Profiling powder behaviour in new materials
7 minute read
How to move from a single flowability number to a multi-parameter profile that predicts production behaviour across the conditions that matter.
The case for a structured protocol
When a new powder material enters a laboratory for characterisation, the natural tendency is to reach for the most familiar test - often Carr's Index, or perhaps a single cohesion measurement - and use that result to make a judgement. This approach produces a number, but not a prediction. It tells you something about the powder at one condition; it says nothing reliable about how the powder will behave at line speed, after storage, or in comparison to the previous batch.
A powder fingerprint is a structured multi-parameter profile built from a defined sequence of measurements. Each level of the protocol adds a specific dimension of information. Together they produce a characterisation that is directly predictive of production behaviour - and that provides a reproducible reference for comparison against future batches, alternative suppliers, and reformulated versions of the same material.
The protocol described here uses four levels. Not every powder needs every level. The decision to progress to the next level is driven by what the previous level reveals.
|
The principle A powder fingerprint is not a collection of all available measurements. It is the minimum set of measurements that answers the three questions that matter: will it fill consistently at speed, will it discharge reliably from hoppers, and will it remain stable through storage and transport. |
Level 1: Baseline classification - always run first
The baseline fingerprint establishes the powder's behavioural family and identifies the dominant failure mode. It should be run on every new material before any process decision is made.
Standard Cohesion test
Run the single-speed cohesion test and record both parameters:
- Cohesion Index (CI) - the work required to lift and move the powder at a defined speed. Reflects inter-particle attraction and resistance to separation.
- Bridging Factor - the ratio of the peak resistance to the mean resistance during lifting. Reflects whether resistance is smooth and continuous (cohesive bonding) or irregular and event-driven (structural arching and force-chain formation).
These two parameters together classify the powder into one of four behavioural families:
| CI | Bridging Factor | Behavioural family | Primary risk |
| High | Low–Moderate | Cohesive fine | Wall build-up, poor initiation, humidity sensitivity |
| Low | High | Structural/granular | Arching, ratholing, geometry-driven discharge failure |
| High | High | Combined | Both mechanisms - highest severity, most complex to address |
| Low | Low | Free-flowing | Generally reliable - check geometry if problems occur |
Conditioned bulk density
Using the split vessel, the PFA calculates conditioned bulk density automatically after the conditioning cycle. Record this alongside the cohesion parameters. It provides the packing context for all subsequent measurements and is the most reliable single parameter for batch-to-batch comparison.
After Level 1, write a one-line summary of the powder: its behavioural family, its packing state, and its most likely production risk. This summary should appear at the top of any characterisation report and should be updated as further levels are completed.
Level 2: Process fit - run when filling or speed sensitivity is a concern
Level 2 characterises how the powder behaves under the dynamic conditions of production - specifically, whether behaviour changes with throughput speed and whether it remains stable during extended handling. This level is essential before any scale-up decision and for any material used in high-speed filling or dosing applications.
Powder Flow Speed Dependence (PFSD)
Run the PFSD test and record three parameters:
- Speed Sensitivity Ratio (Comp100/Comp10) - the ratio of resistance at the highest tested speed to resistance at the lowest. Values significantly above 1.0 indicate increasing resistance at speed; below 1.0 indicate decreasing resistance. Either extreme represents a production risk.
- Flow Stability - the ratio of resistance at the same speed at the end of the test compared to the start. Values significantly different from 1.0 indicate that the powder changes during handling - hardening, breaking down, or segregating.
- Compaction Coefficient at low speed - the absolute baseline resistance that feeders and hoppers must overcome at start-up and low throughput.
| Result | Product implications |
| SSR >> 1.0, stable | Under-fill risk at high speed - test at actual process speed before scale-up |
| SSR << 1.0, stable | Surge/flooding risk at high speed - check feeder design for high throughput |
| SSR ≈ 1.0, unstable | Drift during production runs - likely handling-induced change; investigate particle robustness |
| SSR >> 1.0, unstable | Highest dynamic risk - both speed and handling drive performance change |
Level 3: Storage robustness - run when caking or restart is a concern
Level 3 characterises what happens to the powder after it has been stored under load - whether it consolidates, how much cake forms, how strong that cake is, and how much work is required to restart flow. This level is essential for any material stored in hoppers, silos, big bags, or IBCs between production runs, or transported in bulk.
Caking test
Five compaction cycles measuring cake fraction (Cake Height Ratio) and cake strength (Mean Cake Strength and Cake Strength Work). Record all three outputs. The combination of a high Cake Height Ratio with high Mean Cake Strength represents the most severe storage risk.
Consolidation and Caking Rig
Apply a defined load for a dwell time that matches the realistic storage duration for your application. Measure the Work to Break after dwell. This is the definitive test for restart reliability after storage - it answers directly whether the powder will discharge on Monday morning after a weekend shutdown.
Compressibility
The stress-dependent packing profile - % Compressibility, Stiffness, Relaxation, and Elastic Recovery - provides the mechanistic underpinning for the caking and consolidation results. High compressibility with low elastic recovery identifies powders most vulnerable to transport-induced behaviour change.
Level 4: Troubleshooting - run when results contradict observed behaviour
Level 4 is not a standard protocol - it is a targeted investigation run when the first three levels have produced results that do not explain the observed production problem, or when an environmental variable is suspected to be driving behaviour.
Targeted variations
- Humidity conditioning - repeat Level 1 and Level 2 measurements on powder conditioned at elevated relative humidity. Many powders that appear benign under laboratory conditions show substantially different behaviour under the humidity levels of a production environment or storage facility.
- Extended dwell times - repeat the Consolidation and Caking Rig at dwell times that bracket your actual storage duration. If Work to Break increases strongly with dwell time, storage duration is the critical variable.
- Different consolidation loads - repeat compressibility at loads representative of different silo fill heights or stacking configurations. If behaviour is load-sensitive, maximum hopper fill height becomes a process parameter to control.
- Sequential testing across multiple cycles - repeat the full PFSD test sequence several times in succession. If Flow Stability changes across repeat runs, the powder is undergoing progressive structural change during handling.
The minimum viable fingerprint for routine QC
For incoming quality control and supplier comparison, a full four-level characterisation is often unnecessary. The Minimum Viable Fingerprint provides reproducible multi-parameter data that spans the major risk domains in minimum time:
- Standard Cohesion (CI + Bridging Factor)
- PFSD - at minimum record Comp10, Comp100, and Flow Stability
- Conditioned Bulk Density
- One storage metric: either the cyclic Caking test or the Consolidation and Caking Rig at a standardised dwell time
Run this set consistently on every incoming batch. Trend the data over time. Differences that are invisible to single-point static tests - and that predict production variation - become visible across a well-maintained dynamic database.
