Search results

The nitrided layer with an average grain size of 20 µm was formed on the surface.
The AlN layer shows a wave shape corresponding to the grains shown in Fig.1(a).
These grains relate to the grains of the substrate and are formed by thermo etching.
There are no cracks on the grain boundaries of AlN layer.
No grain boundarys were formed in the AlN layer on the substrate of AC4C.

Introduction It has been known that grain boundaries affect fatigue life of metallic materials.
Hence, it can be said that general grain boundaries contribute to the increase in the fatigue life.
Slip band formations at grain interior were rarely observed.
Twin band (a) With twins Grain boundary T.A.
(b) Without twins Twin band (a) With twins Grain boundary T.A.

Fy is force projected in the vertical direction, E is Young’s modulus, ny the number of yarns in contact with the projectile, Sy the section of the yarn, and n1 the number of layers.
These experimental results are the ballistic penetration results of 4 and 16 grain FSP projectile into Kevlar 29 fabric with 8, 16 and 22 number of fabric plies.
a b Fig. 4 Residual velocity in terms of impact velocity for 8, 16 and 22 fabric plies for a: 4 grain and b: 16 grain FSP projectile, Impact obliquity is 45º References [1] I.

This leads to a remarkable higher Fei concentration in lowly dislocated grains than in the adjacent grain boundaries with high dislocation density.
The Fei image in Fig.1 of a wafer (at 34% ingot height) shows a high contrast between grains and grain boundaries, too.
The large numbers of partially bonded silicon atoms give rise to many dangling bonds, and therefore, a large density of defect levels are found within the band gap near the semiconductor surface.
Including in addition out-diffusion of iron atoms (e.g. from precipitates situated at the grain boundaries), the vicinity of each grain boundary is contaminated, resulting in a decreased lifetime value around the grain boundaries.
Assuming appropriate source concentrations at the grain boundaries, an excellent fit is obtained (red line), which strongly supports the model of out-diffusion.

The grain size was determined by applying the intercepted-segment method.
The grain size of the microspecimens of the batch A (0.28 µm) and B (0.29 µm) are very similar.
In contrast to this the microspecimens of the batch C show with 0.42 µm a clearly higher grain size.
An additional growth of the grain size was determined on the microspecimens of the batch D (0.67 µm).
In small numbers larger pores (5-10 µm) are located at edges.

The effects of TMCP variations will be evaluated in terms of MFS’s and final grain sizes.
Finally, schedule (C) utilizes the same temperature range of schedule (A) but achieves the same final strain using a reduced number of passes and larger applied strains per pass.
Table 3- Below displays the grain size measurements.
This effect can be confirmed in the lower standard deviation of the grain size measurement.
The analysis of MFS’s evolution and final grain sizes has shown that temperature reductions and large strains per pass led to smaller grain sizes in the final product.

The initial grain size in the scanned area varies in a wide range with the mean size of 10.98μm.
The grains are refined to a certain extent with the mean size of 6.99μm after deformation at high temperature compared with the initial grain size, which is obviously the result of dynamic recrystallization during corner forming[13].
But as shown in fig. 6(b), the grains are refined less significantly with the mean size of 8.82μm.
It takes 1100s to fully fill the corner and some grains grow up during this time.
ACKNOWLEDGEMENT The authors would like to thank for the supports from National Natural Science Foundation of China (Project numbers: 51075100 and 51105203) and Program for Changjiang Scholars and Innovative Research Team in University (No.IRT1229) 5.

One can observe an equiaxed β structure, with an average grain size of about 40 - 50μm.
Around 650 MPa stress level, as showed in Fig. 5, one can see that the initial twin line is growing until has occupied almost the entire grain and also that the number density of twins increases with the imposed stress.
First twin lines which do not cross entirely the whole grain, are observed, as seen in Fig. 6, at 650 MPa stress level.
SEM image with first appeared twin lines, at 650 MPa, which do not cross entirely the whole grain.
Also, the structure presents a favourable condition for twinning because grain size is relatively large.

Asaro & Needleman [2] modified Taylor model and assumed that all grains have equal volume and the deformation gradient within each grain has a uniform value throughout the aggregate.
But the compatibility of Taylor model can only satisfy with the equilibrium inside each grain, cannot remain the intergranular equilibrium.
In this study, the large commercial software-ABQUS was applied to model the texture development of BCC IF steel sheet under the cold rolling deformation, by directly inputting the initial grain orientations into subroutine UMAT.
The constitutive equation for elastic deformation in the crystal can be expressed as [3] )1()1( ELT ::::= ,       −⋅= IFFE T **)1( 2 1 , ( )( )( ) TFFFT − − = ***)1( det 1 σ , ( )∑= = N k kkw 1 σ σ (2) where LLLL is the fourth order elastic tensor, )1(E is an elastic strain tensor obtained by polar decomposition, I is the second-order identity tensor, σ is the Cauchy stress in the single grain, ( ) TF −* is the transposed inverse matrix of *F , σ represents a volume average of Cauchy stress, N is the number of total grains of polycrystalline, kw is the volume fraction of each crystal, and kσ is the Cauchy stress in the kth crystal.
Small amount of pearlite distributes at boundaries of ferrite grains.

Self-information reflects only symbol of uncertainty and information entropy can be used to measure the entire source X integral of uncertainties, the definition is: (2) Among them, i is all possible symbol numbers for x.
The attribute of Class label has m assume different values and m different classes Ci(i=1，2，…，m). si is the number of class Ci.
Sij is the tuple number of class Ci in subset Dj.
Expectation information (entropy) divided into subset according to A is depicted as follow: (4) Whereis weight of j subset and equals to the total number of tuple in D divided by the number of tuple subset.
Step 1: data standardized From lumber based database take 10 historical data in table 2, that is to say data tuple number s = 10.