Analytical microplastic recovery: Are smaller particles really harder to recover, or is the filter just saturated?

There is general consensus in the microplastic communitity, that smaller microplastics are more easily lost during sample pre-treatment; and there is evidence for this (Fig. 1)...

Fig. 1 - Excerpt of a scientific paper, possibly showcasing confusion between true and analytical recovery. The question is: were larger particles more 'effectively recovered' or were they simply more easily observed because they were affected by particle agglomeration?

At least, in terms of analytical microplastic recovery... See, there's a difference between true microplastic recovery and analytical microplastic recovery. True microplastic recovery defines how many particles were successfully transferred from the sample onto the filter membrane. Analytical recovery, on the other hand, is simply put, the observed recovery rate.

For this reason, analytical recovery is strongly influenced by particle agglomeration, coverage, and overlap. For obvious reasons, smaller particles are especially exposed. Particle agglomeration will exaggerate the analytical recovery of larger particles and underestimate that of smaller particles. This effect is even more pronounced when a filter is highly saturated with particles (Fig. 2).

Fig. 2 - Figure reproduced from Hagelskjær et al. (2023), showcasing that particle agglomeration is more likely to occur when a filter is more saturated and that this effect 'favors' the identification of larger particles.

We figured this out by staging a simple experiment, wherein we transferred microplastic fragments from a filter into water and back onto a new filter, 5 times. Three replicates of two different size groups were prepared: PE fragments from 5-50 µm and PP fragments from 50-500 µm. The experiment demonstrated that there was no difference in the recovery rate of smaller or larger microplastics (Fig. 3).

Fig. 3 - Figure reproduced from Hagelskjær et al. (2023). The diagram plots average recovery rate (RR) of fine microplastics (5-50 µm) and small microplastics (50-500 µm) within specified size groups, based on 12 transfers. The total number of transferred particles within these groups is marked by triangles.

We did, however, discover that the analytical recovery of smaller microplastics decreased as the saturation on the filter surface increased. (Fig. 4).

Fig. 4 - Figure reproduced from Hagelskjær et al. (2023). The figure illustrates that the relative particle size distribution (PSD) of small microplastics (50-100 µm) increased with each transfer, while the PSD of larger microplastics (100-150 µm) decreased. We interpret this trend as a result of decreasing filter saturation, thus resulting in a decreased likelihood of particle agglomeration.

However, this observation only applied if the filter was saturated above a certain point. The percentage of the filter covered by particles was the determining factor, which we named 'A%'. What we discovered was that at A% > 5, the analytical recovery suffered a substantial decrease (Fig. 5).

Fig. 5 - Figure reproduced from Hagelskjær et al. (2023). When A% was above 5%, a significant drop in the analytical recovery was observed.

Applying this observation to environmental samples, we knew that our filters used in microspectroscopic analysis should not exceed 5% A. We determined the A% of some previously analyzed samples and found that samples that we had previously discarded due to particle agglomeration (from failed sample pre-treatment) demonstrated A% > 5. On the other hand, some of our most successfully treated samples demonstrated between 3-5% A. With this information, we came up with a 'visual guideline' for acceptable A% on a filter meant for microspectroscopic analysis. (Fig. 6).

Fig. 5 - Figure reproduced from Hagelskjær et al. (2023). Visual guideline for target A%, which should not surpass 5% on filters intended for microscopic analysis of microplsastics below 500 µm in diameter.

Conclusively, if A% remains below 5, microplastics should have sufficient spread to reliably represent all present size ranges whilst avoiding excessive MP loss. We therefore adviced against exceeding A% > 5 on any filter intended for microscopic analysis of MPs (< 500 µm). Finally, we suggested that microplastisc recovery experiments should try to respect the following recommendations: (i) The precise number of initially added spike microplastics should be known. (ii) The recovery experiment should be replicated at least three times. (iii) Spike with a significant number of fragments in different sizes, in order to establish recovery rate within specified size groups. If the use of fragments is not feasible, spike with at least two different size groups of spheres or beads. (iv) The size of the spike microplastics should match the approximate size of the relevant environmental microplastics.

Read the full study: https://doi.org/10.1186/s43591-023-00071-5