Results and Discussion

Depth of penetration assessment

Figure 3 shows the depth of penetration for the ‘as received’ material (serial one) and samples which has been subjected thermal cycling (serial 2). Across all samples used during serial one, the depth of penetration remains within a range of between 130 and 155 mm (17.54% difference). The velocity during baseline firings equated to an average of 239.1 ms− 1 with a variance of 8.88% between the lowest (226 ms− 1) and highest (247 ms− 1) velocities recorded. Comparison of the velocity and depth of penetration raw data shows non-linear trends (Annex A – Depth of Penetration and Velocity Raw Data), highlighting no correlation between speed and depth of penetration measurement, as evidenced by the lowest velocity recording (Block 1–132 mm) and the highest velocity recording (Block 6–133 mm). The raw data instead highlights an input velocity of 245 ms− 1 resulted in the highest depth of penetration measured at 155 mm, whereas the lowest depth of penetration (129 mm) was achieved at 241 ms− 1 providing numerical evidence of concerns regarding ‘as supplied’ material calibration and/or homogeneity.

Fig. 3: Perma-Gel Depth of penetration comparison between as-supplied material (Serial 1) and material which has undergone thermal cycling(Serial 2) in order of severity

On completion of the recorded baseline, an assessment of the change in depth of penetration measurement as a result of melt and re-cast was undertaken. Figure 3 shows that after 5 melts an average change of 9 mm (6.25%) was witnessed, whilst 10 and 15 melt cycles provided average results of 20 mm (14.79%) and 22 mm (17.18%) respectively when compared to the as supplied material depth of penetration values. These findings highlight that the material is no longer behaving like the original after being melted and as a result, may produce falsified data. The penetration depth of viscoelastic materials is traditionally governed by mechanical properties such as density, elasticity, viscosity, yield strength and resistance to deformation [2, 11, 14, 15, 23, 25, 58–60], a 12.74% increase suggests that the material has become less resistant (softer). It is also noted that a 12.74% increase in wound ballistics is beyond ‘typical’ experimental tolerances which are referenced within the widely accepted Fackler 10% gelatine calibration method which explicitly states +/- 0.5 cm to the 8.5 cm depth of penetration calibration value [14, 18, 19, 43]. Concerns are therefore raised with the materials ability to withstand higher volumes of melt cycles for repeatable experimentation.

Cavitation depth and diameter differences

To further investigate the influence melting and recasting the samples had on the materials ability to provide the user with reliable source data, a comparison of pre and post melt cycle maximum temporary cavity diameter was conducted. This area of investigation was chosen primarily due to the temporary cavity’s importance in the assessment of energy transfer from projectile to tissue simulant and how that energy can damage structures beyond that of the immediate projectile pathway (permanent cavity) [61–69].

The temporary cavity has been proven to provide the most accurate real-world indicator of the amount and rate of energy that a projectile transfers into the material, as evidenced by high strain rate impacts resulting in greater energy transfer and therefore larger temporary cavities. This provides the user with an assessment on the potential for secondary indirect injuries through stretching or tearing of tissue commonly associated with organs, vessels and nerves [64–69]. Further, the temporary cavity assessment provides the user with increased data fidelity when compared to assessment of entry wounds alone. Assessment of entry wounds remains important, but different types or calibres of ammunition may result in very similar appearance on the sample surface, however the severity within the material is often influenced by velocity, projectile design (fragmenting, tumbling etc.), and stability during flight to, and travel within the sample [67, 70–72]. The outcomes of the ability to assess temporary cavity dynamics ultimately influences enhancements to projectile design, medical treatment and the ability to replicate forensic crime scenes where penetration or perforation of the human body as occurred [11, 64, 67–69, 73].

It is therefore vital to understand any potential impact on the ability for Perma-Gel to accurately provide users with an assessment on the temporary cavity which include not only material performance but also its ability to maintain transparency throughout the experimental procedure.

Table 4 highlights the values recorded during both depth of penetration and cavitation assessment. Assessment of the average cavity diameter data provides averages of 13.63 mm, 12.15 mm and 14.55 mm during baseline firings for the samples identified for 5, 10 and 15 melt cycles respectively. When compared to the post melt values of 13.42 mm, 14.62 mm and 14.50 mm this data provides evidence that an average increase of 5.36% in average cavity diameter after melting highlights a reduced reliability and repeatability of the material, this is an important finding when considering energy-transfer as small differences in material performance will result in later tests within the same series no longer reflecting earlier tests. This therefore results in reduced comparability between studies which may result in misinterpreted forensic interpretation, indirect comparisons between ammunition performance (where ammunition may appear to be more damaging), and ultimately mislead manufacturers to either falsely claim compliance with design requirements or focus on material enhancement in areas which are not required.

