Dl. Youchison et al., THERMAL FATIGUE TESTING OF A DIFFUSION-BONDED BERYLLIUM DIVERTOR MOCK-UP UNDER ITER-RELEVANT CONDITIONS, Fusion technology, 29(4), 1996, pp. 599-614
Thermal response and thermal fatigue tests of four 5-mm-thick berylliu
m tiles on a Russian Federation International Thermonuclear Experiment
al Reactor (ITER)-relevant divertor mock-up were completed on the elec
tron beam test system at Sandia National Laboratories. The beryllium t
iles were diffusion bonded onto an oxygen-free high-conductivity coppe
r saddle-block and a dispersion-strengthened copper alloy tube contain
ing a copper porous coating. Thermal response rests were performed on
the files to an absorbed heat flux of 5 MW/m(2) and surface temperatur
es near 300 degrees C using 1.4 MPa wafer at 5 m/s flow velocity and a
n inlet temperature of 8 to 15 degrees C. One tile was exposed to incr
ementally increasing heat fluxes up to 9.5 MW/m(2) and surface tempera
tures up to 690 degrees C before debonding at 10 MW/m(2). A second til
e debonded in 25 to 30 cycles at <0.5 MW/m(2). However, a third file d
ebonded after 9200 thermal fatigue cycles at 5 MW/m(2), while another
debonded after 6800 cycles. Posttest surface analysis indicated that f
atigue failure occurred in the intermetallic layers between the beryll
ium and copper. No fatigue cracking of the bulk beryllium was observed
. It appears that microcracks growing at the diffusion bond produced t
he observed gradual temperature increases during thermal cycling. Thes
e experiments indicate that diffusion-bonded beryllium files can survi
ve several thousand thermal cycles under ITER-relevant conditions. How
ever, the reliability of the diffusion-bonded joint remains a serious
issue.