Ten pieces of the middle age “Jomon-doki”, a prehistoric unglazed earthenware, which were found by C. Kawashima in the ground of the Tokio University of Engineering, were finely ground and examined for the change of weight on heating up to about 770°C by a thermobalance, while the contents of H₂O, C, CO₂ were determined for fine powder of five pieces of the earthenware. For the latter experiment, the weights. of H₂O and CO₂ given off from the samples on heating at 107°C, 400°C, and 750°C in a current of dry air were determined. The content of H₂O was determined for the ranges of 107-400° and 400-750°C. The content of C was estimated from the amount of CO₂ evolved at 400°C since the authors confermed in a former study that amorphous C burns at about 350°C, whereas the CO₂ given off at 750°C was attributed mainly to carbonates in the samples.
Except a loose immature sample, the curves showing the relation between the temper ature of heating and the loss of the weight were not like those of usual clays, showing no characteristic slopes.
The amounts of dehydration of the dry samples were 5.6-7.8% for 107-400°C, 1.8-2.8% for 400-750°C, and 7.5-10.5% for 107-750°C. The contents of free C and CO₂ were found to be 0.6-1.4% and 0.7-2.0% respectively.

View full abstract

Several poor plastic or non plastic refractory materials such as, silica, alumina, magnesia and steatite were studied on their physical properties, effected by their various moulding pressures and quantities of the binder. In this paper, however, some experimental data with only a steatite are reported.
The steatite used for electrical insulating materials of vacuum tubes of many kinds was a non plastic material. Therefore, how to mould or how to form the insulators was a very important problem. The raw material was temporarily given various quantities of some organic binders for the purpose of forming green products, and then pressed in an iron mould by a hydraulic press under various pressures. Mechanical strength, thermal expansion coefficients, porosities, densities and moulding pressures of the burned and green products were measured, and their relations to each of them were investigated and explained under some assumptions of distribution of moulding pressures in the specimens, with an aid of some experimental formulas applicable to testing materials, and summarised as following:
1 When moulded with no binder, the transvers strength of the burned products could be linearly increased by being increased the moulding pressure.
2 In the case of an excess quantity of binder was added, the transvers strength of the burned products showed a rapid approach to the saturating constant value with increasing the moulding pressure.
3 In the case of a moderate quantity of binder was used, it was found that there were maximum and minimum values in the process of increasing the moulding pressure.
4 The transvers strength of the burned products was found to have a strength of nearly maximum value when they had a saturated mechanical strength in the green state.
5 Expansion coefficients of the burned products were constant independently of their moulding pressures, whiles the cristobalite effect at about 270°C (extraordinary expansion due to the transition of α→β cristobalite) had a maximum value for a certain moulding preasure.
6 A decrease in porosities were produced by a decrease in the quantity of the binder and an increase in the moulding pressures. The mechanical strength of the burned products of lower porosity was larger than that of higher one.
7 The expansion of the green products when knocked out freshly from the mould could be decreased by being increased their moulding pressures and the quantities of the binder.

View full abstract

The general glass composition being the 72% SiO₂, (12-x) CaO, x Fe₂O₃, 16Na₂O, the iron oxide being added to the batch as ferric oxide, and a study has been made as to the variation of the state of oxidation of the iron oxide content and the variation of shades of colours, as well as the transmission curves of the glass in case the value of x being changed in various proportions.
The melting has been carried out in a 20 Kin clay pot (about 20 Kg capacity), the batch being charged in at a temperature in all cases of 1400°, and the melting operation continued of the temperature of 1400-1450°C.
(A) The glasses have been analysed, the ferrous and ferric oxide both being determined. The results are given in the following table.
From the above result it will be observed that the percentage ratio of FeO/FeO+Fe₂O₃ of the resultant glasses depends on the content of iron oxide first added if melting be performed under substantially similar condition, that is, similar melting temperature, time of heating, the condition of the furnace. The percentage ratio of FeO/FeO+Fe₂O₃ in found large in case the content iron oxide first added be small, and it becomes gradually smaller if the content added be increased. However, there is a tendency that the said ratio gradually reaches a certain definite value when the content of iron oxide exceeds an amount corresponding to about 2% in the glass. Jebsen-Marwedel and A. Becker dissociation theory as to ferric oxide will be explain this fact.
(B) The color of glasses containing iron oxide varies according to the proportion of FeO/FeO+Fe₂O₃ contained therein and it also varies with a total iron content.
By this experiment it has been found that when the content of iron oxide is small the color is light bluish, and as the iron oxide content increases it becomes greenish, then changing to dark green, finally it becomes brownish.
I have tested the transmission curve of the glasses and found that it varies with the iron oxide contained in the glasses, the transmission of glasses will be lowered as the iron content in glasses is increased, the limit showing a maximum transmission being shifted toward the red in the spectrun.
The limit of the wave length the glass transmitted will be shifted to one having longer wave length as the amount of iron oxide contained in glasses is increased. It has been known that this tendency is so considerable even for a slight variation in the iron content in case the iron oxide content in the glasses is small, and glasses containing a larger proportion of iron oxide are not so much influenced by variation of the iron oxide contents.

