To test whether the established B/Cl ratio (0.2414 mg·kg–1·‰–1) derived for the open ocean is applicable to coastal systems globally, we measured the B concentration and salinity of seawater samples (n = 245) collected from the coastal waters of the Yellow, East China, and East Seas. We also measured the B concentration of freshwater samples (n = 10) from six rivers discharging into these seas and analyzed along with the B concentrations of other rivers globally, obtained from literatures. Our results show that the supply of riverine B (2–6 μmol B/kg) did not discernably alter the B/Cl ratio in most estuarine systems. The impact of this riverine B input on the interconversion between total alkalinity ([HCO3–] + 2[CO32–] + [B (OH)4–] + others) and carbonate alkalinity ([HCO3–] + 2[CO32–]) is less than a few micromoles per kilogram for most estuarine and coastal systems, although exceptions in the Baltic Sea do occur.

Following heat treatments of Pt encapsulated Ni80Cr20 thin films on silicon substrates at temperatures ranging from 300 to 500 °C, it has been discovered that Cr atoms segregate at the original NiCr/Si interface to form an amorphous layer, while Ni atoms diffuse into the Si to form Ni silicide. The Cr-rich amorphous layer acts as a ‘‘semipermeable membrane’’ which selectively passes Ni to form a very uniform NiSi layer.

PACS 72.15.Eb, 75.20.Ck, 75.47.Pq, 75.50.Gg Mn-Zn ferrites among iron oxide magnetic materials are widely used for high frequency magnetic core materials because they have low power loss and high initial permeability up to several MHz range. In this paper, the small amount of Nb 2 O 5 and TiO 2 were added to the Mn-Zn ferrites to study their effects on electrical and magnetic properties of the Mn-Zn ferrites. The amount of Nb 2 O 5 and TiO 2 was in the range of 0 ~0.02 wt% and 0-0.6 wt% respectively. The effects of additives were studied in view of change in core loss, maximum flux density, specific resistivity, inductance and permeability. Adding both TiO 2 (0.6wt%) and Nb 2 O 5 (0.01wt %), the core loss was lowest. It was decreased to 50% at 500 kHz. In this case, the maximum flux density was depressed by only 2%. This means that the additives could drastically decrease the minimum core loss with penalty of a little decrease in magnetic flux density. IntroductionIn general, the core loss of magnetic core materials increases when they are used at high frequency. This fact limits the efficiency of magnetic core materials. This requires the development of iron oxide magnetic materials with low core loss. For transformers, especially, magnetic core materials with high magnetic flux density, high initial permeability and low core loss are essential. Mn-Zn ferrites among iron oxide magnetic materials are used for high frequency magnetic core materials because they have relatively low power loss and high permeability up to several MHz [1]. However, their magnetic characteristics such as core loss are not good enough for high frequency operation. To improve electro-magnetic properties of Mn-Zn ferrites, many researchers tried to add additives to them. There are three approaches for these additives. First one is to segregate additives on grain boundaries to increase electrical resistance. It was known that core loss decreased adding Nb 2 O 5 up to 0.03 wt % to Mn-Zn ferrites. By precipitation along grain boundaries, core loss was decreased by 30% at 100 kHz/100 mT compared to the Mn-Zn ferrites without additives. Other additives such as V 2 O 5 , B 2 O 3 , SiO 2 and ZrO were also reported to decrease the core loss [2]. Second one is to form solid solution and then improve magnetic characteristics. Additives such as TiO 2 , SnO 2 , Al 2 O 3 , CoO and MgO were tried for this purpose [3,4]. Third one is to form liquid phase and then enhance grain growth. Additives such as Bi 2 O 3 and In 2 O 3 were used for this purpose [5]. As mentioned above, one additive was used for each case. It will also be very interesting to add two additives simultaneously. Therefore, in this paper, the additive effects of both TiO 2 and Nb 2 O 5 on electromagnetic properties of Mn-Zn ferrites were investigated.

Sputter-deposited Ni,,C!r,, films on sputter-cleaned Si substrates contain an amorphous layer at the substrate/film interface whose composition is a mixture of all the elements present at the interface. Subsequent thermal processing at 300 "C for 30 min produces a new segregated Cr-rich amorphous layer as Ni atoms preferentially diffuse through and react with the initial amorphous layer and the silicon substrate. Further annealing results in the growth of uniform nanoscale Nisi layers, as long as the segregated a layer is sustained. The amorphous layers eventually crystallize at -500 "C and Kirkendall voids are observed at 550 "C. Whereas the formation of intermixed amorphous layers from metal-metal or metal-silicon systems has been reported by several authors, the segregated amorphous layer arising out of the interdiffusion and reaction between a metal alloy and Si is of both fundamental and technological interest due to its thermal stability and ability to control the silicide growth. In this work, we describe the evolution of both kinds of amorphous layers, i.e., intermixed and segregated, so as to elucidate their origins. The evolution of the two a layers is also observed when monolayers of Pt are introduced prior to NiCr deposition. In this case, the growth of the segregated amorphous layer is retarded and it dissolves earlier during thermal annealing.