraction normalisation, the carbon nanotube-modified carbon fibre composite offers improvements of 146% increase in Young’s modulus; 20% increase in ultimate shear stress; 74% increase in shear chord modulus and an 83% improvement in the initial fracture toughness. The addition of CNTs imparts electrical functionalisation to the composite, enhancements in the surface direction are 400%, demonstrating a suitable route to sizing-free composites with enhanced mechanical and electrical functionality.

Micro- and nano-scale crystalline indium-tin-oxide (c-ITO) patterns fabricated from amorphous ITO (a-ITO) thin films on a glass substrate using a (low NA 0.26) femtosecond laser pulse that is not tightly focused are demonstrated. Different types of c-ITO patterns are obtained by controlling the laser pulse energies and pulse repetition rate of a femtosecond laser beam at a wavelength of 1064 nm: periodic micro c-ITO dots with diameters of ~1.4 μm, two parallel c-ITO patterns with/without periodic-like glass nanostructures at a laser scanning path and nano-scale c-ITO line patterns with a line width ~900 nm, i.e. ~1/8 of the focused beam׳s diameter (7 μm at 1/e 2 ).

Introduction Plasma concentration monitoring is commonly used to adjust immunosuppressant dosage in transplant recipients, but adjustment is often based on clinical experience rather than rigorous quantitative indicators. Methods We examined the effect of mycophenolate mofetil (MMF) dosage on graft survival by pathologic and immunologic analysis of 88 kidney recipients who were given a postoperative immunosuppressive regimen of tacrolimus (FK506), MMF, and corticosteroids. Patients were given a conventional dosage (≥1.5 g/d; n = 40) or a reduced dosage (n = 48) of MMF owing to postoperative adverse side effects. Results The reduced-dose group included patients given low doses (≤1.0 g/d; n = 27), ultra-low doses (≤0.5 g/d; n = 15), and those who discontinued MMF (n = 6). The dose reduction group had increased acute rejection, chronic rejection, and graft dysfunction, poorer pathologic scores, and increased cell infiltration of graft tissue (CD4, CD8, CD68, and CD138 positivity) and expression of interleukin-2R and HLA-DR. Finally, hazard analysis indicated that patients given low doses and ultra-low doses of MMF had poorer long-term kidney grafts survival (hazard ratios of 1.52 and 1.78, respectively). Conclusions These results indicate the importance of using an appropriate dosage of MMF in kidney transplant recipients.

A novel CuO-ZnO-B2O3-Li2O glass-ceramic was prepared by solid phase method, and. Its phase transition, sintering and dielectric performance was studied. No crystalline phase is observed for the specimen sintered at 560 °C. However, the samples sintered at 580 °C and 600 °C, consist of ZnB4O7 and Cu2B4O7. And the Li2CuB4O8 appears with further increasing temperature. The er and τf change slightly but the Q × f value alter greatly with firing temperature. The glass-ceramic fired at 620 °C possesses optimum properties with er = 3.33, Q × f = 17,724 GHz and τf ≈ 0 ppm/°C. Furthermore, the as-prepared glass-ceramic exhibits good co-firing compatibility with Ag and Al electrode. These properties qualify the dielectric material as promising alternatives for wireless communication applications.

In this study, a method for the fabrication of microstructures on the surface and inside Foturan glass by femtosecond laser-induced modification was developed. This technique was followed by heat treatment to crystallize the modified area, and the specimen was then placed in an 8% HF acid solution for chemical etching. The fabricated microstructures were observed using scanning electron microscopy (SEM). The results demonstrated that the etching time is an important parameter in the fabrication of microstructures on Foturan glass. An example of a tapered U-shaped microchannel with a minimized neck diameter of about 5 μm at the central point for cell detection is presented.

