Distribution and properties of gunshot residue originating from a Luger 9mm ammunition in the vicinity of the shooting gun

Article Outline

• Abstract
• 1. Introduction
• 2. Experimental
  • 2.1. Materials
  • 2.2. Methods
• 3. Results and discussion
  • 3.1. Dependence of the chemical composition of GSR on the distance from the muzzle
  • 3.2. Dependence of the size of GSR on the distance from the muzzle
  • 3.3. Combined inspection into the dependence of size and chemical contents of GSR on the distance from the muzzle
  • 3.4. Assessment of the repeatability of parameters describing metallic particles collected in the vicinity of the shooting gun
  • 3.5. On the mechanism of dispersion of GSR
• 4. Conclusions
• Acknowledgements
• References
• Copyright

Abstract 

Examinations of various features of gunshot residue (GSR) collected from targets in a function of the shooting distance as well as from hands and the forearm, front and back parts of the upper clothing of the shooting person were performed with SEM-EDX. GSR samples were obtained using Walther P-99 pistol and Luger 9mm ammunition of Polish production. The experiments were designed in such a manner that the substrates for collecting GSR reminded the ones usually obtained for examinations within criminal cases. Results of the performed examinations in the form of parameters describing GSR particles: the number of GSR, proportions of their chemical classes as well as their sizes revealed a dependence on the shooting distance both, in the direction of shooting and backwards, i.e. on the shooting person. The analysis of the distribution of particles in the vicinity of the shooting gun may be utilised in description of the general rules of the dispersion of GSR as well as in the reconstruction of a real shooting case.

Keywords: Forensic science, Gunshot residue, SEM-EDX, Distribution of properties of particles in a function of the shooting range

Back to Article Outline

1. Introduction 

Gunshot residues (GSR) are related to the place and time of firing a gun. The analysis of GSR is being routinely performed in order to solve particular problems such as differentiation of entry and exit wounds, estimation of the shooting distance, establishing the kind of ammunition used, tracing the trajectory of a projectile and relating an individual to a shooting incident. Fulfilling any of these tasks may contribute to a shooting incidence reconstruction.

Metallic particles originating from the primer of ammunition are thought to be the most characteristic gunshot residues for their exceptional chemical content and morphology that reveals features of molten and suddenly cooled matter. That makes them highly valuable evidence in relating a person with the fact of using a firearm, however they may also be very informative in other aspects of shooting incidence investigations. Although many sensitive analytical methods can be applied for identification of the characteristic gunshot residues, only scanning electron microscopy with energy dispersive X-ray spectrometry (SEM-EDX) was recognised as the most specific one. It enables an examiner to observe simultaneously the characteristic morphology of a GSR particle and check its elemental content without prior damage of the object and so is a well-established method of identification of this kind of evidence in forensic laboratories. It does not require complicated sample preparation. Among few methods of collecting GSR for examinations with SEM-EDX method the most popular one is lifting microtraces by multiple pressing of an adhesive material to the substrate of interest, e.g. the suspect's hands, face, hair and clothing. One of the important studies concerning the quality of an expertise in the field of GSR examinations by means of SEM-EDX focuses on proficiency testing aiming at the precision and sensitivity of the analytical instrumentation. A great progress was recently achieved in the development of identical artificial GSR test samples fulfilling the requirements of standards ISO 5725 and ISO 13528 [1], [2].

The fundamental problem of GSR examinations for forensic purposes is an evaluation of the analytical results. A classification scheme of metallic particles introduced by Wolten et al. [3], [4], [5], later modified by Wallace and McQuillan [6] was widely accepted by gunshot residue examiners and used for more than two decades. According to the scheme three-component particles containing lead, antimony and barium are unique primer residues. Accompanying two- and one-component particles are indicative for gunshot, since particles of similar elemental contents could have been found in other circumstances. This formal approach is now being under discussion in the view of recent publications on the residue of fireworks, the dust of car brake pads as well as on the residue of the newest non-toxic ammunition. Although still very rarely and in different environment, one can find a single lead–antimony–barium particle that reveals a similar to GSR morphology [7]. This inspired the European GSR examiners to rename the ‘unique’ and ‘indicative’ particles back to ‘characteristic’ and ‘consistent with GSR’, and revise the chemical composition of the appropriate classes of particles [8]. The scheme includes now particles originating from special brands of ammunition marked with gallium or gadolinium (being supplied, e.g. for German and Dutch police) as well as from some non-toxic ammunition types in addition to the traditional ones containing lead, antimony and barium compounds.

However, the classification scheme does not provide answers to questions: what quality and quantity of particles would be significant evidence in certain case and how this particular evidence would contribute to achieving an opinion by the decision-makers. Thus, Romolo and Margot suggested another approach to the assessment of GSR as an evidence called a case-to-case approach, e.g. based on Bayesian principles [9]. In 2005 Avermate presented interesting examples of this manner of interpretation of GSR examinations in casework and demonstrated that despite of lack of numerical data likelihood ratio can be evaluated qualitatively and utilised in formulation of conclusion in the expert's report [10]. Recently, Biederman and Taroni proposed to introduce a probabilistic approach such as Bayesian network for a joined evaluation of two types of evidence: the identification of the firearm (marks on the projectile) and the assessment of the shooting distance (GSR) [11]. The authors provided a theoretical framework for this combined evaluation of the uncertainty of the evidence. For the estimation of shooting distance they suggested visualisation of nitrite compounds as the residue of the propellant using a modified Griess test and counting the number of the revealed particles within patterns of various shooting distances. Unfortunately, the authors used hypothetical numbers instead of empirical data leaving the reader with doubts, whether quantitative results to be obtained in reality would be enough repeatable for the proposed statistical requirements. So far, the shooting distances are in practice estimated by visual inspection of the pattern of interest in relation to a series of test shots patterns. This applies also to patterns visualised by means of analytical instrumentation such as millimetre-X-ray spectrometry [12] and imaging in infrared light [13].

