Understanding powder behaviour during preparation, transfer and filling of dies is important in powder metallurgy as each step influences the quality of the final component. For high productivity and quality during manufacture, it is essential that process and powder characteristics are compatible.
In this investigation of the bulk, dynamic flow and shear properties of four different materials, we show that die filling effectiveness correlates well with many powder properties. Thus the availability of accurate information about the flow and bulk properties of powders allows reliable prediction of die filling efficiency, removing any need for the more usual ‘trial and error’ approach.
Components made from metal powders are manufactured in distinct stages:
die filling - powder drops from a feed shoe into the die cavity
powder transfer - powder is transferred within the die and through a series of tool motions to produce a compact approaching the final shape
compaction - powder in the die is compressed to form a green body
ejection - after compaction, powder is ejected from the die
sintering - following compaction, the green body is sintered in a reducing atmosphere
component is sized or machined in order to ensure the maintenance of dimensional tolerance
The compaction, ejection and sintering stages in particular have a significant impact on the properties of the finished components. A knowledge of the behaviour of powders in preparation, die filling and powder transfer is important, since their packing structure and density distribution may influence later stages, and affect the integrity of the final components.
Powder Flowability and Die Filling
Various experimental techniques have been used to measure powder flowability in relation to die filling. These include:
Hausner ratio and Carr index (poured and tapped bulk density)
Hall flowmeter and Flodex flowmeter (mass flow rate or time required to discharge through an orifice)
Angle of repose; and
shear cell, which measures the yield strength of a consolidated bulk solid.
While all may be useful in specific process environments, none predicts the behaviour of a powder during die filling.
Factors Influencing Powder Behaviour
Powder behaviour is complex. It is influenced by a combination of physical properties and the characteristics of the processing equipment, and powder flowability cannot be expressed adequately as a single value or index. Die filling is a dynamic process, so any powder characterisation methods used should closely reflect the real industrial situation.
Case Study - Die Filling Behaviour of Tungsten, Aluminium and Glass Beads
Here we examine the die filling behaviour (die filling ratio) of tungsten, aluminium, and two kinds of glass beads of different nominal size. Bulk, dynamic flow and shear properties were characterized using the
FT4 Powder Rheometer. ®
Particle Size and Morphology
Particle size distribution for each material was determined using a Mastersizer 2000 (Malvern Instruments, Malvern, UK). Particle morphology was characterised with a JEOL 6340F Scanning Electron Microscope (SEM). The results are shown in table I.
Table 1. General powder properties.
D 50 (µm)
GL Glass beads
GS Glass beads
Granules Aluminium Powder
Obtaining repeatable data requires powders to be in a homogeneous packing state. When a powder arrives for testing, it has a unique history that has been influenced by factors such as consolidation, segregation, aeration, caking, or vibration. The ‘conditioning’ needed to remove this history involves gentle displacement of the whole powder sample, loosening and slightly aerating it into a homogenised and reproducible state.
All samples for dynamic, bulk and shear testing were conditioned and ‘normalised’ before measurement, using the
FT4 Powder Rheometer.
Please click here if you would like more information on the instrument used in this article or a quote
Measuring Powder Flowability
FT4 Powder Rheometer is a universal powder tester. Accessories such as blades, pistons and shear heads can be rotated and simultaneously moved axially into a powder sample whilst axial and rotational force are measured. A number of control modes are available on both axes, including velocity, force and torque. Standard dynamic tests, aeration testing and shear testing are automated with no operator involvement apart from sample preparation.
Dynamic testing was performed using a 48 mm diameter blade and a 160 ml powder sample contained in a 50 mm bore, borosilicate test vessel (figures 1a and 1b). An automated, 18 segment, 48 mm diameter rotational shear cell accessory (figure 1c) was used for all shear testing, using 85 ml sample.
Figure 1a. Downwards testing mode showing bulldozing action along the entire blade length.
Figure 1b. Upwards testing – shearing with minimal consolidation.
Figure 1c. Shear cell above sample vessel.
