Influence of Powder Geometry in Powder Injection Molding Process Parameters

Influence of Powder Geometry in Powder Injection Molding Process Parameters L. M. Resende1, A. T. Prata2, A. N. Klein2, L. H. S. Almeida2 1 Universida...
Author: Camron Quinn
5 downloads 1 Views 696KB Size
Influence of Powder Geometry in Powder Injection Molding Process Parameters L. M. Resende1, A. T. Prata2, A. N. Klein2, L. H. S. Almeida2 1 Universidade Estadual de Ponta Grossa Departamento de Engenharia de Materiais - Ponta Grossa - PR Brazil 84031-510 2 Universidade Federal de Santa Catarina Departamento de Engenharia Mecanica – Florianópolis - SC Brazil 88040-900 Key-words : rheology, powder injection molding, granulometric distribution, powder geometry Abstract: Powder injection molding (PIM) is an alternative route to produce components from metallic powder that have been growing in last 10 years. The process may be resumed in five basic steps, that is powder-binder mixing, mixture injection, binder extraction and siterization. The greatest advantage of the metallic powder processing is that it surmounts difficulties from conventional powder metallurgy regarding feasibility of tridimensional shapes and density homogeneity. Powder selected to PIM process must provide a high powder packing and sinterability, which is necessary to obtain adequate final density levels. In this way, spherical powders not higher than 20 µm are usually used. In the present work, the possibility of using nonspherical powders for PIM was investigated. To provide basis for comparisons, spherical and nonspherical powders were tested. Powder influence was analyzed concerning viscosity behavior. Introduction One of the most limitation of Powder Metallurgy technique is to obtain parts with a complex geometry, due the limitation of a uniaxial pressing. Powder Injection Molding process (PIM), inspired in polymer injection process, consists on injecting powder into cavities with complex geometry. That is possible mixing the powder with an organic binder. The mixture powder-binder must have a rheological behavior similar to polymers used in injection process. One other problem in uniaxial press process, not found in PIM, is density gradient which is uniform in injected components. Fine and spherical powder are generally used due to its high sinterability and high packing [1]. Powders with irregular geometry, although cheaper are not desired due its low packing, which results in a high mountain of binder to obtain optimum rheological properties [2]. In this work the influence of powder geometry was studied concerning to viscosity mixture varying deformation level and amount of non-spherical powder. Materials and Methods In this work spherical iron powder were used with different size distribution. Powders used are listed in Table 1. Powder fraction was constant in 55%. Three powder characteristics were analyzed com concerning to viscosity: powder mixture tap density, powder deformation level and amount of non-spherical powder used Tap density was determined as MPIF 46 Standard. Powder mixtures were classified in 6 different levels of tap density and disposed in a diagram as illustrated in Fig. 1. Spherical powder were deformed by milling. A ball mill (one hour cycle) and a high energy mill (one and two hours

cycle) were used. Different mills in different cycle times were used in order to obtain 3 different deformation levels to the powders. This was the second factor analyzed. The analysis of mills and milling time was done in previous work [3]. Table 1 – Powders Used Fine Powder (SM) Medium Powder (CL) Coarse Powder (QB) Supplier Geometry Composition (max)

Basf Corp. - Basf SM Spherical 0,1%C, 0,5%O, 0,01%Ni

Basf Corp. - Basf CL Spherical 0,05%C, 0,2%O, 0,01%Ni

True density (g/cm3) Average Particle Size Size distribution

7,7 -------

7,8

Quebec Atomet 1001 Spherical 0,003%C, 0,08%O, 0,01%S, 0,18Mg 7,8

6-8µm

68µm

10% - 0,81µm 50% - 1,91µm 90% - 3,66µm

10% - 5,0µm 50% - 10,0µm 90% - 25,0µm

10% - 30,33µm 50% - 68,28µm 90% - 141,85µm

100% CL

Tap Density [g/cm3]