It ought to be admitted that the spatial expansion of GSR, the time of their presence in the crime scene as well as dispersion of their properties depending on the distance from the muzzle are still not thoroughly recognised. Some ideas on the range of GSR and time sedimentation were gathered performing single experiments in controlled conditions, including preparation and distribution of horizontal accumulating targets around the shooter at various heights [14], [15], [16]. In shooting cases, however, often a person that becomes a victim is initially in the upright position. In such cases experts receive for examinations the victim's clothes that acted as vertical targets as well as the shooter's clothes in addition to samples collected from his hands. Sometimes also cartridge cases found on the crime scene are available. An attempt was done by Cheylan et al. to explore the forensic potential of linking suspects to a crime through investigations of the chemical contents of GSR samples collected from the cartridge case, the barrel, the target and the breech [17]. The samples were compared taking into account the proportion of lead, antimony and barium measured by Inductively Coupled Plasma-Atomic Emission Spectrometry (ICP-AES) as well as ²⁰⁸Pb/²⁰⁶Pb and ²⁰⁸Pb/²⁰⁷Pb isotope ratios measured by Sector Field-Inductively Coupled Plasma Mass Spectrometry (SF-ICP-MS). The authors revealed variation of GSR composition along the gun to target path, e.g. approximately the same proportion of lead originating from the primer (about 80%) was found in samples taken from the barrel and the breech. About 93% of lead recovered in the cartridge case came from the primer and the remaining 7% from the core of projectile, whereas as much as 44% of the lead found on the target came from the core of the projectile.

A comparative study of the chemical content and morphology of GSR collected from the shooter's hands and from inside of a spent cartridge obtained as a result of a single shot with full metal jacket (FMJ) Luger 9mm ammunition of various producers was performed using SEM-EDX method [18]. Examples of ammunition representing the following three types of primer were examined: the oldest one (but still in use) based on mercury fulminate, the traditional one based on lead styphnate or lead azide and one of the newest lead-free types. The study has shown some similarities as well as great differences between residue collected from the two substrates. Whereas the residue present inside the cartridge case revealed the chemical composition directly resulting from the composition of the primer, to the chemical composition of the airborne particles contributed materials of the core and the jacket of the projectile. The greatest modification of the chemical composition of the airborne GSR, in comparison to that of particles found inside the cartridge case, took place in the case of the lead-free ammunition with a tin plated projectile as well as in the case of the mercury fulminate primed ammunition. Tin originating from the surface of the projectile in the lead-free ammunition and similarly, lead originating from the uncovered part of the projectile of the mercury fulminate primed ammunition were present in GSR collected from the shooter's hands in the appropriate experiments. Thus, it has been concluded that the distribution of the chemical elements in the gunshot residue is determined by the direction of the movement of the products of detonating primer and burning propellant as well as the kind of materials that were used to construct the ammunition and gun.

The presented study was aimed at establishing whether application of SEM-EDX method would provide an information on possible changes of sizes of GSR as well as the proportions of their chemical classes depending on the distance from the muzzle, both in the direction of shooting and the opposite one. An example of mercury fulminate primed ammunition was chosen for the examinations for two reasons. Firstly, this type of ammunition is being produced and widely used in Poland and so becomes a frequent subject of forensic expertise. Secondly, in this ammunition lead is not present in the primer mixture, however the metal jacket does not cover the bottom of lead bullet core. Thus, it was expected that lead containing particles to be possibly detected in the surroundings of a shooter could be related to the origin of the bullet core and so to provide an information on the mechanism of dispersion of GSR using this type of ammunition.

Back to Article Outline

2. Experimental 

2.1. Materials 

The subjects of the study were inorganic gunshot residues collected from hands of the shooter, from the sleeves, the front and back upper parts of his upper clothing and from a target in the distance of 50cm, obtained within the same single shot. Moreover, microtraces were collected from targets shot in the following distances: 10, 20, 30, 70 and 100cm, in a separate experiment.

In order to obtain the investigative material in the form of metallic particles originating from firing a gun two types of experiments were performed. The first experiment was performed in such a manner that a person, who did not shoot for more than 3 days before, dressed in a white cotton lab coat, made a single shot. After a shot had been performed, the traces were collected from the shooter's hands immediately at the shooting gallery and the clothing and targets were secured for further investigation in the laboratory. The second experiment consisted in test shots to targets placed at distances of 10, 20, 30, 70 and 100cm. In the experiments each target was covered with a fresh paper target and a 40cm×40cm fragment of a white cotton fabric. There were performed three rounds of each experiment.

