From Handling to Interpretation: Quantifying Transfer Dynamics of Smokeless Powder Residues

A recent study published in the journal Science & Justice addresses a critical gap in forensic science by quantifying the transfer of smokeless powder (SLP) residues to hands following direct handling (1). Using a streamlined filter-and-shoot method, Matteo Gallidabino, a Lecturer in Forensic Chemistry at King’s College London and a core member of the King’s Forensics group, and his team measured levels of three common additives, diphenylamine (DPA), dibutyl phthalate (DBP), and ethyl centralite (EC), and found a strong linear relationship between the amount handled and residue recovered (1,2). These data enabled the development of regression models to estimate expected residue levels, supporting more robust and evidence-based interpretations in investigations involving suspected SLP handling.

As part of “From Sample to Verdict,” Gallidabino sat down with LCGC International to discuss his study’s findings.

What prompted this study on smokeless powder residue transfer, and why was it important to focus specifically on low-order explosives like smokeless powders (SLPs) rather than the more commonly studied high-order explosives?

This study was motivated by a major knowledge gap in forensic science. Although explosive residue analysis is routinely conducted in the aftermath of bomb-related incidents, we have very little data on how those residues get onto a person’s hands, especially for low-order explosives like SLPs.

SLPs are of special forensic interest because they are widely accessible and frequently used in improvised explosive devices (IEDs), such pipe bombs. For instance, recent data from the U.S. Bomb Data Center show that SLPs were the main charge in up to 6% of all recorded explosions in 2023 (3). Despite this, most research has focused on high-order explosives like TNT or RDX, which are more relevant in military rather than civilian contexts.

Our aim was to fill this evidential gap by generating empirical data on residue transfer from SLPs during realistic handling scenarios. This represents an important step toward developing more robust and transparent activity-level interpretation, helping forensic experts assess not just whether residues are present, but how they may have been deposited, and ultimately, whether the suspect did or did not handle an IED.

Your study identified diphenylamine (DPA), dibutyl phthalate (DBP), and ethyl centralite (EC) as target compounds. What criteria were used to select these additives, and how representative are they of typical SLP formulations?

We selected DPA, DBP, and EC because they are three of the most used organic additives in modern SLP formulations. These compounds serve key functional roles: DPA and EC act as stabilizers, while DBP is used as a plasticizer. All three are typically present at concentrations high enough to be reliably detected in trace analysis.

Importantly, these additives are also chemically stable and easily ionizable, making them highly compatible with a range of chromatographic and mass spectrometric techniques (especially the GC–MS method we ultimately selected). This analytical robustness was particularly valuable for our study, which aimed to ensure sensitive, reliable detection of residues across multiple samples.

Admittedly, SLPs contain a broader range of chemical constituents. However, while no single set of compounds can fully represent the diversity of all SLPs, DPA, DBP, and EC, they are broadly representative. Studying their transfer characteristics provides a strong starting point for understanding residue behavior and offers insights that can be cautiously generalized across many formulations encountered in forensic casework.

Can you walk us through the filter-and-shoot method you developed for residue collection? What advantages does this streamlined approach offer over traditional sampling and analysis techniques?

The "filter-and-shoot" method is a simplified sample preparation technique we developed to maximize recovery while reducing complexity. After swabbing the participant’s hands, the residues are back-extracted in a small volume of methanol, filtered, and directly injected into the GC–MS, without any need for evaporation or pre-concentration to lower the detection limit.

This approach has several advantages:

• Speed: minimal sample preparation, which is ideal for high-throughput workflows.
• Reduced losses: fewer handling steps, which is beneficial for preserving trace-level analytes.
• Reproducibility: good inter-day repeatability, which is crucial for comparative studies.

It works particularly well for relatively clean matrices, such as hand swabs collected shortly after handling. For more complex or contaminated samples (such as gloves or environmental surfaces) more selective clean-up methods like solid-phase extraction (SPE) may be better suited. In fact, our follow-up work compared filter-and-shoot with SPE to assess performance in such contexts (4).

Overall, filter-and-shoot is a pragmatic, reliable option for analyzing large numbers of hand swabs in forensic experiments, offering a balance between simplicity, sensitivity, and reproducibility.

One of your key findings was a strong linear relationship between the quantity of SLP handled (MSLP) and the quantity transferred to the hands (q_T). How might this relationship be leveraged in real-world forensic casework?

This relationship is important because it establishes a quantifiable link between the amount of explosive material handled and the amount of residue recovered. In forensic casework, this kind of information can support evidence interpretation in two main ways:

1. Contextualizing trace findings during investigation: If a certain amount of residue is recovered from a suspect’s hands, our data can help assess whether that level is more consistent with direct handling of bulk powder, or with a lower-contact scenario such as indirect transfer or environmental exposure. This enables more informed and objective decisions during the early stages of an investigation.
2. Supporting structured interpretation at trial: once the case reaches trial, our data provide an empirical basis for assessing how likely it is to observe a given residue level under different activity-based hypotheses (for instance, whether the suspect handled an explosive device or not). This enables more transparent and consistent reasoning when evaluating competing accounts.

Of course, real-world interpretation must also consider additional factors like time since contact or handwashing. Nevertheless, this foundational relationship between MSLP and q_T marks an important step toward evidence evaluation grounded in data rather than assumption.

Your analysis revealed that individual handler variability (HID) had a significant effect on residue transfer. What factors do you believe contribute to this variability, and how can forensic practitioners account for it during interpretation?