Centerline heat transfer measurements along two opposite ribbed walls in three rotating rectangular ducts roughened by 45° staggered ribs with channel aspect ratios (AR) of 1:1, 2:1 and 4:1 are performed at Reynolds ( Re ), rotation ( Ro ) and buoyancy ( Bu ) numbers in the ranges of 5000–30,000, 0–2, and 0.005–8.879, respectively. These channel geometries are in common use as the internal cooling passages of a gas turbine rotor blade and the tested Ro and Bu ranges are considerably extended from the previous experiences. This study focuses on the heat transfer characteristics in response to the change of AR under the parameter ranges examined. With zero-rotation ( Ro = 0), the local Nusselt numbers ( Nu 0 ) along the centerlines of two opposite ribbed walls increase as AR increases due to the increased rib-height to channel-height ratio. The Bu impact on heat transfer appears to be AR dependent, i.e. the increase of Bu elevates Nusselt number ratios Nu / Nu 0 in the square channel but impairs heat transfer in the rectangular channels of AR = 2 and 4. Acting by the Coriolis effect alone, all the leading edge Nu values in the present Ro range are lower than the zero-rotation references but started to recover as Ro increases from 0.1 in the channels of AR = 1, 2 and from 0.3 in the channel of AR = 4. The trailing edge Nu / Nu 0 ratios increase consistently from unity as Ro increases but their responses toward the increase of AR are less systematic than those found along the leading edge. The above findings, with the aids of extended Ro and Bu ranges achieved by this study, serve as the original contributions for this technical community. The Nu / Nu 0 ratios in the rotating channels of AR = 1, 2, and 4 fall in the ranges of 0.6–2.2, 0.5–2.7, and 0.5–2.1, respectively. A set of heat transfer correlations is derived to represent all the heat transfer data in the periodically developed flow regions of three rotating ducts.

Membranous nephropathy is an autoimmune-mediated glomerulonephritis and a major cause of nephrotic syndrome. We studied the kinetics of adaptive immunity in the pathogenesis of membranous nephropathy in T1/T2 double transgenic mice (T1/T2 TG mice) that express human Thy1 protein under the control of interferon-gamma (INF-gamma) and mouse Thy1.1 protein under the control of interleukin (IL)-4. Nephropathy was induced by cationic bovine serum albumin. We found that splenocytes expressed a progressive Th2 response and a subsequent compensatory T-helper 1 (Th1) response, with a gradual augmentation of IL-4-producing Th2 cells and INF-gamma-producing Th1 cells. Increased Th2 marker expression was seen in peripheral blood and kidney cells, with the immunoglobulin G1 (IgG1) antibody isotype predominant in the serum and kidneys. We found that CD8+ T cells contribute more to the augmented INF-gamma production than CD4+ T cells. Moreover, CD19+ B cells demonstrated a greater production of IL-4 than the CD4+ T cells. Cytokine-related gene expression in kidneys and splenocytes showed an upregulation of proinflammatory Th1 and Th2 cytokines. Th2 cells but not Th1 cells were significantly correlated with serum cholesterol and proteinuria. Our study shows that both peripheral and renal immune reactions are strongly polarized toward Th2-type immune responses during the course of membranous nephropathy. The T1/T2 mouse model may help decipher the kinetic changes of adaptive immunity in glomerulonephritis.

We investigate the effect of conversion from a cyclosporine (CsA) based-regimen to a tacrolimus (FK506)-based regimen with respect to graft renal function induced by chronic allograft nephropathy (CAN). Thirty-one patients with a histological diagnosis of CAN were included after other causes of chronic graft dysfunction had been excluded. Conversion to FK506 was undertaken at an initial dose of 0.15 mg/kg/d, which was subsequently adjusted to maintain FK506 whole blood trough levels between 5 and 10 mug/L. The rate of decline of renal function before and after the FK506 conversion was represented by regression lines (slope) of the reciprocal of serum creatinine versus time. To evaluate the effect of conversion on allograft function, we gathered data on serum lipids, blood glucose, proteinuria, and hypertension. When postconversion slopes were compared to preconversion slopes for each patient, 20 patients (64.5%) showed positive regression lines and four patients (12.9%), less negative. Seven patients (22.6%) displayed an increased rate of decline in renal function with regression lines becoming more negative. FK506 was associated with a significant decrease in lipid levels, proteinuria, and hypertension. No patient returned to dialysis at the end of the 36-month follow-up. Conversion from a CsA-based regimen to a tacrolimus-based regimen was an effective alterative for salvage of patients with abnormal graft renal function induced by CAN.