The shooting experiments were performed in a police shooting gallery with a Walther P-99 pistol (F.B. “Łucznik”, Radom, Poland). Prior to the experiment the gun was cleaned with a cleaning-brush and cloth wetted in petroleum spirits away from the shooting gallery.

The appropriate ammunition Luger 9mm FMJ (full metal jacket) of production lots 2003 and 2004 (Mesko Metal Works, Skarżysko-Kamienna, Poland) was applied.

All samples were collected using aluminium stubs with conductive carbon adhesive tabs (TAAB Laboratories Equipment Ltd., Great Britain) making 100–200 pressings of the stubs to the substrate in the case of hands (including the thumb and the index finger) and clothing, and 12 pressings to the targets around the gunshot hole in the radius of 10cm.

Stubs with the collected firearm discharge residues were covered with the conductive layer of carbon using a SCD 050 sputter, BAL-TECH, Lichtenstein in order to avoid charging of organic debris such as hair, fragments of epithelium, cotton fibres, particles of incompletely burned propellant, etc.

2.2. Methods 

The prepared samples were placed inside the sample chamber of a scanning electron microscope JSM-5800, Jeol with an energy dispersive X-ray spectrometer Link ISIS 300, Oxford Instruments Ltd. (Si(Li) detector, ATW—atmospheric thin window, resolution 131eV for Mn Kα at 10000 counts).

The automatic identification of GSR was performed with a GunShot program, i.e. an application of the Link ISIS 300, Oxford Instruments. The program automatically searches for particles of defined features subsequently analysing rectangular frames, into which the whole area of a stub is being divided; the number and the size of frames depends on the magnification set. Firstly, the program requires from an operator setting up the layout of the stubs as well as the Mn–Pd standard (for establishing the range of the back-scattered electron signal registered with the detector of the SEM). In the next steps the set of the expected chemical classes of the particles, the limits of particle size and their number within a frame have to be established. The chemical classes of particles are defined by setting the list of contributing elements within broad ranges of their composition, typically between 0 and 100%, including the border values. Since many particles were expected in the samples collected immediately after shooting, the measurement setting aimed at particles of 1μm in diameter not to be missed (Table 1). The detection calibration of the program was performed using a synthetic particle specimen SPS 521C, Plano W. Plannet GmbH, Wetzlar. The standard sample consists of 43 randomly distributed PbSb particles on silicon wafer of known positions and the following sizes: 1.2μm, 2.5μm and 6μm. The measurement settings were adjusted so that all particles in the standard sample were detected. The entire area of each stub was analysed. The agreement of the chemical composition of a particle (resulting from its X-ray spectrum) with the class assigned to it by the program was checked and corrected manually.

Table 1. Analytical conditions of the automatic search for GSR by means of SEM-EDX method.

Back to Article Outline

3. Results and discussion 

As a result of the automatic search of the whole area of the samples taken from the shooter's hands, clothing and the targets varying numbers of metallic particles, from several dozens to several thousands, were obtained (Table 2). In order to make a comparison between samples, taking into account either the size or the chemical content of particles, the numbers of particles of certain classes were transformed into the frequencies of occurrence expressed as fraction against the total number of the metallic particles found in a sample:

where N_i, number of particles of ith class.

Table 2. Numbers of the metallic particles revealed by means of SEM-EDX method in the studied samples.

3.1. Dependence of the chemical composition of GSR on the distance from the muzzle 

The frequencies of occurrence of the following chemical classes of the revealed particles were calculated for each of the studied samples: PbSbBa, PbBa, PbSb, SbBa, Pb, Sb(Sn) and Ba.

Results obtained for samples taken from the shooting person, i.e. his hands and clothing: the lower parts of the sleeves, the front upper parts and the back upper parts are presented in Fig. 1a–d. It has been estimated that for the shooting person (an adult male 185cm tall) the places of GSR collection were in the following distances (s) from the pistol muzzle: hands in about 10cm, sleeves −30cm, the front upper part of clothing −70cm and the back upper part of clothing −90cm. The assessed values of the distance were given the negative sign as they were opposite to the direction of shooting. Results obtained for samples collected from targets placed in the following distances from the muzzle: 10, 20, 30, 50, 70, and 100cm are presented in Fig. 2a–f.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Frequencies of occurrence f of chemical classes of particles in samples taken from the shooting person: from the back (a) and front (b) upper parts of the upper clothing, the sleeves (c) and hands (d).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2. 

  Frequencies of occurrence f of chemical classes of particles in samples taken from targets in the following shooting distances, s: 10cm (a), 20cm (b), 30cm (c), 50cm (d), 70cm (e) and 100cm (f).

One can observe a similar distribution of the chemical classes of particles collected from the shooter's clothing (Fig. 1a–c), i.e. the prevailing, nearly equal amount of Sb- and Pb-particles, small amounts of PbSb-particles and some minor quantity of particles of other classes. However, distribution of the chemical classes of particles collected from the shooter's hands, differs significantly: the Sb class outnumbers the other six classes of particles (Fig. 1d).

In the case of the samples taken from targets (Fig. 2a–f) one can observe the presence on particles only of classes PbSb, Pb and Sb. In the sample collected from the target placed at 10cm from the muzzle (Fig. 2a) frequencies of occurrence of these classes of particles are about 0.3. The proportions are gradually changing with the shooting distance, so that at distances 70 and 100cm fractions of Pb- and PbSb-particles decrease to less than 0.1, whereas the fraction of Sb-particles increases to about 0.9.