Handler variability (HID) was indeed significant, which aligns with findings from other trace evidence disciplines, such as DNA.This variability likely stems from a combination of physiological and behavioral factors, including skin hydration, sweat composition and how individuals handle materials. These factors are rarely controlled, or even observable, in casework, making HID a particularly challenging source of uncertainty.

From an interpretive standpoint, this underscores the need to treat residue levels probabilistically rather than deterministically. The same quantity of residue could arise from different activities depending on who was involved. Forensic practitioners should therefore avoid rigid thresholds and instead adopt interpretive frameworks that explicitly account for individual variation.

To support this, we recently presented a Bayesian Network model at the latest European Association of Forensic Science (EAFS) conference (the largest forensic science meeting in Europe) which incorporates HID alongside other contextual factors, such as time since contact (5,6). This type of modeling enables a more transparent and realistic evaluation of competing activity-level propositions.

Ultimately, acknowledging HID enables more defensible inferences and helps prevent over-interpretation of trace findings in isolation.

While the type of SLP (TSLP) had a comparatively weak effect on q_T values, were there any differences between formulations that surprised you or that might have forensic relevance in certain scenarios?

Yes, we were somewhat surprised that TSLP had only a limited effect on q_T. Intuitively, one might expect different SLP formulations to produce markedly different transfer behaviors.

That said, it’s important to note that the SLP we selected, while distinct, were not dramatically different in their content of the three target additives (DPA, DBP, and EC). This likely contributed to the overall similarity observed in their transfer profiles. To better understand the role of TSLP in influencing q_T, further studies involving more chemically diverse SLP formulations are needed.

Still, we did observe some SLP-specific effects, both in the total amount of residue transferred and in the composition of the residues. Although these differences were modest, they suggest that formulation can play a role, even if it was not the dominant factor under the conditions we studied.

How did gas chromatography–mass spectrometry (GC–MS) and liquid chromatography–tandem mass spectrometry (LC–MS/MS) compare in terms of analytical performance for detecting SLP residues, and what guided your choice of instrument for the final analysis?

We initially compared GC–MS and LC–MS/MS across a broader panel of SLP additives, including both thermally stable and thermally labile compounds. As expected, LC–MS/MS performed better for the more thermally labile analytes, which tend to degrade or volatilize poorly under GC conditions.

However, for the three target compounds selected for our study (DPA, DBP, and EC), GC–MS showed superior performance, particularly in terms of limits of detection (LODs) and repeatability. These additives are thermally stable and sufficiently volatile, making them well suited to GC analysis.

This is why we ultimately chose GC–MS for the study. It also offered practical advantages for our “filter-and-shoot” workflow: the method was rapid, cost-effective, and compatible with direct injection of alcohol-based extracts, making it ideal for high-throughput analysis of hand swabs in a controlled setting.

That said, LC–MS/MS remains a valuable tool in the broader context of SLP analysis, both at the bulk and trace levels. It is particularly advantageous when targeting a wider range of additives or working with more complex matrices. Our findings simply reinforce that method selection should be guided by the analytes of interest and the practical constraints of the application.

In what ways do you envision the regression models and empirical q_T data developed in this study shaping future forensic protocols or courtroom interpretations involving suspected SLP handling?

Our findings offer a much-needed empirical foundation for interpreting trace-level findings involving SLP residues. Historically, activity-level assessments have often relied on expert judgment with limited data support. This study helps shift the field toward more transparent, data-driven interpretation.

In particular, our findings allow forensic scientists to:

• Contextualize residue level by helping assess whether a given q_Tis more consistent with direct handling or with lower-contact scenarios such as secondary transfer.
• Support expert testimony in Court by grounding opinions in empirical data rather than subjective judgement or rigid thresholds.
• Develop standardized protocols for both sample collection and interpretation that explicitly account for variability and uncertainty.

Ultimately, we hope this work contributes to the development of more robust and transparent evaluative frameworks, similar to those now used in DNA or fingermark evidence, helping ensure that trace findings are interpreted with appropriate nuance and scientific rigor.

References

1. Standley, T.; Putruele, C.; Lambert, G.; et al. Quantification of Smokeless Powder (SLP) Additives on Hands After Direct Handling of Bulk Samples via a Filter-and-Shoot Method. Sci. Jus. 2025, 65 (4), 101262. DOI: 10.1016/j.scijus.2025.101262
2. King’s College London, Dr. Matteo Gallidabino. King’s College London. Available at: https://www.kcl.ac.uk/people/matteo-gallidabino (accessed 2025-07-28).
3. United States Bomb Data Center, Explosives incident report (EIR), 2023.
4. Putruele, C.; Kelly, C.; Standley, T.; et al. Evaluation of Solid-phase Extraction (SPE) for the Analysis of Smokeless Powder (SLP) Residues on Hand Swabs with a Proof-of-Concept Comparison between Gloves and Bare Hands. In preparation, 2025.
5. Standley, T.; Putruele, C.; Gavriilidi, I.; et al. Towards a Robust Activity-level Assessment in Explosive Residue Analysis: Assessment and Modelling of the Primary Transfer of Smokeless Powder (SLP) Residues on People's Hands, in: EAFS 2025, Dublin, 2025.
6. Gavriilidi, I.; Standley, T.; C. Putruele, C.; et al. Can we Identify Bomb Handlers? Proposition of a Bayesian Network (BN) to Model the Traceogenesis of Explosive Residues on Hands and Support Activity-level Significance Assessment, in: EAFS 2025, Dublin, 2025.