Die Filling Rig
A model die filling rig was designed to mimic commercial die filling processes (figure 2). It consists of a stationary die and a motorised unit which drives the shoe at a steady velocity from 50 to 300 mm/s. In this study, the feed shoe was cylindrical with a fixed volume of 160 ml and diameter of 50 mm. The cylinder die had a 10 ml volume and 25 mm diameter.
Figure 2. Schematic drawing for die filling process.
Experiments were carried out in air at shoe velocities of 50 mm/s to 250 mm/s. 160ml samples were conditioned on the
FT4 Powder Rheometer and then carefully transferred to the rig for translation over the die. The mass transferred into the die was then measured to determine the filling ratio. This was repeated three times, each time using a newly conditioned sample of powder.
Results and Discussion
Table 2 summarises the most important material properties, which are further discussed in this section.
Table 2. Powder flow properties characterized under different status.
Basic Flowability Energy, BFE (mJ)
Stability Index, SI
Flow Rate Index, FRI
Conditioned Bulk Density, CBD (g/ml)
Bulk Density, consolidated by 20 taps, (g/ml)
Specific Energy, SE (mJ/g)
Aeration Ratio, AR
Pressure Drop across the powder bed at 2mm/s air velocity, PD 15 (mbar)
Consolidation Index, CI 20Taps - factor by which flow energy increases relative to BFE
Volume change -20 taps (%)
Volume change -15kPa direct pressure, Compressibility (%)
Shear Stress, t 2 (kPa)
Shear Stress, t 1 (kPa)
Unconfined Yield Strength, UYS (kPa)
Angle of Internal Friction (°)
Cohesion, Co (Pa)
Flow Function, FF C
Specific Energy (SE) is a measure of how easily a powder will flow in an unconfined or low stress environment. It is calculated from the energy needed to establish a particular flow pattern in a conditioned, precise volume of powder. During measurement, this flow pattern is an upward clockwise motion of the blade (see figure 1b), generating gentle lifting and low stress flow. SE is calculated from the work done in moving the blade through the powder from bottom to top of the vessel (upward traverse). Gravity dominates in this test, so to compensate for varying bulk densities the flow energy is expressed as Specific Energy, mJ/g.
Specific Energy depends primarily on the shear forces acting between particles. Cohesion is often the most influential property in low stress environments. Sample GS has the lowest SE indicating it flows most readily in a low stress, conditioned state. Tungsten has the highest SE suggesting it behaves most cohesively. Interestingly, GL has a higher SE than GS, indicating higher cohesion.
Basic Flow Energy (BFE)
In these conventional dynamic tests, a previously conditioned powder was consolidated using a bulldozing blade action (figure 1a) that forced the powder downwards towards the bottom of the containing vessel. The Basic Flowability Energy (BFE) value is a key parameter that is highly sensitive and differentiating in relation to small differences in flow properties. Results varied from 5964 mJ (tungsten) to 899 mJ (Glass GS) as shown in Table 2.
The presence/absence of air greatly affects a powder’s flow properties; air is added naturally when powder is moved freely. When a powder is aerated less energy is normally required to move it. This reduction in flow energy is described by the Aeration Ratio (AR).
Cohesive powders do not allow air to permeate through the powder bulk – instead channels or rat-holes occur. The resulting change in energy is therefore small. In less cohesive powders air permeates the entire bulk, with a consequent large reduction in energy. In some cases virtually all particles separate and the bed fluidises. The sensitivity of a powder to aeration relates well to its performance in a gravity feed system and for processes such as volumetric filling.
The four powders behave very differently during aeration when subjected to the same range of air velocities (see figure 3). Above a certain air velocity the aluminium, GL and GS glass beads fluidize. Because of its permeability, the GL sample requires the highest velocity, releasing air rapidly and needing considerable air flow to separate the particles.
Figure 3. How flow energy varies as a function of air velocity.
The permeability of the smaller glass beads, GS, is lower by a factor of 6, allowing fluidisation at relatively low air velocity. Thus a small amount of entrained air will greatly improve the flowability of GS, compared with GL. However, GL will rapidly release a large amount of entrained air. The aluminium powder is also highly permeable, requiring high air velocity to fluidise. The tungsten with its poor permeability, does not fluidise, as shown by the high flow energy requirement (figure 3). Close bonding of the small tungsten particles makes the bed cohesive, with injected air forming channels to escape.