The manufacturing process for a new Printed Circuit Board (PCB) design is often instable and might generate a number of defects during the complicated production process. Defects reduce the yield rate and increase the production costs. Although skilled engineers can predict the possible defect items for a new PCB product, this approach requires strong engineering experience and is time consuming. To conquer this problem, this research applies case-based reasoning (CBR) methodology to develop a defect prediction system for new PCB products. In the CBR system, each case is represented using the design specifications, defect items and corresponding costs. A vantage-based case indexing mechanism is developed to accelerate the case retrieval efficiency. In addition, a reasoning algorithm that considers the defect cost is proposed to infer the defect items that are interesting to PCB manufacturers. The system performance is analyzed to show the efficiency and accuracy of the proposed system. A practical implementation using a case-base provided by a PCB manufacturer is demonstrated.

Previous research in finite element formulation of flexible mechanisms usually neglected high order terms in the strain-energy function. In particular, the quartic term of the displacement gradient is always neglected due to the common belief that it is not important in the dynamic analysis. In this paper, we show that this physical intuition is not always valid. By retaining all the high order terms in the strain-energy function the equations of motion naturally become non-linear, which can then be solved by the Newmark method. In the low-speed range it is found that the dynamic responses predicted by non-linear and linear approaches indeed make no significant difference. However, when the rotation speed increases up to about one-fifth of the fundamental bending natural frequency of the connecting rod, simplified approaches begin to incur noticeable error. Specifically, for a connecting rod with a slenderness ratio of 0·01 the conventional simplified approaches overestimate the vibration amplitude almost 10-fold when the rotation speed is comparable to the fundamental natural frequency of the connecting rod. Therefore, non-linear finite element formulation taking into account the complete non-linear strain is needed in analyzing high-speed flexible mechnisms with slender links.

A general approach is proposed for determination of direct relations between: (i) the principal curvatures and directions of two surfaces being in continuous tangency along a line at every instant, (ii) the normal curvatures and torsions of two surfaces with conditions of tangency mentioned above. The surface of relative normal curvature, its principal curvatures and directions, and the torsions are determined as well. The instantaneous contact ellipse for two elastic surfaces is determined. The proposed approach is illustrated with a numerical example of hypoid gears.

The authors analyze the tooth surface contact and stresses for double circular-arc helical gear drives. The geometry of such gear drives has been represented by the authors in their previous paper [1]. The proposed approach is based on application of (i) computerized simulation of meshing and contact of loaded gear drives, and (ii) the finite element method. Load share between the neighboring pairs of teeth is based on the analysis of position errors caused by surface mismatch and elastic deformation of teeth. The authors have investigated the conditions of load share under a load and determined the real contact ratio for aligned and misaligned gear drives, respectively. Elastic deformation of teeth and the stress analysis of the double circular-arc helical gears are accomplished by using the finite element method. The finite element models for the pinion and gear are constructed, respectively. Contact pressure is spread over elliptical area. The stress analysis for aligned and misaligned gear drives, respectively, has been performed. The numerical results have been compared with those obtained by other approaches.

Interfacial reactions between a W thin film and a single-crystal (001) β-SiC substrate on rapid thermal annealing from 600 °C to 1100 °C for 60 s were investigated by backscattering spectrometry, X-ray diffraction, secondary ion mass spectrometry and cross-sectional transmission electron microscopy. Backscattering spectrometry shows that W reacts with SiC at 950 °C. The product phases identified by X-ray diffraction are W5Si3 and W2C. At 1100 °C no more unreacted W is detected. Current-voltage measurements show that ohmic contacts are already obtained on as-deposited W. Contact resistivity measured using the circular transmission line model is about 10−3 Ω cm2. Thermodynamic studies of the solid phase stability in the ternary WSiC system help us to understand the chemical stability of W thin film.