View full abstract

A study was made of the effect of some special auxiliary constituents of portland cement on the clinkering temperature of its raw mixture and on its quality. The basal raw mixture was compounded from chemically pure calcium carbonate, silica, alumina, and ferric oxide so as to give a clinker consisting of 23.44% SiO₂, 4.69% Al₂O₃, 3.12% Fe₂O₃, and 68.75% CaO. The theoretical contents of the auxiliary constituents of clinkers were as follows: 1%, 3%, and 5% MgO; 0.5%, 1%, and 3% Cr₂O₃; 1% Ti₂O₃, 1% Mn₂O₃, 1% Mo₂O₃; 3% MgO+0.5% R₂O₃ (R=Cr, Mn, Ti, Mo). Extra amounts of lime corresponding to the assumed compounds, 2CaO·R₂O₃, were added to the raw mixtures for Cr, Mn, Ti, and Mo. Some of small brickets made of the raw mixtures were fired at different temperatures in a gas-fired muffle kiln. The clinkers thus obtained were analyzed for the content of free lime in order to compare the fluxing action of the auxiliary constituents. The remainder of the brickets were fired at temperatures ranging from 1390° to 1460°C so as to obtain clinkers containing about 0.5% of free lime. An observation was made on the relation between the auxiliary constituents and the color of the clinkers. The cements prepared from the clinkers and 2% of gypsum were tested for the crushing strength of 1:3 mortars by a small-piece method. The test pieces were aged for 3 to 28 days in water, a set of them being heated at temperatures between 139° and 140°C for 6 hours in an autoclave after 28 day aging in water.
The results seem to justify the following conclusions:
1. MgO acts as a strong flux at comparatively low temperatures, although its action diminishes at clinkering temperatures. The action is only slightly affected by its amount. A small content of Cr₂O₃ has no effect on the clinkering of the raw mixtures while amounts more than 0.5% retard the combination of CaO strikingly. The effect of Ti₂O₃, Mn₂O₃, and Mo₂O₃ is not marked.
2. Each clinker has a color peculiar to its auxiliary constituent. The basal clinker has a yellow color which changes to a greyish or greenish black characteristic to portland cement clinkers by an addition of MgO or Mn₂O₃. Ti₂O₃ and Mo₂O₃ have no effect while Cr₂O₃ produces a bright green. The color of these clinkers is affected by the nature of kiln atomosphere, e.g. the clinkers with MgO, Mn₂O₃, or Cr₂O₃ are brown when fired with a reducing flame.
3. MgO increases long age strength of the cement and the effect increases with its content. A small amount of Cr₂O₃ such as 0.5% produces a high early strength cement, although its higher content reduces the effect and increases long age strength. Both Mn₂O₃ and Mo₂O₃ seem to increase early and long age strengths while Ti₂O₃ favors only early strength. However the effect of these three constituents is not manifest when MgO is contained. The strength is affected by the nature of flame, e.g. early strength of the cement with 1% Mn₂O₃ is obviously increased by firing its clinker with a reducing flame.

View full abstract