Number of Melts | Sample No. | Serial 1 (mm) | Serial 2 (mm) | Difference between Serial 1 and Serial 2 DoP (%)
5 |  |  |  | 
 | Block 7 | 150 | 143 | −4.67
 | Block 8 | 125 | 142 | 13.60
 | Block 9 | 125 | 140 | 12.00
10 |  |  |  | 
 | Block 1 | 125 | 163 | 30.40
 | Block 2 | 140 | 142 | 1.43
 | Block 3 | 134 | 153 | 14.18
15 |  |  |  | 
 | Block 4 | 140 | 169 | 20.71
 | Block 5 | 125 | 150 | 20.00
 | Block 6 | 125 | 138 | 10.40

Despite the 5.36% increase being a small difference when reviewing the raw data, it does provide concern that the long-term use of Perma-Gel undergoing melting and re-casting within the conditions used in this study does result in deterioration of the mechanical properties. The data suggests that the material has become more elastic meaning that the material tends to overpredict the temporary cavity as the number of melt cycles increases. It is therefore recommended that during future use of this material this finding be factored into data error to mitigate against misleading findings being published within the wider academic literature. Additionally, the number of melt cycles is recorded to capture any degradation in performance.

To further investigate this finding, an assessment of the depth at which the temporary cavity maximum diameter occurred was undertaken to further contextualise its impact. In this assessment, pre-melt samples provided average depth of 61.12 mm, 60.26 mm and 63.46 mm (samples identified for 5, 10 and 15 melt cycles respectively) using factory supplied material versus the 67.91 mm, 53.44 mm, and 57.88 mm measurements for 5, 10 and 15 melt cycles respectively. These findings highlight a non-linear change relative to the number of heating cycles which is hypothesised to be due varying severities of molecular change during heating, but it is also noted that variance in input velocity may have influenced these findings. Additionally, this finding further emphasises that the materials mechanical performance has changed when compared to the original factory supplied blocks and provides further evidence that the melt-cycles have altered the polymer structure reducing in a material which is slightly softer (reduced yield strength) which also raises concerns around maintained use of a calibrated test medium during a prolonged test series.

When reviewing the findings surrounding maximum temporary cavity diameter changes post melt-cycle, it is recommended that a limit be placed on the number of times Perma-Gel can be melted and recast before disposal is required to ensure reliable and repeatable findings are produced and data inaccuracy within the open literature is minimised.

Material use

This study has reported that the effects of melt/recasting cycles have a detrimental impact on Perma-Gels performance, however some additional concerns with the materials use have been noted. The literature does not currently provide detailed review of the materials constituent parts and their role in enabling the material to perform advantageously (exact polymer type, crosslink density, fillers and stabilisers etc.). This is important to understand as small changes in polymer mixes can dramatically change the materials viscoelastic properties (elasticity), yield stress and damping rate which all contribute to the measurement of both permanent and temporary cavity, and depth of penetration measurement [15, 74–79]. Additionally, in most cases, the material is used ‘as supplied’ and no verification of manufacturers claims appears to be undertaken prior to use in experimental campaigns [24, 34, 35, 39]. This introduces risk into test programmes by not establishing a baseline. Mitigation is provided in the form of a calibration certificate from the supplier stating that the batch of material supplied has been evidenced to conform to the FBI protocol, however from a researcher prospective assumptions must then be used including those concerned with material quality control, homogeneity, and polymer crosslinking consistency which can be summarised under ‘batch to batch variation’.

Inconsistencies in polymer crosslinking (main driver in controlling elasticity and energy dissipation) have been reported to result in variable stiffness and failure behaviour of other gelatines used within this field of study and should therefore be monitored throughout test to ensure molecular breakdown is not attributing to inaccuracies in data [80–83]. Additionally, lack of universally accepted calibration or lot-certification for synthetic gels may lead to material being received which is visually and identifiably identical but have differences in mechanical responses leading to inconsistencies within the test series and potentially across labs [12, 43, 53].

