news

We use cookies to improve your experience. By continuing to browse this site, you agree to our use of cookies. More information.
When a traffic accident is reported and one of the vehicles leaves the scene, forensic laboratories are often tasked with recovering the evidence.
Residual evidence includes broken glass, broken headlights, taillights, or bumpers, as well as skid marks and paint residue. When a vehicle collides with an object or person, the paint is likely to transfer in the form of spots or chips.
Automotive paint is usually a complex mixture of different ingredients applied in multiple layers. While this complexity complicates analysis, it also provides a wealth of potentially important information for vehicle identification.
Raman microscopy and Fourier transform infrared (FTIR) are some of the main techniques that can be used to solve such problems and facilitate non-destructive analysis of specific layers in the overall coating structure.
Paint chip analysis begins with spectral data that can be directly compared to control samples or used in conjunction with a database to determine the make, model, and year of the vehicle.
The Royal Canadian Mounted Police (RCMP) maintains one such database, the Paint Data Query (PDQ) database. Participating forensic laboratories can be accessed at any time to help maintain and expand the database.
This article focuses on the first step in the analysis process: collecting spectral data from paint chips using FTIR and Raman microscopy.
FTIR data were collected using a Thermo Scientific™ Nicolet™ RaptIR™ FTIR microscope; complete Raman data were collected using a Thermo Scientific™ DXR3xi Raman microscope. Paint chips were taken from damaged parts of the car: one chipped from the door panel, the other from the bumper.
The standard method of attaching cross-sectional specimens is to cast them with epoxy, but if the resin penetrates the specimen, the results of the analysis may be affected. To prevent this, the paint pieces were placed between two sheets of poly(tetrafluoroethylene) (PTFE) at a cross section.
Prior to analysis, the cross section of the paint chip was manually separated from the PTFE and the chip was placed on a barium fluoride (BaF2) window. FTIR mapping was performed in transmission mode using a 10 x 10 µm2 aperture, an optimized 15x objective and condenser, and a 5 µm pitch.
The same samples were used for Raman analysis for consistency, although a thin BaF2 window cross section is not required. It is worth noting that BaF2 has a Raman peak at 242 cm-1, which can be seen as a weak peak in some spectra. The signal should not be associated with paint flakes.
Acquire Raman images using image pixel sizes of 2 µm and 3 µm. Spectral analysis was performed on the principal component peaks and the identification process was aided by the use of techniques such as multi-component searches compared to commercially available libraries.
Rice. 1. Diagram of a typical four-layer automotive paint sample (left). Cross-sectional video mosaic of paint chips taken from a car door (right). Image Credit: Thermo Fisher Scientific – Materials and Structural Analysis
Although the number of layers of paint flakes in a sample may vary, samples typically consist of approximately four layers (Figure 1). The layer applied directly to the metal substrate is a layer of electrophoretic primer (approximately 17-25 µm thick) that serves to protect the metal from the environment and serves as a mounting surface for subsequent layers of paint.
The next layer is an additional primer, putty (approx. 30-35 microns thick) to provide a smooth surface for the next series of paint layers. Then comes the base coat or base coat (about 10-20 µm thick) consisting of the base paint pigment. The last layer is a transparent protective layer (approximately 30-50 microns thick) which also provides a glossy finish.
One of the main problems with paint trace analysis is that not all layers of paint on the original vehicle are necessarily present as paint chips and blemishes. In addition, samples from different regions may have different compositions. For example, paint chips on a bumper may consist of bumper material and paint.
The visible cross-sectional image of a paint chip is shown in Figure 1. Four layers are visible in the visible image, which correlates with the four layers identified by infrared analysis.
After mapping the entire cross section, individual layers were identified using FTIR images of various peak areas. Representative spectra and associated FTIR images of the four layers are shown in Figs. 2. The first layer corresponded to a transparent acrylic coating consisting of polyurethane, melamine (peak at 815 cm-1) and styrene.
The second layer, the base (color) layer and the clear layer are chemically similar and consist of acrylic, melamine and styrene.
Although they are similar and no specific pigment peaks have been identified, the spectra still show differences, mainly in terms of peak intensity. Layer 1 spectrum shows stronger peaks at 1700 cm-1 (polyurethane), 1490 cm-1, 1095 cm-1 (CO) and 762 cm-1.
Peak intensities in the spectrum of layer 2 increase at 2959 cm-1 (methyl), 1303 cm-1, 1241 cm-1 (ether), 1077 cm-1 (ether) and 731 cm-1. The spectrum of the surface layer corresponded to the library spectrum of alkyd resin based on isophthalic acid.
The final coat of e-coat primer is epoxy and possibly polyurethane. Ultimately, the results were consistent with those commonly found in automotive paints.
Analysis of the various components in each layer was performed using commercially available FTIR libraries, not automotive paint databases, so while the matches are representative, they may not be absolute.
Using a database designed for this type of analysis will increase the visibility of even the make, model and year of the vehicle.
Figure 2. Representative FTIR spectra of four identified layers in a cross section of chipped car door paint. Infrared images are generated from peak regions associated with individual layers and superimposed on the video image. The red areas show the location of the individual layers. Using an aperture of 10 x 10 µm2 and a step size of 5 µm, the infrared image covers an area of ​​370 x 140 µm2. Image Credit: Thermo Fisher Scientific – Materials and Structural Analysis
On fig. 3 shows a video image of a cross section of bumper paint chips, at least three layers are clearly visible.