The primer of Luger 9mm ammunition by Mesko consists in mercury fulminate playing the role of the detonator, antimony trisulfide—the fuel and potassium chlorate being the oxidiser. However, no more than 1% of particles in each sample contained minor quantities of mercury and particles containing solely mercury or its significant amounts were not observed. This observation remains in agreement with results of earlier studies on the chemical contents of particles taken from the cartridge case and hands of a person shooting with a Mesko Luger 9mm ammunition [18] as well as other types of mercury fulminate primed ammunition [19], [20], [21], [22]. The primer cup is sealed with tin and so, in the majority of particles containing antimony also tin was observed. Significant amounts of lead containing airborne particles are attributed to an interaction of the reacting primer and propellant materials with the bottom of projectile, where lead core is uncovered by the copper jacket as no lead compounds were present in the primer mixture. It is worthwhile mentioning that, for the same production lots of Mesko Luger 9mm ammunition no lead was found among the residue from inside the cartridge cases, instead it contained sulphur, chlorine, tin, antimony, potassium and mercury [18]. This finding suggests that neither the primer nor the propellant are the sources of minor quantities of barium containing particles observed in samples taken from the shooters hands and clothing as well as the short distant targets. They may originate from the barrel as the contamination from previous shootings since there are no perfect methods of cleaning the weapon from the GSR.

3.2. Dependence of the size of GSR on the distance from the muzzle 

Particles detected in each sample were sorted according to the ranges of their equivalent circle diameter d, expressed in micrometers, established by the program GunShot:

There were chosen the following ranges of particle diameters: (0–1.00), (1.01–1.50), (1.51–2.00), … (6.01–7.00), (7.01–10.00), (10.01–20.00), (20.01–30.00)μm.

Frequencies of occurrence of particles of certain size range obtained for samples taken from the shooting person, i.e. his hands and clothing: the lower parts of the sleeves, the front upper parts and the back upper parts are presented in Fig. 3a–d. Results obtained for samples collected from targets placed in the following distances from the muzzle: 10, 20, 30, 50, 70, and 100cm are presented in Fig. 4a–f.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 3. 

  Frequencies of occurrence f of the equivalent circle diameter d of particles in samples taken from the shooting person: from the back (a) and front (b) upper parts of the upper clothing, the sleeves (c) and hands (d).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 4. 

  Frequencies of occurrence f of sizes of particles in samples taken from targets in the following shooting distances, s: 10cm (a), 20cm (b), 30cm (c), 50cm (d), 70cm (e) and 100cm (f).

For all of the examined samples, both taken from the shooting person and from the targets, the majority of particles are of the smallest size in the ranges of sub-microns to about 1.5μm.

The distributions of the size of the particles become more flat and the relative contribution of larger particles grows with the distance from the muzzle, both, in the direction of shooting and in the opposite direction. However, the tendency is more pronounced for the direction of shooting as the frequency of occurrence of the smallest particles detected diminished from about 0.56 for target s=10cm (Fig. 4a) down to about 0.23 for target s=100cm (Fig. 4f). One can also observe that there were not present particles bigger than 4.0–4.5μm in the sample collected from target in the distance s=10cm. Particles bigger that these occurred at longer distances and their sizes as well as the amounts gradually increased with the distance from the muzzle. In the direction opposite to shooting the frequencies of occurrence of the smallest particles diminished within smaller interval from about 0.35 for samples taken from the shooter's hands, i.e. s=−10cm (Fig. 3d) to about 0.25, on average, for samples taken from back parts of his clothing, i.e. s=−90cm (Fig. 3a). Within all samples collected from the shooting person there were present particles of sizes representing all the considered ranges.

In order to obtain a general view on changes of the size of particles depending on the distance from the muzzle mean values of the equivalent circle diameter d were calculated for all the collected samples according to the following formula:

where d_j, diameter of the jth particle [μm]; N_j, number of particles in a sample.

It can be observed in Fig. 5 that the mean values of the equivalent circle diameter of particles increase with the distance from the muzzle in both, the direction of shooting and the opposite direction. Moreover, the mean values of the equivalent circle diameter of particles tend to be bigger in samples taken from the shooting person than in these taken from targets. That shows that there is a relatively greater contribution of larger particles in the population of particles collected from the shooting person, i.e. in the direction opposite to shooting. The smallest mean value of diameters was observed for the target placed in the distance of 10cm from the muzzle, i.e. 1.08μm that remains with an agreement with data in Fig. 4a. A relatively large value of the mean diameter of particles collected from the shooters hands (s=−10cm), i.e. 1.80μm on average was observed. It ought to be expected that a significant amount of GSR escaping from the gun via the ejection port contribute to the population of the GSR depositing on hands. Thus, one can conclude that relatively large particles are being formed in the region of back parts of the gun, similarly to these observed inside the cartridge case. It might be so due to more likely condensation processes in the limited space prior to cooling in the air. The greatest value of the mean diameter of particles, i.e. 2.82μm was observed for the back parts of the shooters clothing (s=−90cm), however with a great standard deviation resulting from rather poor statistics of the low numbers of particles found in this place. Among particles of the same density the greatest kinetic energy would have these of bigger sizes. That could also partially explain the presence of rather large particles, if any, in further distances from the shooting gun in the direction opposite to shooting.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 5. 