QB

CL

100% SM

100% QB SM

Figure 1. Tap density diagram with different powder compositions

Figure 1 – Tap Density diagram with different powder composition Milled powder was aggregated to the spherical powder in 3 different levels of fractions, 10, 20 and 30% (keeping the total amount of powder constant in 55%). The fraction of non-spherical powder used was the third factor studied Powders blends used are listed in Table 2. The amount of fine, medium or coarse powder used in each powder blend is a function of tap density level desired to the blend Powder micrography of spherical and milled powders are shown in Fig. 2. The binder system used is described in Table 3. Components (binder ad powder) mixture was done in a sigma mixer (Haake Polylab System). In preparing the mixtures the powder was pre-mixed at room temperature with stearic acid and paraffin. Polypropylene and EVA were melted and the pre-mixed powder was added and mixed for 30 minutes at 170oC. The mixture was then cooled to room temperature and mixed once more during the same period of time and at the same temperature. Viscosity data were taken in a capillary rheometer coupled to an extruder. The capillaries employed had diameters of 1.0, 1.2 and 1.5mm with a constant L/D ratio of 40. Mixtures were injected in an Arburg 320S injection machine, clamping force of 50T. The same injection parameters were used to all mixtures injected

System number 000 012 021 101 110 122 202 211 220

Table 2 – Powder systems studied Fine Medium Coarse Tap Density Fraction of Powder Powder Powder Powder [g/cm3] non-spherical Deformation level amount (%) amount (%) amount (%) powder [%] (form factor) 80 10 10 3,82 10 0,79 80 10 10 3,82 20 0,55 80 10 10 3,82 30 0,64 40 40 20 4,37 10 0,64 40 40 20 4,37 20 0,79 40 40 20 4,37 30 0,55 20 50 30 4,75 10 0,55 20 50 30 4,75 20 0,64 20 50 30 4,75 30 0,79

(b)

(a) Figure 2 – Powder micrografy after (a) and before (b) milling

Component Polypropylene Paraffin Wax EVA Stearic Acid

Table 3 - Binder Data Volume Density (g/cm3) Content (%) 0.90 40 0.82 40 0.94 15 0.92 5

Mass Content (%) 41.1 37.5 16.1 5.3

Chemical debinding was done in atmosphere and than immersion in Hexane in a total time of 5 hours, in an apparatus projected at the Materials Laboratory at the Federal University of Santa Catarina – Brazil [4]. This system was the most effective for this binder system [5]. After a chemical debinding, a thermal debinding was done, in hydrogen atmosphere, followed by a presintering. Sintering was done in hydrogen atmosphere at 1150oC, for 30 minutes. The density parts were determined by MPIF 42 Standard.

Results and Discussions To analyze viscosity data, an confounding 33-1 factorial design was used. The nine different mixtures used are those described in Table 2, and the influence of 3 main factors in viscosity were analyzed: powder tap density, powder deformation level and fraction of non-spherical powder . The viscosity curves obtained are shown in Fig. 4. Mixture System 000 012 021 101 110 121 202 211 220

Viscosity [Pa.s]

1000

100

10

1 10

100

1000

10000

100000

-1

Shear rate [s ]

Figure 4 – Viscosity curves for different mixtures with non-spherical powder The viscosity curves seems to have different values when analyzed to low shear rates and same values when analyzed to high shear rates. In Fig. 5 curves with the lowest, the highest and with an intermediate value of viscosity are shown.

Viscosity (Pa.s)

1000

Systems 122 211 000

100

10

1 10

100

1000

10000

100000

-1

Shear Rate (s )

Figure 5 – Viscosity – shear rate curves of 3 systems with non-spherical powder Analyzing Fig. 5, it is possible to notice that powder geometry has a higher influence in viscosity function for low shear rates than for higher shear rates. A high level of shear rate (up to 10000 s-1) minimize the influence of powder geometry, with an overlap of viscosity curves (in any level of powder deformation or quantity of non-spherical powder aggregated here studied). It seems that powder geometry influences in viscosity level up to the point shear rate is high enough to