Contact resistivities of as-deposited and annealed Pt, Re and Ta films on n-type single-crystalline β-SiC(001) are characterized using a circular contact pattern and the circular transmission-line model method. The β-SiC substrates used in the experiment are n-type doped either non-intentionally to a carrier concentration of about 10 17 cm −3 , or by nitrogen implantation and annealing to a concentration of 5 × 10 19 cm −3 . The effect of a finite resistance along the circular contact rings on the measured potentials is corrected with a resistance network model. On the non-intentionally doped β-SiC substrates, Pt contacts are non-ohmic regardless of the heat treatment. The as-deposited Ta and Re contacts are ohmic with contact resistivities of 5 × 10 −5 Ω cm 2 and 4 × 10 −4 Ω cm 2 respectively. Upon annealing at 500 °C for 30 min, the resistivity of Ta increases slightly while that of Re decreases slightly. Both Ta and Re contacts become non-ohmic by annealing at 900 °C for 30 min. The as-deposited Ta, Pt and Re contacts are all ohmic on the nitrogen-implanted β-SiC substrate. The contact resistivity of the as-deposited Ta contact is the lowest and in the order of high 10 −7 Ω cm 2 , stays about the same at 500 °C and degrades to 4.3 × 10 −6 Ω cm 2 at 1000 °C. The as-deposited Re contact has the highest contact resistivity of 1 × 10 −4 Ω cm 2 but it improves to 1 × 10 −5 Ω cm 2 upon annealing at 900 °C. The contact resistivity of the as-deposited Pt contacts is 6 × 10 −6 Ω cm 2 and increases to 1 × 10 −5 Ω cm 2 at 500 °C. After annealing at 900 °C for 30 min, the Pt contact on the nitrogen-implanted β-SiC is no longer ohmic. The results are compared with the reactions that take place in those systems.

We deposited TaN films by reactive r.f. sputtering from a Ta target with an N 2 Ar gas mixture. Alloys over a composition range 0–60 at.% N have been synthesized. We report on their composition, structure and electrical resistivity before and after vacuum annealing in the temperature range 500–800 °C. We found that the film growth rate decreases with increasing ratio of the nitrogen flow rate to the total flow rate, while the nitrogen content in the films first increases with the N 2 partial flow rate and then saturates at about 60 at.%. B.c.c.-Ta, Ta 2 N, TaN and Ta 5 N 6 appear in succession as the nitrogen content rises, with Ta 2 N being the only single-phase film obtained. The atomic density of the films generally increases with the nitrogen content in the film. Transmission electron micrographs show that the grain size decreases from about 25 to 4 nm as the nitrogen concentration increases from 20 to 50 at.%. The Ta 2 N phase can exist over a wide range of nitrogen concentration from about 25 to 45 at.%. For as-deposited films an amorphous phase exists along with polycrystalline Ta 2 N in the center portion of that range. This phase crystallizes after vacuum annealing at 600 °C for 65 min. A diagram of stable and metastable phases for TaN films based on X-ray diffraction and transmission electron microscopy results is constructed. The resistivity is below 0.3 m ohms cm for films with 0–50 at.% N and changes little upon vacuum annealing at 800 °C.

The mechanism of stress corrosion cracking (SCC) of Cu-30Au in 0.6 M NaCl was investigated by a series of experiments in which samples were de-alloyed by potentiostatic anodic polarization at 475 mV (Ag/AgCl) and at zero applied stress (i) for varying lengths of time (10 to 30 min) and then impact bent, and (ii) for 30 min followed by a period of time (5–30 s) at the open circuit potential and then impact bent, and (iii) for 30 min followed by anodic polarization at +150 mV (Ag/AgCl) for a period of time (5–30) and then impact bent. The results indicate that de-alloying at zero applied stress produces a surface layer that is capable for a brief period of time of inducing intergranular cleavage failure of the normally ductile substrate. In addition, samples were de-alloyed and simultaneously stressed at various nominal values (0.5–1.7 σ y ). At low values of stress, failure occurred by brittle intergranular cracking, and at high values of stress, failure occurred by brittle transgranular cracking. The results indicate that the electrochemical contribution to the SCC of Cu-30Au in 0.6 M NaCl is the same for both intergranular stress corrosion cracking (IGSCC) and transgranular stress corrosion cracking (TGSCC) and that the mode of fracture is dictated by the magnitude of the applied stress.

The thermal reaction of sputter-deposited Ta thin films with polycrystalline CVD-grown diamond substrates at temperatures between 700°C and 1100°C for 1 h is investigated by MeV He backscattering spectrometry, X-ray diffraction, and scanning electron microscopy. The product phases are polycrystalline Ta2C at 900°C and TaC at 1000°C with an initially 135nm thick Ta film. The first phase formed conforms to the rule of Bene for metal-metal bilayers. No coexistence of Ta2C and TaC is observed during reaction. The second phase formed, TaC, is in thermodynamic equilibrium with carbon.

