Material Preparation:

Two commercially available doped lanthanum manganite powders ((Lao.8Cao2)o.98Mn03, hereafter called LCM-20) were evaluated: Powder X (made by a combustion synthesis process) and Powder Y (made by a co-precipitation process). Powder X was calcined at 1000'C for one hour and ball milled for 24 hours to break up hard agglomerates.

Consolidation and Densification:

The powders were mixed with 10 weight percent o f total solids of an organic binder sys tem. The resulting mixtures were added to a 50.8 mm ID steel die preheated to 149"C and uniaxially pressed

♦Operated by Battelle Memorial Institute for the U.S. Department of Energy under contract DE-AC06-76RLO 1830.

at 8.8 MPa while at temperature. The resulting Powder X and Powder Y discs varied in thickness, averaging 1.975 and 1.414 mms in dimension, respectively. The average thicknesses were normalized to a 10 gram starting weight. After pressing, the Powder X and Powder Y discs were debindered in a three day cycle and heated to the sintering temperature (1150'C and 1450'C, respectively) at a rate of 120'C/hour and held at the selected temperature for two hours in air. The sintered discs were subsequently creep flattened at the sintering temperature for a duration of five hours. All binder burn-out, sintering, and creep flattening studies were made using Lindberg Blue M or Deltech DT-31-FL-10 box furnaces.


Surface areas were measured by a Micrometrics ASAP 2000 BET. Particle and agglomerate size distributions were determined by a Microtrac Series 9200 Full Range Analyzer. Phase was measured using Cu Ka radiation and a Philips Wide Ränge Vertical Goniometer. Sintered densities were measured with the Archimedes submersion technique using ethanol as the liquid medium. Linear expansion was measured on an Anter Model 1161 dilatometer at a rate of 2*C/minute in air. Elemental analysis was measured using inductively coupled plasma auger electron spectroscopy (ICP AES) and energy dispersive x-ray fluorescence (ED XRF) techniques.


If both powders (X andY) had comparable phase compositions, particle sizes and distributions, surface area and morphologies, it would be predicted that the powders would have similar green densities after pressing and thus similar sintered densities. No differences between Powder X and Powder Y were evident from preliminary x-ray diffraction phase analysis. Phase analysis did not identify any major secondary phases in either powder. Surface area and particle size distribution, however, were indicative of the synthesis methods and varied accordingly. Powder X had a surface area of 6.2 m2/gm, and the particle size distribution, which was multimodal, ranged from 0.95 to 434.35 pm with a dso at 9.04 pm. This particle size range was reflective of the large, hard agglomerates formed during synthesis and the effect of subsequent milling. Powder Y's surface area was measured at 1.78 m2/gm. The particle size was monodisperse. The particle size distribution was Gaussian with a range from 2.52 to 15.07 pm with a djo at 5.47 pm.

Further evidence of powder property differences was found in the dilatometric data (Figure 1). The absolute linear shrinkage was measured using bars pressed from both Powder X and Powder Y. The Powder X bar exhibited an expansion at 1250"C whereas the Powder Y bar sintered to complete density at 1450"C without any expansion behavior. This expansion behavior has been observed in lanthanum chromite and strontium doped lanthanum manganite samples when the A/B ratio was approximately 0.90 and when large additions of zirconia (>40 wt%) were mixed in the cathode (7,8). The A/B ratio refers to the cation ratio in the ABO3 perovskite structure. In addition, this type of expansion can occur in the doped manganites with minor phase additions of silica and alumina. Again, evidence of powder differences was observed in the measured green densities. The Powder Y discs had an average green density of 3.49 gm/cc while the Powder X discs had green densities of 2.5 gm/cc. The low green densities of the Powder X discs were indicative of the large particle sizes, high surface area, and agglomerate morphology. As anticipated from the higher green densities, the linear shrinkage, and the higher achievable sintering temperatures, the Powder Y discs densified to higher sintered densities of 5.8 (92.4% of the 6.27 gm/cc theoretical) versus the 5.04 gm/cc sintered densities (80.4% theoretical) for the Powder X discs. In addition to poor sintered densities, the Powder X discs warped and bowed considerably during sintering and frequently cracked during the process of creep flattening. However, since the Composition B discs did not exhibit any linear expansion during the sintering event, these LCM-

20 discs had minimal waipage and flattened without cracking. A Jet Propulsion Laboratory (JPL) designed separator plate which was characteristic of the as-fabricated discs that were pressed is shown in Figure 2.


Figure 1 : Absolute Linear Expansion of Powder X and Powder Y LCM-20


Figure 1 : Absolute Linear Expansion of Powder X and Powder Y LCM-20

Figure 2: As-Fabricated Doped Lanthanum Manganite Separator Plate (71.8 mm O.D.)

To determine the A/B ratios and whether minor quantities of silica and alumina were present that were not visible in the preliminary x-ray analysis, elemental analysis was measured using ED XRF and ICP AES. The reference used for the ICP AES was a LCM-20 sample synthesized using the glycine nitrate process (GNP). The A/B cation ratios for Powders X and Y were 0.95 and 0.99, respectively (Table 1). The silica content was negligible in both powder samples, however, Powder X had twice the alumina content (0.2 wt%) of Powder Y. Both samples contained trace amounts of zirconia (approximately 0.02 wt%). It is unknown at this time if such trace amounts of alumina and zirconia are sufficient to cause the expansion behavior seen in the Powder X sintering curves or if the alumina contributed to the formation of a glassy, secondary phase. The formation of this glassy phase could seal off the surface of the Powder X discs and subsequently trap the gas evolved during the sintering process.

Table 1: Summary of ICP AES and ED XRF Results
0 0

Post a comment