How to accurately measure particle size in agglomerated powders

Key takeaways:

The most common particle sizing techniques for agglomerated materials are laser diffraction, sieving, and image analysis.
An instrument’s suitability for agglomerate assessment depends on its physical design and its underlying measurement principles, including how particles are dispersed and detected as well as how size data are reported.
Often, a combination of techniques, along with careful dispersion protocols and morphological insight, yields the most accurate and actionable view of a material’s true particle size and structure.

Agglomeration is a pervasive challenge across powder processing industries, occurring during key steps such as drying, granulation, and milling. These clusters of smaller primary particles can drastically impact powder flow, dissolution, and product performance. From a particle sizing perspective, agglomerates are notoriously difficult to characterize due to their diverse structures — from fragile, loosely bound clusters to hard, fused masses — and their irregular shapes and broad size distributions.

The most common particle sizing techniques — laser diffraction, sieving, and image analysis — can provide insight into these granular ensembles; however, each technique can produce data that may not be representative of the process if not thoughtfully selected. By understanding the strengths and limitations of these methods, powder handlers and process engineers can make more informed decisions to ensure reliable, application-relevant particle size data that targets the goal of the testing and the material performance.

Understanding agglomerates

The vocabulary surrounding agglomerated material can be complex on its own. These structures are known by various names, including aggregates, flocs, clusters, and granules, depending on the context or industry. However, regardless of terminology, agglomerates are ultimately defined by the nature and strength of the forces that bind the primary particles — ranging from weak physical interactions (e.g., van der Waals forces, electrostatic attraction, or liquid bridging) to strong, often irreversible bonds formed through sintering or recrystallization bridging, where partial surface dissolution and re-solidification at particle contact points create solid bridges (see Figure 1).

This distinction is critical and can often be done with the assistance of general microscopy imaging, as the type of agglomeration present significantly influences both how the material should be prepared for testing and which analytical technique is most appropriate. Ideally, characterization should reflect the state of the material as it exists in the actual processing environment — whether it is handled dry or in suspension, and whether it undergoes high or low dispersion energy during manufacturing or transport.

Figure 1. (Left) Loosely bound agglomerates formed through weak physical forces, showing visible gaps and irregular clustering. (Right) Hard agglomerates (circled in red) formed through recrystallization or solid bridging, where primary particles are fused.

In the world of particle sizing, the initial impulse is often only to consider and quantify the observed intact clusters, especially if they dominate the material’s visible structure. However, this approach can overlook the role of small primary particles, which — with their high surface-area-to-volume ratios — readily form agglomerates through weak interactions. In the processing world, excessive agglomeration can indicate underlying issues such as supplier variability or milling processes. Therefore, depending on the application and the questions being asked, it may be just as important to evaluate and analyze the primary particles themselves. It should not be assumed that the same particle-sizing technique and methodology can be used interchangeably for assessing agglomerates and primary particles. A combination of techniques may be necessary to obtain a comprehensive understanding of the material.

Methods for measuring particle size in agglomerated materials

An instrument’s suitability for agglomerate assessment depends on its physical design and its underlying measurement principles, including how particles are dispersed and detected as well as how size data are reported. Overlooking these factors can lead to misleading results, especially when working with limited sample quantities or powders with broad, heterogeneous size distributions.

For example, some techniques require dispersing the sample in a liquid, applying air pressure, or introducing mechanical shear — all of which can disrupt fragile agglomerates or fail to break apart more strongly bound clusters. If the testing environment is incompatible with the agglomerate type and testing goal, the resulting measurements may not reflect the true structure of the material that was observed. Likewise, the detection and reporting approach — such as number-based versus volume-based distributions — can skew interpretation. Volume-based methods tend to overweight larger agglomerates, potentially masking fine particles, while number-based outputs may understate the impact of a few large clusters.

Understanding how these factors play out in common sizing techniques — such as sieving, laser diffraction, and image analysis — can help ensure that the method selected aligns with both the material's characteristics and the measurement goals.

Sieving

While limited in resolution, sieving is the most traditional method of particle size analysis and remains widely used for its simplicity, low cost, and direct physical separation by size. The process involves passing a known quantity of powder through a stack of wire mesh screens with defined declining aperture sizes and measuring the mass retained on each screen to generate a size distribution (reported by percentage of total sample mass). For agglomerated materials, sieving offers the advantage of analyzing large sample volumes compared to most other particle sizing techniques, providing a robust representation of heterogeneous powders. However, it also presents challenges.

Figure 2. Macro image illustrating the tendency of fine powders to form low-density agglomerates that bridge or obstruct sieve apertures due to dry screening. Such behavior can bias particle size distributions by preventing accurate passage through smaller mesh sizes.

If agglomerates remain intact, sieving can provide a direct measure of their physical size, but soft or loosely bound clusters may break apart under mechanical vibration or air entrainment, resulting in data that reflect smaller primary particles rather than the intended agglomerate structure. Conversely, cohesive fine powders may form stable agglomerates that resist passing through fine meshes or clog apertures, shifting the apparent distribution toward coarser size fractions (see Figure 2). And while this technique physically separates particles or agglomerates, it relies on a geometric assumption: particle size is approximated based on the smallest dimension that allows a particle to pass through a square opening, effectively an equivalent spherical diameter. This can introduce inaccuracies for irregularly shaped particles or agglomerates, as the farther their geometry deviates from a sphere, the more likely they are to orient unpredictably during passage, leading to biased size measurements.