The performance of W40Re40B20 metallic amorphous thin film alloys as diffusion barriers between monocrystalline silicon substrates ( ) and aluminum or copper overlayers is reported. The films are deposited using d.c. and r.f. magnetron co-sputtering. Their crystallization temperature and resistivity are about 900°C and 200 μohmscm, respectively. The /W40Re40B20/Al and /W40Re40B20/Cu metallizations were vacuum-annealed at different temperatures for 30 min and characterized by 4He backscattering spectrometry, X-ray diffraction, scanning electron microscopy in combination with energy-dispersive analysis of X-rays, and electrical measurements on shallow junction diodes. The /W40Re40B20/Al structure was stable up to 500 °C and failed at 550 °C due to chemical reactions between the barrier and aluminum. A formation of cracks in the bilayer /W40Re40B20/Cu structure is responsible for its failure at 600 °C. The results demonstrate that it is fruitless to raise the crystallization temperature of an amorphous metallic film if its ability to react with the adjacent materials is not suppressed as well.

Polycrystalline CuInSe 2 /Cd(Zn)S heterojunction solar cells deposited on Corning 7059 or soda-lime glass are characterized structurally and chemically by scanning electron microscopy and transmission electron microscopy in conjunction with energy-dispersive analysis of X-rays. Scanning electron micrographs reveal rough and uneven surfaces and cross-sectional morphologies of the Cd(Zn)S and CuInSe 2 layers. The crystallography and defect structure of the individual Cd(Zn)S, CuInSe 2 and molybdenum layers are examined by conventional and high resolution transmission electron microscopy. The crystal structures for Cd(Zn)S, CuInSe 2 and molybdenum are wurtzite, chalcopyrite and b.c.c. respectively. The Cd(Zn)S layer exhibits stacking faults on hexagonal basal planes. Planar defects such as twins and stacking faults on {112} chalcopyrite planes are identified in the CuInSe 2 layer. The most significant features obtained from these cross-sections are (i) the lateral non-uniformity of the Cd(Zn)S and CuInSe 2 layers, (ii) the intimate bonding between these two layers, and an epitaxial relationship between grains of Cd(Zn)S and CuInSe 2 at the interface ({0001} Cd(Zn)S ∥ {112} CuInSe 2 ), and (iii) the presence of voids and fractures in the CuInSe 2 layer. A correlation between the formation of fractures and voids and the defect structure in CuInSe 2 layer, and the mechanical stresses induced by differential thermal contraction of the substrate/film assembly is discussed.

The mechanism of transgranular stress corrosion cracking (TGSCC) that was proposed by Edeleanu and Forty and Sieradzki and Newman recognized an often neglected aspect of TGSCC, namely, the brittle, cleavage-like appearance of the fracture surface. In this paper it is suggested that anodic processes produce a surface layer that induces cleavage of the normally ductile substrate. Demonstrated in experiments by Newman et al. on thin foils of {alpha}-brass. Foils that were fractured at 77{degrees}K without prior corrosion failed in a ductile manner as did foils that were corroded, dried, and then fractured in air. Thus, brittle fracture did to occur merely as a result of immersion in the aqueous solution, nor did it occur simply as a result of deformation at low temperature. In fact, failure occurred by local fracture at the surface where the alloy reacted with the environment followed by brittle crack propagation through the normally ductile substrate.

Semi-volatile organic contaminants such as pesticides, PAHs and PCBs, as well as their more volatile counterparts, are being found in ground and surface waters with increasing frequency. It has been shown that removal of the volatile organic contaminants such as carbon tetrachloride and trichloroethene can be performed by membrane-based solvent extraction using a non-volatile oil as the solvent. The membrane is microporous polypropylene, coated to prevent wetting of the membrane pores by the oil phase. Solutes volatilize from water into the gas-filled membrane pores and are transferred to and concentrated in the oil phase. The hollow fiber shape of the membranes leads to a high specific surface area, and a very rapid removal rate; limited in most cases only by diffusion of the contaminant through water to the membrane surface. In this paper the applicability of this technology for removal of less volatile compounds from water is assessed. It is shown that the pesticide lindane (γ-hexachlorocyclohexane) and two other isomers of hexachlorocyclohexane can be removed from water at a rate similar to that of the volatile species. A transport model presented earlier is updated to include transfer through the membrane by surface diffusion, a mechanism proposed to explain the rapid transfer of species of low volatility. The model predicts rapid removal from water of a wide range of organic compounds.