  Mean values of the equivalent circle diameter d of particles in each sample in a function of the distance s from the gun muzzle.

3.3. Combined inspection into the dependence of size and chemical contents of GSR on the distance from the muzzle 

The frequency of occurrence and the mean values of diameters of particles belonging to certain classes were calculated in a function of the shooting distance and presented in one chart. As examples the data for Sb and Pb particles are presented in Fig. 6a and b. As it can be seen from these figures, both the frequency of occurrence as well as the dimensions of particles behave differently in the function of distance from the muzzle, depending on their chemical contents, and so the density.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 6. 

  Frequencies of occurrence f and mean values of the equivalent circle diameter of particles d_i [μm] calculated for particles of the chemical classes Sb(Sn) (a) and Pb (b) in a function of the distance s from the gun muzzle.

Thus, an attempt was done to perform a combined inspection into, both the size and the chemical content, in a function of the shooting distance.

Schwoeble and Exline [23] showed that movement of solid particles in the air could be described in terms of Stockes law, assuming static conditions and neglecting turbulence. In this approach the retardation of a particle moving in the air is being expressed as a difference between the gravitational force and the buoyancy force, and so, depending on the properties of the air and the parameters of a particle, such as its size and density. As the properties of air at certain conditions are constant, the retardation, and so the distance being travelled by a particle, depends mainly on its density and its volume. Assuming the spherical shape of GSR particles one can find that the retardation depends reciprocally on the density and the diameter of a particle.

where d, diameter of a particle; C_D, air buoyancy coefficient; ρ_L, air density; v, particle velocity; ρ_p, particle density.

The density of material is related to its chemical content and the state. Knowing the chemical classes of the detected particles one can assess their densities in solid state taking into account data available in the physical tables [24]. In Table 3 there are listed densities of the elements and their combinations present in the particles being calculated as mean values of the densities of the present elements assuming their equal contribution. The diameters of the particles as well as the chemical classes assigned to them are present in the output of a GSR searching program.

Table 3. Estimated densities of particles of the chemical classes found in the samples.

Thus, a parameter, called here ξ, being a reciprocal value of multiplication of the particle diameter and its density, can be calculated from the GSR measurement data:

where ρ_i, mean density of a particle in the ith chemical class [g/cm³], d_i, mean value of diameters of particles in the ith chemical class [μm].

The parameter ξ_i was calculated for each of the seven chemical classes of the detected particles for each sample, i.e. in a function of the distance from the shooting gun. Results for the highly populated classes of Pb, Sb and PbSb particles are presented in Fig. 7 and for the remaining poorly represented classes of PbSbBa, PbBa, SbBa, and Ba in Fig. 8.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 7. 

  Values of the parameter ξ_i [cm³/(gμm)] calculated for highly populated classes of particles: Pb, Sb and PbSb in a function of the distance s from the gun muzzle.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 8. 

  Values of the parameter ξ_i [cm³/(gμm)] calculated for weakly represented classes of particles: PbSbBa, PbBa, SbBa, and Ba in a function of the distance s from the gun muzzle.

One can see in Fig. 7 that values of the parameter ξ_i calculated for Pb and PbSb classes are higher in the direction of shooting than in the opposite one, and remain similar in the case of Sb class in both directions. Thus, Pb and PbSb particles experience higher retardation in the direction of shooting than in the opposite direction, whereas Sb particles undergo similar retardation in both directions. It can be related to the sizes of particles: Pb and PbSb particles are generally smaller in direction of shooting than in the opposite direction and the distribution of sizes of Sb particles is similar in both directions (Fig. 6a and b). Generally, values of the parameter ξ_i of the three classes show a decreasing tendency with the distance from the muzzle in both directions, pointing on an increase of the sizes of particles with the distance.

Values of the parameter ξ_i calculated for PbSbBa, PbBa, SbBa, and Ba particles that occur in small amounts (Fig. 8) seem to be less regular than these of Pb, Sb and PbSb particles, however, some general observations can be made. ξ_i of PbSbBa particles remain similar in both directions, generally decreasing with the distance from the muzzle. The lightest Ba particles were observed only on the shooting person and they showed the highest values of parameter ξ_i for the smallest and the biggest distance from the muzzle, i.e. for s=−10cm and s=−90cm, respectively. PbBa particles were found only in samples taken from the shooter's hands and sleeves as well as from targets s=10cm and s=20cm and they showed higher values of ξ_i for targets than for the shooting person, significantly decreasing with the distance from the muzzle. SbBa particles reveal the parameter ξ_i similar to that of Ba particles on the shooting person and only it has two values in the direction of shooting, i.e. at s=20cm and s=100cm.

Moreover, the weighted mean value of ξ was calculated for the entire population of particles in a sample according to the following formula:

where ξ_i, parameter ξ of the ith chemical class of particles in a sample, f_i, frequency of occurrence of particles of ith chemical class.

In the case of the repeated experiments an average value and the standard deviation were calculated and presented in Fig. 9.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 9. 

  The weighted mean value of 〈ξ〉 [cm³/(gμm)] calculated for the entire population of particles in a sample in a function of the distance s from the gun muzzle.