orientates powder particles. The higher the shear rate the higher the orientation of the powder, parallel to shear direction. When using only spherical powder, this effect does not occur, no matter the size of the particles analyzed. As there is not an preferential direction of orientation to spherical powder, small or coarse particles have the same behavior under the same levels of shear rate [6]. The pseudoplastic behavior of viscosity function is changed with the use of irregular powders. Fluidity index changes when powder characteristics are studied. But, in a certain level of shear rate, viscosity values converge to the same range, despite the fraction of non-spherical powder or the powder deformation level used. Statistical analysis shows that the high the shear rate the low the influence of any factor studied in powder geometry. To a shear rate of 300 s-1 the most influence is due to powder deformation level. The more non-spherical powder is, the higher its contribution to enhance viscosity levels. The second significant factor is powder tap density. The lower tap density, the higher viscosity level to a certain shear rate. Interactions among the 3 factors studied is the third significant factor (interaction is understood as the influence that one factor make on other when acting both simultaneously). Non-spherical powder fraction is the lowest influence factor in viscosity levels. Up to 30% fraction of non-spherical powder (the highest level studied), low influence in viscosity function was observed. Analyzing data to an intermediate shear rate value, 1000 s-1, conclusions are very close to a shear rate of 300s-1. To higher shear rate values analysis, ranging about 8000s-1, deformation level factor influence viscosity values the most. Interaction among the factors studied followed by nonspherical powder fraction influence viscosity. Powder tap density does not influence viscosity levels for high shear rate ranges. That means that the most important factor to be analyzed is how nonspherical the powder is. To minimize viscosity values enhances, when using non-spherical powders, the less deformed the powder is, the better. In this way, a higher powder volume fraction is possible to be used. In Fig. 6, viscosity curves of systems 202 220, and a system with only spherical powder, all with the same tap density (4,73g/cm3) are shown. Systems 202 and 220 have different non-spherical powder fraction and deformation level,. System 220 has the lowest powder deformation level but the highest non-spherical powder volume fraction. System 202 otherwise, has the lowest non-spherical powder volume fraction and the highest powder deformation level. So, it is possible to notice that deformation level enhance viscosity more than non-spherical powder volume fraction. spherical powder system 202 system 220 system 202 system 220 system spherical powder system

1000

Viscosity [Pa.s]

100

10

1 10

100

1000

10000

-1

Shear rate [s ]

Figure 6 – Viscosity – shear rate curves of systems with same tap density and same powder fraction, with and without non-spherical powder

Comparing these factors, i.e., non-spherical powder fraction and deformations level, with the system with only spherical powders, it is easier to make some analysis. Mixtures with spherical powders and mixtures with an amount of non-spherical powder may have the same levels of viscosity (both with the same powder fraction), if geometry of non-spherical powder is not so much deformed. In comparing viscosity curves of systems 220 and 202 to viscosity curve of a system with spherical powder (same powder fraction and same tap density), 220 system curve overlap spherical system curve, although it has a high amount of non-spherical powder (30%). 202 system has a lower amount of deformation powder (10%) but in a higher level of deformation. That indicates deformation level as the most important factor influencing viscosity. Conclusions In this work an analysis of powder-binder system viscosity, used in PIM was done. Three powder characteristics were analyzed: powder tap density, powder deformation level and amount of deformation level used, using a confounding factorial design. These powder properties influence viscosity function depending on the range of shear rate analyzed. The lower the shear rate, the bigger the powder influence. To very high shear rate values (10000s-1) powder characteristics seem not to influence viscosity values. To this shear rate range, mixtures viscosity have very close values, no matter how high amount of non-spherical powder and deformation level the powder have, for the levels here studied. That is because higher shear rates tend to align powder geometry to the shear rate direction minimizing powder geometry influence Powder deformation level is the most significant factor to influence viscosity followed by powder tap density. The less significant factor is non-spherical powder amount. References [1] A. R. Erickson, H. E. Amaya, International Conference - Materials by Powder Technology (1993 : Dresden). Anais. Dresden : Oberursel, p.145-155, 1993 [2] K.M Kulkarni; Advances in Powder Metallurgy, 1990, vol. 3, [3] L.H.S. Almeida; L.M.Resende; P.A. P.Wendhausen, Proceedings of VI CREEM - Congresso Brasileiro de Estudantes de Engenharia Mecânica, Brasilia, 1999. [4] J.L.C.Marques; D.F.S. Junior, J.G. Justino, Proceedings of V CREEM - Congresso Brasileiro de Estudantes de Engenharia Mecânica, Vitória, 1998. [5] P. A. P Wendhausen, M. C. Fredel, J. G. Justino, L. M. Resende, R. M. Nascimento, A. N. Klein, Proceedings of the Powder Metallurgy World Congress, 1998, Granada, Spain, vol. 5, p. 387-392 [6] L.M Resende, A. N. Klein, A. T. Prata, Advances in Powder Metallurgy, 2000, vol 4.

Suggest Documents