Using purified sand and the materials of reagent grade K₂O-Na₂O-SiO₂ glasses were melted in a platinum crusible.
The quenched samples were subjected to reheating by holding at a constant temperature in the range of 300°-525°C, and the change of the density was measured by floating method.
The data obtained were analysed in the light of the new theory posturated by the author (cf. J. Ceram. Assoc. Japan, 66, 117, 1958).
The density change ΔD may be represented by
ΔD=_??_p_ie^-t/φ_i=pe^-t/φ₄+qe^-t/φ₃+re^-t/φ₂+se^-t/φ₁+…
where t is the time, φ relaxation time, P, q, …are the constants whose absolute values increase monotonously with decreasing temperature, having the increments falling off below a certain temperature. φ′_s have the respective apparent activation energies and increase with decreasing temperature. The apparent activation energies, however, are possible to become more or less smaller with decreasing temperature.
The activation energy of the largest time constant, φ₄, have the value close to that of viscosity giving φ₄≅480 min at the temperature corresponding to 10^14.3 pois. Assuming that the part of the density change represented by this term takes place with the flow of the same structural units as in viscous flow it is possible to calculate the size of the units by applying the Eyring's formula with a result that the volume is about 15 ų containing 1-2 O^2- ions.
Around the strain point it may fairly be presumed that the mobility of separate ions would predominate that of molecular groups.
The other terms than the first correspond to the elastic after effect in the range of transformation temperature and below, being probably caused by the expansion, contraction, and rotation of silicate chain anions. There is a possibility of giving a wide distribution of comparatively low activation energies which take part in the density change at low temperatures.
In the range of composition covered by the present paper the amount of contraction by heating decreased with increasing alkali content, the trend which reversed at the alkali content of 20-25 mol per cent. This decrease of contraction may be interpreted by the obstructing effect of alkaline ions to the change of network structure, and the point of inflection would correspond to the same trend of the composition-density diagrams.
It is highly probable that the density change caused by the migration of alkaline ions would be less than 10^-3g/cm³. In mixed alkali-glasses, therefore, the influence of thermal history would be highly reservable owing to the small mobility of mixed alkaline ions, and the larger contribution of the place exchange of alkaline ions to the density change would bring about the larger thermal after effect.
The author modified his model representing the energy state of a glass structure by an assembly of small cubes of equal volume each being assigned the minimum free energy of the level, L₁, L₂, …as follows:
(1) L_i is composed of the levels l₁₁, l₁₂ having the values of free energy close to each other.
(2) l_ij are composed of l_ij1*, l_ij2*.
And the transformation L↔L corresponds, for example, to the growth and disappearance of microphases which bring about the change of structure, l↔l to the expansion, contraction, and rotation of the chain structure-units, and l*↔l* to the movement of alkalineions.

抄録全体を表示

主としてスラグ系セメントと対照としてのポルトランドセメントについて配合特に石膏添加量が硬化におよぼす影響を調べ, さらに水和度と液相組成の変化を測定し, それらの問の関係について研究した.
各種セメントの強さに対し適当なSO₃量はポルトランドセメントでは2.6%, 高炉セメントでは1,8%高硫酸塩スラグセメントでは7%附近に見出された. このような配合においてモルタルの液相のSO₃濃度はポルトランドおよび高炉セメントでは1日で200-300ppm程度, 高硫酸塩スラグセメントでは3日以後でほぼ一定の500-600ppm程度となった.
高硫酸塩スラグセメントの液相のCaOとSO₃は3日までは急に減少するがSO₃はCaOの当量以上に存在する. また連結養生でCaOとSO₃濃度は非常に増し, 一度結合した石膏が再び遊離するのを認められ, これに対応して曲げ強さは相当に低下した.
液相のpHは各セメント共1日以後はほぼ一定の値をとる. 高硫酸塩スラグセメントでアルカリ刺激剤が適量であると始めからpHは12附近に保たれるが, 過剰であると特に極く初期のpHが高くなり, 水和度はほとんど変わらないが強度の増進がさまたげられる.
どの型のセメントにも強度をf, 水和度をIとした場合に回帰曲線f=aIⁿがよく適合した. ポルトランドおよび高炉セメントではaが小さくnが大きいが, 高硫酸塩スラグセメントでは逆の傾向が見られる. またスラグ含有量の多いセメント程強さを負担する水和物が低温で脱水しやすい結合水を多く含むようであった.