How Consolidation Flow Energy
Closer packing and the loss of entrained air mean that compacted or consolidated powders are less likely to flow freely under gravity. Consolidation of all four powders was determined after subjecting each to 20 taps. Many processes (including die filling) impose vibration and the vulnerability of a powder to this is important. The standard dynamic test was used (as in the BFE measurement) to determine the extent to which the flow energy increased as a result of consolidation.
The results (Table 2) show an increase in normal flow energy (BFE) of 230% for the tungsten with a 16% volume reduction. This compares with an energy increase of 10% for GS glass and only a 2% volume reduction.
Compressibility is a bulk property measurement that examines the volume change of a conditioned sample when it is slowly compressed so that entrained air can escape. While not a direct measure of flowability, it can indicate whether a powder is cohesive or free flowing.
Tungsten is the most sensitive to compression (figure 4). GL and GS have very low compressibility, reflecting the relatively efficient packing of the particles when in a conditioned, low stress state. This is due to their spherical shape, low cohesion and high permeability. Aluminium powder has a median compressibility between glass beads and tungsten powder since their large particle size and irregular shape allows a degree of realignment and closer packing.
Figure 4. Compressibility as a function of applied normal stress.
Permeability indicates the ease with which a material can transmit a fluid (in this case air) through its bulk. For powders, influencing factors include physical properties such as particle size and distribution, cohesivity, particle stiffness, shape, surface texture and bulk density. External factors, consolidation stress for example, also have an effect by changing porosity and particle contact surface areas.
Generally, cohesive powders consisting of mainly sub 30 micron particle size are the least permeable; granular powders are typically most permeable. Powders with large particles and fines may form a tight packing structure, with fines filling the spaces between particles, reducing the permeability of the powder bulk. Permeability determines the discharge rate of entrained air during die filling.
Figure 5 shows the air pressure difference required across the depth of the powder bed to maintain a constant air flow (2 mm/s) while consolidating the powder at increasing normal stresses. Increasing stress reduces the permeability of the tungsten but has little effect on the other powders due to their lower compressibility and greater particle size.
Figure 5. Pressure drop through powder bed at a constant 2mm/s air velocity as a function of applied normal stress.
Granular aluminium and GL are more permeable than the other two samples because of their relatively large void structure.
The primary flow mode of powders is shear, where particles slide relative to one another. High shear strength means greater resistance to flow. Figure 6 shows the relatively high yield loci of tungsten and aluminium compared with glass. Shear tests were carried out at 2, 1.75, 1.5, 1.25 and 1 kPa normal stresses, with samples pre-consolidated to 3 kPa normal stress. The derived unconfined yield strength or compression strength of the powders determines whether discharge from a hopper will cease due to the formation of a stable bridge. Cohesion value and internal angle of friction are also derivable from figure 6 (see values in Table 2).
Figure 6. Shear stress versus normal stress of a sample consolidated and pre-sheared at 3kPa normal stress.
Shear properties are important in die filling because powders retained in the dispensing shoe must shear under gravity in order to flow into the die, and must not bridge. These factors will determine the mass flow rate during the die filling process. The higher the shear stress and internal angle of friction for a given normal stress, the more problematic the powder during flow initiation in the process environment.
Please click here if you would like more information on the instrument used in this article or a quote
Die Filling Behaviour
Die Filling Behaviour of Conditioned Powder
Gravity filling comprises two successive stages: powder flow from the shoe, and powder packing/ settling in the die. Several major factors influence the die filling ratio, including powder flow properties, air escape from the die and escape of entrained air in the powder.
For the initial tests all powders were conditioned before die filling. The die filling ratios of conditioned powder were calculated from filled mass and conditioned bulk density (see figure 7). The four powders have very different die filling abilities: GL glass the best and tungsten the worst. At the lowest shoe speed even tungsten filled well, but at the highest speed the filling ratio deteriorated to below 20% compared with GL glass which remained above 80%.
Figure 7. Die filling ratio of four different powders at different shoe speeds.