Concerns were also held during the process of melting and recasting of each sample which have been noted during this study, most notably thermal degradation of the polymer, incomplete homogenisation, microbubbles, material contamination and loss of clarity.

Investigation of the effects of thermal degradation of Perma-Gel was the primary aim for this study and has reported that under the test conditions used in this study, the data provided by this work confirms the hypothesis that increased numbers of melting cycles may detrimentally impact the performance of the material [12, 24, 33, 39]. The initial hypothesis centred on the logic that polymer chains are often susceptible to heat causing chain-scission), resulting in reduced elastic modulus and tensile strength [84–86]. Whilst this is an important consideration, thought must also be given to the degradation of plasticisers and stabilisers within the material which may have caused shifts in viscosity resulting in differences in mechanical performance, with the findings within this study pointing toward changes to viscosity [87–89]. This work, whilst providing an introductory understanding to the effects of thermal cycling on Perma-Gel is not without its limitations. The primary limitation of this study is that no physical testing occurred between heating cycles which may provide additional understanding on the degradation mechanisms and its severity. Additionally, it also remains unclear whether the physical work done during impact of a projectile impacts the rate of degradation when paired with thermal cycling.

Optical clarity was also monitored during the melt/recast cycles via visual inspection to ensure no micro-bubbles were being introduced to the material thereby being a detriment to the ability to examine high-speed video footage, but also introducing the potential for premature material failure due to localised weak spots [14]. During the manufacture of samples and pre/post-test visual inspection no micro-bubbles were found the be present. However, slight yellowing to one sample which had undergone 15 melt-recast cycles was apparent which is predicted to be as a result of oxidation and breakdown of stabilisers and plasticisers within the material resulting in discolouration (Fig. 2) [88, 90, 91]. This phenomenon was not apparent on the other two samples which underwent the same amount of cycling and were exposed to the same conditions, thereby confirming the concerns surrounding homogeneity of the material during the melt/recast process and the potential for mechanical drift . .

During the melt and recast process, it was observed that moderate quantities of fumes were emitted from the blocks, which did not reduce with the amount of cycles it underwent potentially linking the findings listed above to the marginal decreases in performance presented in Table 5. The fumes were cloudy in nature presenting an opaque visual reference and was acrid in smell, which required the oven to be relocated to an area with increased ventilation and additional personal protective equipment utilised to mitigate against inhalation. It is recommended that this observation be noted and considered during future works to ensure experimentation is conducted safely.

Number of Melt Cycles (Block No.) | Average Input Velocity Pre Melt (m/s) | Pre-Melt Temp Cavity Depth (mm) | Pre-Melt Temp Cavity Dia. (mm) | Average Input Velocity Post Melt (m/s) | Post-Melt Temp Cavity Depth (mm) | Post-Melt Temp Cavity Dia. (mm)
5 (7) | 245 | 66.65 | 15.84 | 228 | 59.81 | 13.78
5 (8) | 242 | 57.81 | 12.52 | 222 | 71.41 | 12.69
5 (9) | 244 | 58.91 | 12.52 | 230 | 72.50 | 13.78
10 (1) | 226 | 58.18 | 12.15 | 236 | 46.04 | 15.23
10 (2) | 229 | 61.12 | 12.15 | 230 | 59.09 | 14.5
10 (3) | 246 | 61.49 | N/A3 | 232 | 54.38 | 14.14
15 (4) | 232 | 65.17 | N/A4 | 230 | 58.36 | 16.31
15 (5) | 241 | 58.18 | 12.15 | 226 | 59.09 | 13.78
15 (6) | 247 | 67.02 | 16.94 | 224 | 56.19 | 13.41

Contamination during melting and recasting was also considered but not witnessed, with the ball bearing and any sabot fragmentation carefully removed using a surgical scalpel before the melt cycles began. This was undertaken using the human eye, with no microscopy being utilised to confirm successful removal at a molecular level. Although minimised, the risk of introducing contamination to the material during the melting cycles is one which requires careful control. Metal flakes caused by fragmentation have the potential to act as nucleation sites for tearing or local stiffening, but they also have the potential to increase the rate of oxidative degradation [92, 93]. This study has utilised AISI 420 grade stainless steel ball bearings that have a reported hardness of 52–55 HRC Rockwell Scale to minimise against the potential for fragmentation [94] and as such concern was discounted for this study.