S. Liet al. Journal of the European Ceramic Society 21(2001)841-845 Load-point deflections to center-point deflections of stantially to the observed creep response, although most creeping specimens were measured directly using linear of the primary strain reportedly derived from redis variable displacement transducer (LVDT). In the tribution of the intergranular matrix glass. In the pre- and before the load was applied, the time of constant BN interlayer facilitated the redistribution of stress l, c experiment, the rate of rising temperature was 500C/h ent study, the weak interfacial bond provided by temperature was 15 min. Samples were creep tested similar fashion during primary creep, particularly in the between 1000 and 1200C at stresses between 250 and vicinity of the stress concentration, such as micro 600 MPa under an air atmosphere. After testing, the cracks. Moreover, the glassy phases concentrated in the ensile zones of the specimens were examined using X- BN layer and which might happen to plastic flow on the ray diffraction (XRD) to determine the phases in the interface layer. The two mechanisms may, meanwhile materials and by scanning electron microscopy to char- effect the material creep Tertiary, or accelerated, creep acterize the creep deformation mechanisms. is usually attributed to 'distributed" damage caused by nucleation, growth, and coalescence of cavities microcracks, which gradually deteriorates the load 3. Results and discussion bearing capacity of the specimen. The lack of tertiary creep may suggest that fracture was dominated by 3.1. Creep response localized damage due to the growth of preexisting defects, e.g. macrocracks or voids, even though other In this study behavior of Si3 N4/BN fibrous damaging mechanisms may operate concurrently. In monolithic ceramic systematically investigated in other words, the growth of a defect to its critical size the temperature 1000-1200 C. Test conditions may usually happen before other damaging mechanisms and rupture times are shown in Table 1 and typical start to have a significant effect on creep. However creep curves at 1200C are shown in Fig. I there was a difference between monolithic ceramic and An obvious feature exhibited by these creep curves is fibrous monolithic ceramic. We know that fibrous mono- their extensive primary creep and the lack of tertiary liths are not governed by weak link statistics like mono- creep. There was a substantial primary creep response, lith materials. For example, a few cells of a fibrous which, in fact, accounts for most of the measured strain. monolith can fracture without catastrophic failure The large primary creep response related to stress redis- occurring if the remaining cells can support the applied tribution processes occurring in the composite and to the stress, so the tertiary creep of the Si3N4/BN fibrous intrinsic response of glass-ceramic. 8. 9 This redistrution ceramic does not appear of stress may involve one or more of the following pro- There are many creep laws and thus many creep cesses: redistribution of residual intergranular glas equations, none of which can describe the whole creep phase and/ or compliance of interface. Mayer et al. process. Usually, the creep strain rate(e)can be described systematically explored the role of the interface response in by the power law: the composites. From their experiment, they concluded that plastic flow of an interface layer of siliconous glass emulated a debonded interface and contributed sub Where A is a constant, o the stress, n the stress exponent The rupture times of the composites at different temperatures under a for creep, Q the activation energy for creep, r the gas variety of stresses constant, and T the absolute temperature. The stress Temperature (C Creep stresses(MPa) Rupture times(h) 350MP .2 1200 Creep Time(h) 0.7 Fig 1. The creep curves at 1200.C under different stresses.   0     0    #     8  #        <JK/=7 F  1 !     #   ),,

!        !         ()  7      (,,,  (+,,

    +),  3,, 45     7 .  #!    6      1   # G 8  :  <G'K=           8  #      8    6       7     /-%-    F   8!                8 8  #    # (,,,2(+,,

7 /             / (  8     (+,,

    ; #7 (7 .    1   8      1   8       8 7 /      8   !  !  !          7 / #  8               #                # 7>!B /       8         #   D        # #       7 48  7B 8 8 1              7 ;   1 ! 8      0    8       #            8      !  #      8    8            #  1 #7 F    8!          8   8                   #  8 !  8      8      !     7 4 !  #8       8     #     0    87 /    8!   ! :    7 / 8! !   8    L  M #  8  ! # !          !   #8       #  8    7 /    8  8 ##       8   6 #    #   1  # ! 7#7    !   #  # #   8   87 F  !  #           6 8 8    # #        #  :  7(, $ !     :                   7 %             #  8               7 ; 1!                      #     #        !    8              7 /  8      8   !              7 N8!      <
=     8    D
0
 1 *
% 0    !  !    1   ! 1     #8  ! *  #  !      7 /  / ( /            :     8   / <

=
  <45= '   <= (,,, +), A,, ), A,, ,, A,, ),, ), 3,, + ((,, +), A,, ), A,, ,, A,, ),, (3 3,, (7+ (+,, +), A,, ), +(, ,, , ),,  3,, ,7C ; #7 (7 /    (+,,

  : 7 >+
-   - 2 3      
 ,% 4,&&%5 '(%6'(7