Die Filling Behaviour of Aerated and Tapped Powder
The packing condition of powder in the shoe and its impact on air content and bulk density affects die filling performance. To quantify this, aluminium and tungsten were evaluated in both slightly consolidated and lightly aerated states. Samples were aerated at 20 mm/s (aluminium) and 10 mm/s (tungsten) then transferred carefully to the die filling rig.
Aerated Powder had the Best Filling Ability and Consolidated Powder the Worst
The die filling ratio of tapped tungsten powder is shown in figure 8. The effect of tapping is dramatic causing the die filling ratio to drop by a further 50%, and bulk density to increase by 16%. A single measurement on ‘aerated’ tungsten was made at 150 mm/s showing no change from the conditioned data. Evidently aeration in tungsten powder produces channels and does not alter the packing state of the bed.
Figure 8. The die filling properties of Tungsten powder under different powder packing conditions.
Powder Properties and Filling Performance
Thirteen flowability and other powder properties were measured. For most parameters low values mean better flowability while higher mean poorer. The exception is aeration ratio where a high value indicates good aeration and flowability.
The availability of a diverse range of flowability indicators is particularly useful when comparing very similar materials. Because tungsten, aluminium and glass spheres are fundamentally different, in this study it was not absolutely necessary to measure all parameters. However the data (Table II) show excellent correlation of all parameters – low values for the GS glass and the highest values for tungsten – confirming the poor flow properties of tungsten.
Figure 7 shows the filling ratios for the four powders when filling at 230 mm/s. The correlation is clear - tungsten has the worst flow properties and filling ratio, with aluminium the next worst. The glass spheres data is surprising: GS, with smaller spheres, has the best flow properties, yet inferior filling ratios compared with the larger spheres of GL. This is attributed to the higher permeability of GL. Its ability to instantly release entrained air allows more complete filling of the die, confirmed by observing bed collapse rates.
Air is a major influence and particularly important in die filling by gravity. Crucial phases are:
(a) Air content of the powder before filling
(b) Air entrainment as powder falls from shoe to die
(c) Release of the air entrained in the powder in the die
(d) Escape of air from the die upon filling
The first three relate directly to powder properties discussed previously. Venting of the die volume (d) depends on die design and is not part of this study.
The importance of air on flow properties (a) is demonstrated by the aeration data. Tungsten does not aerate due to its fine particle size and cohesivity. The glass spheres however are interestingly different, since GS aerates and eventually fluidizes at relatively low levels of air addition, whereas GL with high permeability requires considerable air to fluidize. GS therefore aerates rapidly on die filling, giving it superior flow properties, but it is relatively poor at releasing air once in the die.
Aerating powder before filling confirmed that aeration improves flow properties (except fine cohesive powders), improving flow into the die and giving higher filling ratios. Consolidation excludes air, increases bulk density, compromises flow properties and reduces the filling ratio.
This study measured the die filling performance of three very different powders and the correlation with their flow properties and other characteristics. A comprehensive set of properties was measured using a universal powder tester, with conclusive results.
There was clear correlation between die filling performance and powder properties. In general powders having low levels of flow energy, flow rate index, consolidation index, shear strength, cohesion, compression strength (UYS), internal angle of friction and compressibility, fill well. However, permeability is also important and there is likely to be an optimum permeability for powders that aerate, since too little allows aeration but inhibits de-aeration. A highly permeable powder may not increase its entrained air or fluidise until considerable air has being applied, but it will de-aerate rapidly.
The four powders have very different die filling performance; 174 µm GL glass is the best and 4 µm tungsten the worst. This is attributed to the good flow properties and excellent permeability of the GL glass beads. The cohesive 4 µm tungsten has poor flow properties and poor permeability.
Die filling performance depends on the packing condition of powders in the dispensing shoe. With 4µm tungsten, consolidation significantly worsens flowability and may result in bridging and zero filling. Powders that readily entrain air will improve their flowability and die filling capacity if aerated.
In summary, die filling efficiency is predictable from information about both flow and bulk properties. Where powders that fill well are available alongside problematic powders, powder characteristics correlating to ‘good’ and ‘bad’ can be measured. New formulations may then be compared with this data and their die filling performance predicted.
This information has been sourced, reviewed and adapted from materials provided by Freeman Technology.
For more information on this source, please visit