Prakt. Met. Sonderband 46 (2014) 171
icrostructures (approx. 40 vol.% and 2um mean grain size) embedded in a silicate based glass matrix.
the possibility Microstructural characterisation of this type of material can be found in previous work [5].
jual stresses Two different LTCC substrates were selected, containing the same glass matrix (approx.
cients of the 90 % of amorphous glass-phase), but different alumina fillers. The filler of the first material,
ristic strength named as LTCC_E, consisted of equiaxed alumina particles, whereas the filler of the
method (the second material, referred to as LTCC_T, was made of platelet-like alumina particles to
ing different achieve a certain texture grade of the microstructure. For the multilayer architectures,
ical laminate combinations of LTCC_E and LTCC _T were selected with different volume ratio between
and electron the E and T layer materials. Two multilayer architectures were fabricated: (i) TET_1, and
o understand (i) TET_2 with volume ratio Ver 1 = 2.7 and Ver 2 = 18.5, respectively, as shown in Fig. 1.
- The coefficient of thermal expansion (CTE) for both LTCC_E and LTCC_T bulk specimens
was measured in a dilatometer between 20°C and 400°C at a heating rate or 2°C/min.
LS
rence in the
ig sintering of
wever, as the
erential strain
asidual stress
alculated as: Figure 1. LTCC systems investigated: (a) bulk LTCC with equiaxed alumina particles, (b)
bulk LTCC with textured alumina particles, (c) TET_1 multilayer system with volume ratio
Ver 1 = 2.7, and (d) TET _2 multilayer system with volume ratio Ver 2 = 18.5.
3.2. STRENGTH AND RAMAN MEASUREMENTS
The biaxial strength (maximum failure stress) was determined in monolithic LTCC _E and
lus, Poisson's LTCC_T samples as well as TET_1 and TET_2 multilayers, using the Ball-on-three-balls
the mismatch (B3B) testing [6]. Samples of “as-received” rectangular plate specimens of dimension ca.
11.0 x 9.6 x 1 mm? were used. In the B3B the specimen
the reference ; . .
\egligible, Tes Is symmetrically supported by three balls on one side
considered as and loaded by the fourth on the opposite side. A pre-
load of 20 N is applied to hold the specimen in the
+ given as an fixture. Then, a compression test is performed until the
specimen fractures (see Fig. 2). More details on the
B3B can be found in [6; 7]. In this investigation, tests
were conducted under displacement control at
0.1 mm/min using a universal testing machine (MIDI 25-
5, Messphysik, Fürstenfeld, Austria). The presence of
humidity during the test affects the strength distribution
in LTCCs, as demonstrated elsewhere [8], due to
‘environmental assisted cracking”, also called Sub- Figure 2. B3B biaxial testing
Critical Crack Growth (SCCG) [9]. Therefore, all tests configuration; with maximal
were performed in water to hold constant testing stresses occurring in the
conditions for all specimens tested. For statistical centre of the plate [5]
significance, 30 specimens were tested for each
Condition. The failure stress (maximum stress, omax) in the specimen is. given by:
3d state by the P
\1203 particles Omax fz
(3)