With so calculated parameter an overall picture of dispersion of GSR particles was achieved. It shows that particles travelling along the shooting direction experience a slightly higher retardation than these moving in the opposite direction. Parameter 〈ξ〉 and so the retardation of particles decreases with the distance in both directions. This tendency reflects an increase of the average diameter of particles 〈d〉 (Fig. 5) as well as the change of the mutual proportion of the particles of the chemical classes, especially the rate of Sb and Pb particles (Fig. 1, Fig. 2, Fig. 6).

3.4. Assessment of the repeatability of parameters describing metallic particles collected in the vicinity of the shooting gun 

The objective of the performed examinations was to obtain information on properties of GSR emerging in the nearest vicinity of the shooting gun, i.e. on the close-distant targets and on the shooting person. The following parameters describing particles were taken into account: the number of the revealed particles, the frequencies of occurrence of their chemical classes and their sizes expressed as the equivalent circle diameter. Moreover, a parameter ξ binding the size and the density of a particle was calculated. All these parameters were observed in a function of the distance from the muzzle of the shooting gun.

In order to evaluate the repeatability of the results obtained from three experiments performed, while a person was shooting to a target at 50cm distance, mean values and the relative standard deviations (RSD) of the above-mentioned parameters were calculated. Examples of the calculations for chosen parameters are listed in Table 4.

Table 4. Mean values and the relative standard deviations RSD for the number of particles, frequency of occurrence f of Sb(Sn) particles, the mean diameter 〈d〉 and mean parameter 〈ξ〉.

The number of particles in a sample is an unrepeatable parameter, since even for highly populated samples such as these collected from hands and the target placed in s=10cm distance the RDS was only marginally lower than 50% and for less populated samples it exceeded 100%.

Much better repeatability was achieved for frequencies of occurrence of particles of certain class, e.g. the chemical class Sb(Sn) as well as for the mean values of particle diameters within a sample 〈d〉. RSD in the case of both parameters was typically about 30% and less.

The best repeatability was presented in the case of the weighted mean parameter 〈ξ〉 as the highest RDS was 24%.

Thus, one can conclude that except for the numbers of particles the other parameters describing GSR particles are worthwhile analysing as they provide a meaningful description of the dispersion of particles in the vicinity of the shooting gun.

3.5. On the mechanism of dispersion of GSR 

Results obtained within the presented work can be utilised to consider the mechanism of dispersion of GSR originating from the investigated type of ammunition, i.e. Luger 9mm, FMJ, Mesko.

The phenomena related to a cartridge discharge are very dynamic and reveal a complex nature. For this reason there are not many studies related to mechanism of gunshot residue formulation as well as their dispersion in the vicinity of the shooting gun.

Studying the morphology of GSR collected from the shooter's hands and distributions of lead, antimony and barium in the cross-sections of particles by means of SEM-EDX, taking into account the temperatures of melting and evaporation of these metals, Basu made inferences on a three-stage combustion process of the primer and the propellant [25].

Wolten et al. have performed examinations of GSR collected from shooter's hands that were obtained using ammunition, especially prepared by addition of various metals or metal compounds to the propellant and the projectile. They found that the materials of the projectile (lead and copper) clearly contribute to the metallic GSR. However, the quantity of these additions depends on properties such as the melting and evaporation temperatures as well as the chemical affinity of metals originating from the primer to lead and copper of the projectile core and jacket, and so the possibility of constituting the appropriate alloy systems [26].

The only theoretical modelling of the phenomenon of GSR deposition on targets in the close-range shooting distances was attempted by Bhattacharyya, who adopted a kinetic theory of gases [27]. The entire organic, inorganic and metallic discharge products can be considered as a gaseous system moving with a high velocity in the direction of shooting, in which the particular velocities of the constituent particles reveal a Maxwellian distribution. The following assumptions were made: no relative motion between the muzzle and the target occurred and also the air friction and the gravitational field did not affect the system of particles. Although very interesting this model did not strictly reproduce analytical data on the concentration of antimony on the subsequent targets, neither did the model later modified for collisions of the metallic and the air particles [28]. The discrepancies between the measured and the calculated concentrations of Sb increased with the shooting distance. The reasons for that, as suggested by the author, could have been variation of shapes, sizes and mass of particles that were not taken into account in the model as well as a possible range of particle velocity that effects in adhesion of a particle on the target. Particles both of lower and of higher velocity than the range would not adhere to the target or reflect from it.

Results obtained within current publication provided information on the proportion of particles of particular sizes and chemical contents that influences the density and so the mass of particles. The following assumptions were made in order to simplify the calculations: all particles were of the spherical shape (that remains in agreement with the observation that as much as 70% of metallic particles are of the spherical shape, see, e.g. [3]), they revealed a homogenous chemical content and the density of particles composed of more than one element was an average value of the densities of the constituting elements.

The type of ammunition selected for the performed experiments represented the one of the old-fashioned recipe of primer, based on mercury fulminate, antimony sulphide and potassium chlorate making, however, possible a differentiation between the primer residue from those originating from the projectile, composed of lead core and copper jacket (FMJ) that did not cover the bottom, hollow part of the projectile core. Also, the unwanted debris in the form of particles containing barium, clearly not originating from the cartridge but from the barrel, could have been utilised in drawing conclusions on the mechanism of GSR dispersion.