Despite these challenges, sieving remains effective for larger, mechanically robust agglomerates such as granules or pellets. And in systems where larger particle fractions dominate performance — such as in flow, compaction, or dosing — sieving remains a practical and robust tool, particularly when minimal dispersion and simple size thresholds are sufficient for decision-making.

Laser diffraction

Laser diffraction is one of the most widely used particle sizing methods due to its speed, reproducibility, and broad dynamic range — typically spanning from submicron particles to several millimeters. The technique works by measuring the angle and intensity of light scattered by a dispersed particle cloud. This scattering pattern is then used to calculate an equivalent spherical diameter, generating a volume-based particle size distribution.

The method can be performed in either a wet or dry dispersion mode, suspending a small quantity of powder (typically approximately 50 mg to 1 gram, depending on the setup) in a liquid or entraining it in an air stream during measurement. While this flexibility is useful, it also introduces challenges when working with agglomerated materials. Soft clusters are particularly susceptible to disruption during dispersion, and preserving their structure can be difficult regardless of the medium.

Conversely, insufficient dispersion may result in persistent agglomerates being recorded as oversized particles, and because laser diffraction reports a volume-based distribution, the presence of large agglomerates can dominate the distribution even if they are relatively few in number. This makes sample preparation critical, as the forces applied — through sonication, surfactants, or compressed air — directly influence the reported size, although systematically varying this energy input — through pressure titration in dry systems or incremental sonication in wet systems — can give an indication of how strongly agglomerates are bound.

Figure 3. The red curve shows the presence of intact agglomerates, while the green and blue curve reflects smaller primary particles released by dispersion. This illustrates how sample preparation influences measured size and highlights laser diffraction’s sensitivity to agglomerate integrity.

A practical illustration of this is comparing particle size distributions before and after applying dispersion techniques. Without sufficient dispersion, intact agglomerates shift the distribution toward larger sizes; after dispersion, the size distribution often shifts leftward, presumably reflecting the breakup into smaller primary particles. While this shift demonstrates the effect of preparation, the technique alone cannot determine whether the original sample contained fused or soft agglomerates — only that dispersion altered the observed size (see Figure 3).

Despite these limitations, laser diffraction is well-suited for routine particle size monitoring in manufacturing and quality control environments. It offers high throughput and robust repeatability but lacks resolution for assessing particle shape, surface texture, or internal structure. With careful attention to dispersion settings, including energy ramping studies, it remains a powerful tool for tracking particle size trends and identifying changes in agglomeration state over time.

Image analysis

Image analysis offers a visual and morphology-sensitive approach to particle sizing, setting it apart from laser diffraction and sieving. Using either static microscopy or dynamic flow imaging, this method captures two-dimensional projections of particles, enabling measurement of size, shape, and surface texture. In many instances, it can distinguish agglomerates not only by their size but also by morphological features such as aspect ratio, circularity, and surface complexity.

This makes image analysis particularly valuable for agglomerated materials, where both size and structure influence performance. By revealing internal and surface features, it can help differentiate between loosely packed clusters, densely bound agglomerates, and fused primary particles. In practice, classification thresholds can be user-defined based on shape metrics — such as circularity or aspect ratio — to separate agglomerates from individual particles or debris, allowing for targeted analysis and specification setting.

Figure 4. (Left) Software-identified agglomerates — particulate that has a circularity of less than 0.95. (Right) Software-identified primary particles — particulate that has a circularity >0.95. By applying filters for shape parameters such as circularity, software can distinguish between single particles and irregular fused agglomerates. This allows for both visual confirmation and quantitative reporting of agglomerate content, supporting specification setting in applications where structure influences performance.

For instance, if a spray-dried material tends to produce fused agglomerates under specific processing settings, quantification of this occurrence can be generated by evaluating the particle size distribution on a circularity basis. By definition, the closer a particle’s circularity value is to 1, the more spherical the particle is considered to be, and under this scenario, the more likely it is to be a single particle. Particles with circularity values below 0.95 may be designated by the user in the software to be fused agglomerates, enabling the software to automatically classify and quantify these populations within a sample (see Figure 4). This type of morphological filtering is especially useful in applications where particle shape and structure are tied directly to product performance, such as inhalable powders, catalysts, or controlled-release granules.

However, image analysis also has limitations. It typically operates at lower throughput than laser diffraction or sieving and requires meticulous sample preparation to avoid particle overlap, poor focus, or agglomerate disruption. Proper dispersion and uniform spreading are essential to ensure representative measurements. Furthermore, because image analysis often examines fewer particles than volumetric methods, statistical confidence may be reduced unless a sufficient number of particles are analyzed, which in turn can be time prohibitive.

Despite these challenges, image analysis remains a critical tool in formulation development and troubleshooting. It complements other particle sizing methods by quantifying size and revealing particle morphology, providing deeper insight in cases where particle structure and shape influence product behavior.

Putting it all together

Agglomeration is a common yet complex feature of powder systems, and its presence complicates both material behavior and particle size characterization. As shown, no single sizing technique can fully capture the nuances of agglomerated structures without deliberate consideration of sample preparation, measurement principles, testing goal, and data interpretation. Commonly used techniques such as sieving, laser diffraction, and image analysis each offer unique advantages, but they also come with limitations that become especially apparent when working with soft, fused, or morphologically irregular agglomerates.

Ultimately, reliable agglomerate assessment depends on matching the right tool to the right question. Whether the goal is to monitor processing consistency, identify formulation issues, or set meaningful specifications, a clear understanding of how each method interacts with agglomerates is essential. In many cases, a combination of techniques — paired with careful dispersion protocols and morphological insight — will offer the most accurate and actionable view of a material’s true particle size and structure.