Infrared cross-sectional images confirm the presence of three distinct layers (Fig. 4). The outer layer is a clear coat, most likely polyurethane and acrylic, which was consistent when compared to clear coat spectra in commercial forensic libraries.
Although the spectrum of the base (color) coating is very similar to that of the clear coating, it is still distinct enough to be distinguished from the outer layer. There are significant differences in the relative intensity of the peaks.
The third layer can be the bumper material itself, consisting of polypropylene and talc. Talc can be used as a reinforcing filler for polypropylene to enhance the structural properties of the material.
Both outer coats were consistent with those used in automotive paint, but no specific pigment peaks were identified in the primer coat.
Rice. 3. Video mosaic of a cross section of paint chips taken from a car bumper. Image credit: Thermo Fisher Scientific – Materials and Structural Analysis
Rice. 4. Representative FTIR spectra of three identified layers in a cross section of paint chips on a bumper. Infrared images are generated from peak regions associated with individual layers and superimposed on the video image. The red areas show the location of the individual layers. Using an aperture of 10 x 10 µm2 and a step size of 5 µm, the infrared image covers an area of ​​535 x 360 µm2. Image Credit: Thermo Fisher Scientific – Materials and Structural Analysis
Raman imaging microscopy is used to analyze a series of cross sections to obtain additional information about the sample. However, the Raman analysis is complicated by the fluorescence emitted by the sample. Several different laser sources (455 nm, 532 nm and 785 nm) were tested to evaluate the balance between fluorescence intensity and Raman signal intensity.
For the analysis of paint chips on doors, the best results are obtained by a laser with a wavelength of 455 nm; although fluorescence is still present, a base correction can be used to counteract it. However, this approach was not successful on epoxy layers because the fluorescence was too limited and the material was susceptible to laser damage.
Although some lasers are better than others, no laser is suitable for epoxy analysis. Raman cross-sectional analysis of paint chips on a bumper using a 532 nm laser. The fluorescence contribution is still present, but removed by baseline correction.
Rice. 5. Representative Raman spectra of the first three layers of a car door chip sample (right). The fourth layer (epoxy) was lost during the manufacture of the sample. The spectra were baseline corrected to remove the effect of fluorescence and collected using a 455 nm laser. An area of ​​116 x 100 µm2 was displayed using a pixel size of 2 µm. Cross-sectional video mosaic (upper left). Multidimensional Raman Curve Resolution (MCR) cross-sectional image (lower left). Image Credit: Thermo Fisher Scientific – Materials and Structural Analysis
Raman analysis of a cross section of a piece of car door paint is shown in Figure 5; this sample does not show the epoxy layer because it was lost during preparation. However, since Raman analysis of the epoxy layer was found to be problematic, this was not considered a problem.
The presence of styrene dominates in the Raman spectrum of layer 1, while the carbonyl peak is much less intense than in the IR spectrum. Compared to FTIR, the Raman analysis shows significant differences in the spectra of the first and second layers.
The closest Raman match to the base coat is perylene; although not an exact match, perylene derivatives are known to be used in pigments in automotive paint, so it may represent a pigment in the color layer.
The surface spectra were consistent with isophthalic alkyd resins, however they also detected the presence of titanium dioxide (TiO2, rutile) in the samples, which was sometimes difficult to detect with FTIR, depending on the spectral cutoff.
Rice. 6. Representative Raman spectrum of a sample of paint chips on a bumper (right). The spectra were baseline corrected to remove the effect of fluorescence and collected using a 532 nm laser. An area of ​​195 x 420 µm2 was displayed using a pixel size of 3 µm. Cross-sectional video mosaic (upper left). Raman MCR image of a partial cross section (lower left). Image credit: Thermo Fisher Scientific – Materials and Structural Analysis
On fig. 6 shows the results of Raman scattering of a cross section of paint chips on a bumper. An additional layer (layer 3) has been discovered that was not previously detected by FTIR.
Closest to the outer layer is a copolymer of styrene, ethylene and butadiene, but there is also evidence of the presence of an additional unknown component, as evidenced by a small inexplicable carbonyl peak.
The spectrum of the base coat may reflect the composition of the pigment, since the spectrum corresponds to some extent to the phthalocyanine compound used as the pigment.
The previously unknown layer is very thin (5 µm) and partly composed of carbon and rutile. Due to the thickness of this layer and the fact that TiO2 and carbon are difficult to detect with FTIR, it is not surprising that they were not detected by IR analysis.
According to the FT-IR results, the fourth layer (the bumper material) was identified as polypropylene, but the Raman analysis also showed the presence of some carbon. Although the presence of talc observed in FITR cannot be ruled out, an accurate identification cannot be made because the corresponding Raman peak is too small.
Automotive paints are complex mixtures of ingredients, and while this can provide a lot of identifying information, it also makes analysis a major challenge. Paint chip marks can be effectively detected using the Nicolet RaptIR FTIR microscope.
FTIR is a non-destructive analysis technique that provides useful information about the various layers and components of automotive paint.
This article discusses the spectroscopic analysis of paint layers, but a more thorough analysis of the results, either through direct comparison with suspect vehicles or through dedicated spectral databases, can provide more precise information to match the evidence to its source.