The majority of the primer Sb(Sn) particles travel with the main and the fastest stream of the combustion products of the primer and the propellant loads, along the cartridge case, the barrel and yet some distance in the free air, ending on the target.

The head of the shock wave of the reacting explosives, while striking the bottom of the projectile becomes to interact with the uncovered lead core. Since it is hollow shaped, it is very likely that droplets of partially molten lead extracted from the surface of the projectile core are pushed away from the centre of the main stream of the gases towards the inner surfaces of the weapon barrel. Thus, after the projectile leaves the barrel, the particles travelling in the outer, slower layer, including the Pb particles together with those polished off the surface of the barrel (Ba containing particles) are prone to deflection from the direction parallel to the axis of the barrel.

Only the smallest Pb and PbSb particles are drawn with the main stream of the gases in the direction of shooting. Due to their small sizes they experience a great retardation causing their diminishing contribution to the population of particles revealed in the subsequent targets with the increasing shooting distance.

The fact of finding only very small particles on the close range targets, e.g. s=10cm seems to support the idea suggested by Bhattacharyya of the reflection of the bigger and so having the highest kinetic energy particles from the target [28].

As soon as the projectile leaves the barrel, the products of the detonating primer and the burning propellant undergo an immediate expansion in all directions as well as the air resistance. This causes local turbulence of gases and vapours that may be observed in films or pictures recorded by means of a high-speed camera [23]. Solid GSR particles in these conditions undergo a rapid deceleration [29], a deflection from the initial direction of movement parallel to the barrel axis making some of them to travel in the direction opposite to shooting, i.e. behind the gun muzzle. It concerns mainly particles travelling in the outer layer of the combustion stream including the ones detached from the bottom part of the projectile (Pb containing particles) and these polished off the inner surfaces of the barrel (Ba containing particles). These kinds of particles travelling in the direction opposite to shooting deposit on the shooting person. However, a special attention to the particles depositing on the shooter's hands ought to be paid, since they significantly differ from these revealed on the shooter's clothing as well as on the targets.

Due to the construction of a semi-automatic pistol utilising the pressure of the combustion products for ejection the used cartridge case and loading a new one, it ought to be expected that the majority of particles depositing on the shooter's hands originate from the ejector and so from the closest vicinity of the cartridge case. Indeed, the particles found on the shooter's hands are similar to the ones taken from the cartridge cases of the same batch of ammunition in the prevailing presence of Sb(Sn) particles [18]. A lack of particles containing lead and barium in the cartridge case and their presence in the sample taken from the shooting hands proves partial origination of particles from the barrel leaving it via the muzzle and travelling in the direction opposite to the shooting one.

Although limited to only one brand of ammunition and one item of weapon, the performed examinations allowed to find that the formation and dispersion of metallic gunshot residue are not accidental and may be utilised to understand similarities and differences of the quantity and quality of the particles originating from physically the same cartridge but collected from various substrates in the place of shooting incident. Moreover, the presented here way of inspection of the routinely obtained information on GSR features may be utilised in casework in some circumstances.

Back to Article Outline

4. Conclusions 

The performed study has shown that SEM-EDX method being routinely used for the detection of characteristic gunshot residues can also be applied for the collection of information on possible changes of sizes of the particles as well as the proportions of their chemical classes depending on the distance from the muzzle, both in the direction of shooting and the opposite one. The observed features of metallic gunshot residue depend on the distance from the muzzle and the kind of substrate they were collected from as well as on the chemical composition and the details of construction of the ammunition.

It has been found that the mechanisms of formation and dispersion of GSR, although dynamic and complex, are immanent features of a discharge of a certain type of a cartridge.

The subject of the study were GSR collected from hands and clothing of the shooting person and from vertical targets covered by cotton fabric in selected distances, the substrates being similar to these normally examined as the evidence in real shooting cases. Thus, in some privileged circumstances of having a gun and ammunition in access the presented way of analysing the features of the examined GSR may contribute to crime reconstruction.

Back to Article Outline

Acknowledgements 

The author is grateful to Mr. Krzysztof Zdeb, Criminalistic Laboratory of the Voivodeship Headquarters of Police in Krakow for help in performing experiments in the shooting gallery.