Thin films of amorphous Ta-Si-N alloys were deposited by reactive RF sputtering of a Ta5Si3 target in an Ar/N2 ambient. These alloy films were tested as diffusion barriers between Al and Si, as well as between Cu and Si. Electrical measurements on Schottky diodes and on shallow n+p junction diodes were used to evaluate the thermal stability of the 〈Si〉 /W48Si20N32(20 nm)/Ta36Si14N50(80 nm)/Al(1000 nm) metallization. The amorphous W48Si20N32 contacting layer was added to raise the Schottky barrier height of the metallization on n-type Si. Both the shallow junctions and the Schottky diodes are stable up to 700°C for 20 min (above the Al melting point of 660°C) which makes this material the best thin-film diffusion barrier on record. Furthermore, the same Ta36Si14N50 amorphous film maintains the integrity of the I–V characteristics of the shallow n+p junctions with the 〈Si〉 /TiSi2(30 nm)/Ta36Si34N50(80 nm)/Cu(500 nm) metallization up to 900°C for 30 min annealing in vacuum. The TiSi2 contacting layer was added to assure an ohmic characteristic of the contact. For comparison, the same shallow junctions with 〈Si〉 /Cu metallizations were shorted after annealing at 300°C.

Bilayers of a slightly copper-rich composition of copper and indium deposited (in that order) at room temperature by vacuum evaporation, electroplating, and r.f. sputtering, were analyzed by MeV 4 He backscattering spectrometry, scanning electron microscopy and X-ray diffraction before and after thermal annealing at 400 °C in vacuum. The surface morphology of as-deposited samples is roughest for evaporated samples and smoothest for samples deposited by sputtering with low r.f. power. All as-deposited samples contain copper and indium phases and the metastable CuIn compound. After 1 h at 400 °C, this metastable phase and that of copper disappear and are replaced by some copper-rich compound (Cu 9 In 4 , Cu 7 In 4 ) in all cases. Simultaneously, the surface morphology smooths out considerably for the initially rough samples. Bilayers of various Cu/In compositions were also prepared. The results are consistent with the absence of stable compounds outside the 26–38 at.% indium range, as indicated by published phase diagrams.

Riboflavin status was surveyed in adults aged 35-64 y in 65 counties (two communes per county) in the Peo- ple's Republic of China by erythrocyte glutathione reductase activity coefficients (EGRACs), by urinary excretion of nbo- flavin 4 h after the oral administration of 5 mg riboflavin (one commune per county), and by direct measurement of food in- take over a 3-d period (one commune per county). EGRAC data were highly correlated between sexes (p < 0.001) and be- tween neighboring communes within the same county (p < 0.001) and with riboflavin intake (p < 0.001). The urinary- riboflavin-load test data were not correlated either with EGRAC or with riboflavin intake. Approximately 90% of the survey subjects were found to be deficient by Western and Chi- nese reference standards and allowances. Upon reconsidera- tion ofthe Western experimental data originally used to estab- lish these recommendations, it is suggested that riboflavin al- lowances are set too high both in China and in Western countries. Am J C/in Nutr 1990;5 1:436-45.

1. 1. In vivo activation of cuticular phenoloxidase (PO) of Florida spiny lobster is apparently mediated by some mechanism other than tryptic activity. 2. 2. Endogenously activated (EAPO) and trypsin activated (TAPO) forms of Florida spiny lobster PO show differences in their secondary structure as evidenced by circular dichroism spectropolarimetry. 3. 3. EAPO and TAPO forms of Florida spiny lobster possess identical antigenic determinants.

In situ electrochemical scanning tunneling microscopy (ESTM) was employed to observe the surface of silver immersed in 1M KCl before and after an oxidation-reduction cycle (ORC). The roughness created by ORC was found to relax quickly. During the first 2 min marked decreases in surface area occur that are accompanied by the disappearance of nanometer size features. The rate of change in surface area decreases significantly after this time. Further decreases in surface energy are brought about by the growth of crystal planes which become apparent during the first 3–5 min. Some of these planes appear to be low index planes {111}; these have the lowest surface energy. The results of the present study suggest that the time evolution of surface enhanced Raman scattering (SERS) following an ORC of silver can be contributed to by changes in atomic scale roughness (⩽ 30 A) as well as submicroscopic roughness (∼ 500 A).