抄録全体を表示

Calcium silicate hydrates, which are one of the principal bonding materials in mortars and concretes, are formed by puzzolanic reactions or hydrations of calcium silicate anhydrates.
This paper is a report of an investigation of the reactivity of silica gel in lime solution, and deals with electronographic morphology, X-ray and electron diffraction analyses of solid phases.
The initial C/S mol. ratios ranged from 0.50 to 3.50, and the solution was kept at the constant temperature of 20°±1°C for 95 days giving occasional agitation by a magnetic stirrer.
For comparing the solid reaction products with the hydrated solid phases of calcium silicates in clinker, ordinary portland cement and β-C₂S were subjected to similar tests.
The results obtained are summarized as follows:
(1) More than 90 days seem to be necessary before the system lime-silica gel-water reaches to neary stable equilibrium. CaO combines rather rapidly in the first week with the velocity decreasing with time. The effect of C/S mol. ratios in starting mixtures on these of the reaction products came to the front on long term experiments, for example, about 60 days.
(2) C/S mol. ratios of the solid reaction products were in the range from 0.45 to 1.24 which showed a rapid increase up to about 0.8, the point of inflection, turned then to a slow increase even if an appreciable amount of CaO remained in the liquid phase. The principal factors governing the reaction were the initial C/S mol. ratio and the time of reaction.
(3) The solid phases thus obtained were exclusively the aggregates of semi-transparent, thin, flexible films or foils being quite likely as having a gell-like structure as it was proved by the observation under electron microscope. The crystals often curled up to rolls, and gave the lattice spacing calculated from the pattern of electron diffraction usually smaller than that estimated by X-ray analysis. It is likely that this kind of difference is due to the shrinkage of crystals produced by the electron bombardment and also by high vacuum in electron microscope. Difference of C/S mol. ratios of the initial mixture did not give any recognizable influence on the properties of the solid reaction products. It is very likely that the variation of the C/S mol. ratios of solid reaction products is caused by the adsorption of CaO but not by the formation of solid solution. The filmy crystals may be the earlier stage of the crystalization of mono-calcium silicate hydrate (low temperature modification).
(4) X-ray analysis showed the systematic absence of h k l for h+k odd. The reflection of (220) was abnormally strong and its peak profile traced by Geiger-counter diffractometer was generally unsymmetrical, which probably shows the two dimensional mistaked structure.
(5) X-ray analyses showed the presence of calcium silicate hydrates in portland cement pastes (W/C=50%) cured 3-7 days. The observation under electron microscope of the paste cured 35 days revealed the existence of the semi-transparent filmy, and also fibrous, or needle-like crystals whose appearances were able to draw a distinction from ettringite. It is highly probable that these are calcium silicate hydrates, although it was impossible to discriminate between mono and dicalcium salts.
(6) According to the results of X-ray analysis and of the observation under electron microscope the hydrated solid phase of dicalcium silicate in large amount of water is monocalcium silicate hydrate identical with that formed in the system lime-silica gel-water. It is also filmy crystals whose C/S mol. ratio was <1.38.

抄録全体を表示

Having confirmed the formation of passably pure titanium nitride by heating the mixture of graphite and titanium oxide in the current of pure nitrogen (1956) the authors have carried out a series of investigations on the reaction of nitrogen with metallic titanium (99.5% or higher).
A titanium plate was heated in a graphite tube furnace to 1100°-1600°C for 1, 2, and 3 hours in the current of pure nitrogen under the pressures 7, 76, and 760mmHg.
The products so obtained were investigated by X-ray, and by microscope. Also the refractive indices were measured by immersion method.
Weight increase by nitrogenation proceeded rapidly from 1400°C, on which the temperature exerted larger influence than furnace pressure. Moreover, larger increase in wight was found by the heating under reduced pressure.
The lattice constant of the surface layer varied from 4.26₂ to 4.30₅ Å, which is higher than the value for TiN and lower than that of TiC. The examination under microscope revealed the existence of the small amount of graphite crystals, from which the authors attributed the origin of the carbon to the heating element of graphite furnace.
By X-ray analysis it was confirmed that the surface layer was composed of the TiN-TiC solid solution having the lattice of NaCl type, while in the inner layer was formed a solid solution of hexagonal crystals containing less amount of carbon and nitrogen. The surface layer, composed almost entirely of the crystals of NaCl type, was obtained by the heat treatment at a temperature higher than 1400°C.
Observation under microscope has disclosed that the crystals of the surface layer may be divided into two main classes, isotropic and biaxial. And a small amount of biaxial crystals were identified in the specimen which was heated at 1600°C.
The refractive index of high refractive isotropic crystals was found to be n=1.72₈, while that of low refractive ones gave the value n=1.49₂. The refractive indices of biaxial crystals ranged from 1.50₀ up to 1.78₆.

抄録全体を表示