Back to Article Outline

References 

1. Niewoehner L, Andrasko J, Biegstraaten J, Gunaratnam L, Steffen S, Uhlig S. Maintenance of the ENFSI Proficiency Test Program on Identification of GSR by SEM/EDX (GSR2003). J. Forensic Sci. 2005;50:877–882
   • View In Article
   • MEDLINE
2. Niewoehner L, Andrasko J, Biegstraaten J, Gunaratnam L, Steffen S, Uhlig S, et al. GSR2005–Continuity of the ENFSI Proficiency Test on Identification of GSR by SEM/EDX. J. Forensic Sci. 2008;53:162–167
   • View In Article
   • CrossRef
3. Wolten GM, Nesbitt RS, Calloway AR, Loper GL. Particle analysis for the detection of gunshot residue. I: Scanning electron microscopy/energy dispersive X-ray characterisation of hand deposits from firing. J. Forensic Sci. 1979;24:409–422
   • View In Article
4. Wolten GM, Nesbitt RS, Calloway AR, Loper GL. Particle analysis for the detection of gunshot residue. II: Occupation and environmental particles. J. Forensic Sci. 1979;24:423–430
   • View In Article
5. Wolten GM, Nesbitt RS, Calloway AR, Loper GL. Particle analysis for the detection of gunshot residue. III: The case record. J. Forensic Sci. 1979;24:864–869
   • View In Article
6. Wallace JS, McQuillan J. Discharge residues from cartridge-operated industrial tools. J. Forensic Sci. Soc. 1979;24:495–508
   • View In Article
   • Abstract
   • Full-Text PDF (1651 KB)
   • CrossRef
7. Garofano L, Capra M, Ferrari F, Bizzaro GP, DiTullio D, Dell’Olio M, et al. Gunshot residue further studies on particles of environmental and occupation origin. Forensic Sci. Int. 1999;103:1–21
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (1505 KB)
   • CrossRef
8. ENFSI–guide for gunshot residue analysis by scanning electron microscopy/energy-dispersive X-ray spectrometry, ENFSI Working Group Firearms, September 2005. (Approved at the 12th Meeting of the ENFSI Working Group Firearms, Oslo, Norway.)
   • View In Article
9. Romolo FS, Margot P. Identification of gunshot residue: a critical review. Forensic Sci. Int. 2001;119:195–211
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (245 KB)
   • CrossRef
10. Avermate S. The application of Bayesian principles to GSR case reports–some typical examples. In: The 12th Meeting of the ENFSI EWG Firearms. Oslo. 2005;
    • View In Article
11. Biederman A, Taroni F. A probabilistic approach to the joint evaluation of firearm evidence and gunshot residues. Forensic Sci. Int. 2006;163:18–33
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (343 KB)
    • CrossRef
12. Berendes A, Neimke D, Schumacher R, Barth M. A versatile technique for the investigation of gunshot residue patern on fabrics and other surfaces: m-XRF. J. Forensic Sci. 2006;51:1085–1090
    • View In Article
    • MEDLINE
    • CrossRef
13. Atwater CS, Durina ME, Durina JP, Blackledge RD. Visualisation of gunshot residue patterns on dark clothing. J. Forensic Sci. 2006;51:1091–1095
    • View In Article
    • MEDLINE
    • CrossRef
14. Davis GH, Poole LL, Springer F. GSR mapping: the study of gunshot residue distribution of known firearms in closed environment. In: California Association of Criminalists Semiannual Meeting. Lake Tahoe, May 10–12. 2001;
    • View In Article
15. Fojtašek L, Kmječ T. Time periods of GSR particles deposition after discharge—final results. Forensic Sci. Int. 2005;153:132–135
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (253 KB)
    • CrossRef
16. Fojtašek L, Vacinova J, Kolar P, Kotrly M. Distribution of GSR particles in the surroundings of shooting pistol. Forensic Sci. Int. 2003;132:99–105
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (220 KB)
    • CrossRef
17. Cheylan P, Dobney AM, Wiarda W, Beijer R, van Leuven E. Comparison of gunshot residue samples: variation composition along the gun to target path. In: Meeting ENFSI Firearms Working Group. Brugge. 2001;
    • View In Article
18. Brożek-Mucha Z. Comparison of cartridge case and airborne GSR—a study of their elemental contents and morphology by means of SEM-EDX. X-Ray Spectrom. 2007;36:398–407
    • View In Article
19. Brożek-Mucha Z. SEM-EDX study of inorganic gunshot residues from Makarov 9mm ammunition. Prob. Forensic Sci. XLI. 2000;62–89
    • View In Article
20. Brożek-Mucha Z. A gunshot following the stopping of passenger car by police—reconstruction of an event on the basis of case files and gunshot residue examinations. Prob. Forensic Sci. LI. 2002;119–136
    • View In Article
21. Brożek-Mucha Z, Jarosz J. Reconstruction of a crime with the use of a firearm from the case files study and GSR examinations. Prob. Forensic Sci. XLV. 2001;109–121
    • View In Article
22. Wallace JS. Discharge residue from mercury fulminate-primed ammunition. Sci. Justice. 1998;38:7–14
    • View In Article
    • Full-Text PDF (3914 KB)
    • CrossRef
23. Schwoeble AJ, Exline DL. Current Methods in Forensic Gunshot Residue Analysis. Boca Raton, London, New York, Washington, DC: CRC Press; 2000;
    • View In Article
24. Praca zbiorowa, Poradnik fizykochemiczny, wyd. II, WNT, Warszawa, 1974.
    • View In Article
25. Basu S. Formation of gunshot residues. J. Forensic Sci. 1982;27:72–91
    • View In Article
26. Wolten GM, Nesbitt RS. On the mechanism of gunshot residue particle formation. J. Forensic Sci. 1980;25:533–545
    • View In Article
27. Bhattacharyya CN. Dispersion of firing discharge residues using a Maxwellian model. Forensic Sci. Int. 1989;42:271–277
    • View In Article
    • CrossRef
28. Bhattacharyya CN. Dispersion of firing discharge residues using a modified Maxwellian model. Forensic Sci. Int. 1990;47:31–37
    • View In Article
    • CrossRef
29. DeForest PR, Martir K, Pizolla PA. Gunshot residue particle velocity and deceleration. J. Forensic Sci. 2004;49:1237–1243
    • View In Article
    • MEDLINE