Sign in

‘How Did THAT Get in There?’ Identifying Particulate Contamination in Products and Packaging

Contact Author Kathleen A. Martin, PhD, McCrone Associates Inc.
Fill out my online form.

The unexpected appearance of dark particles or discoloration in a product raises concerns over the integrity of the product and can cause consumers to reject it. In order to catch problem products before they reach consumers, quality control laboratories typically examine a product’s appearance, as well as other properties, before it is released. Particulate contamination in products can originate from a wide variety of sources via raw materials or from debris in the manufacturing environment. Common types of particulate include metal wear products or corrosion, paint chips, shredded plastic, glass chunks, hairs and insect parts. In some cases, particulates or discoloration also may arise from the agglomeration of ingredients or reactions between ingredients and contaminants.

When particulate contamination or discoloration is observed, the next step is often to pass the sample on to an analytical chemist to identify the nature of the contaminant or discoloration. Samples are generally submitted by quality control or stability testing laboratories, plant engineers and consumer complaint coordinators. Identification is an important step in the process of tracing the origin of the problem so that it can be fixed, safety concerns can be addressed, and in some cases, responsibility for the costs incurred can be assigned.

Isolation of Particulates

Contaminated samples are submitted for particulate analysis in a variety of ways, including raw materials and finished products as well as samples collected from the manufacturing environment in the form of swabs, filters or dismantled equipment. The particulates are usually isolated from these matrices via filtration or manual particle-picking before they can be identified. Particle-picking is usually performed while viewing samples under a stereomicroscope and may be accomplished with forceps for large particles, i.e. > 0.5 mm, or with custom microtools such as a sharpened tungsten needle for small particles, down to about 10–20 µm. In most cases, only visible particles > 50 µm are of interest for identification, although for eye care products, sub- visible particulates may also be of interest.

Excerpt Only This is a shortened version or summary of the article you requested. To view the complete article, please log in or create an account. Registration is Free!

The major considerations in selecting filters for particulate isolation are compatibility with the matrix, pore size and the ability to subsequently remove the particles for analysis. Paper-based filters and woven filters can be effective at trapping particles but frequently are not useful if those particles must be removed from the filter for analysis. Membrane filters are preferred when particles need to be selected for subsequent analysis. Filtration is typically performed using a vacuum apparatus with a polycarbonate membrane filter. Polycarbonate provides an excellent reflective viewing surface at magnifications up to 100x on a stereomicroscope, but it is not compatible with alcohol or other organic solvents. For alcoholic solutions, a polyester filter may be used. Once particulates are isolated, they can be examined with a stereomicroscope and analyzed by one or more instrumental methods.

Analysis and Identification

Stereomicroscopic examination: Whether particles are presented in situ or on a filter or some other substrate, examination under a stereomicroscope is an important first step in characterization. Particle size, morphology, color and reflectivity can all be easily determined with a stereomicroscope for particles as small as 10–20 µm. In addition, particle texture—i.e., gelatinous, hard or elastomeric—can be determined. Different illumination sources are used to observe different types of particulates. Oblique lighting is appropriate for most particles, coaxial, i.e., top light, illumination is useful for thin residues, and transmitted light is best for transparent particles.

Information gleaned from stereoscopic examination is used to decide which instrumental methods to employ. For example, metal particles can be further analyzed by energy dispersive x-ray spectrometry (EDS), which allows for the identification of alloy types, while elastomeric particles can be further analyzed by infrared spectroscopy to determine the nature of the polymer.

Instrumental analysis: Many different types of instruments can be used to identify materials, the descriptions of which are beyond the scope of this article. Therefore, three of the most useful instruments are described here in detail. These include: scanning electron microscopy equipped with energy dispersive x-ray spectrometry (SEM/EDS); infrared spectroscopy (IR); and Raman spectroscopy. SEM/EDS is generally useful for morphology and elemental analysis, i.e., carbon and heavier elements, whereas IR can identify many organic and some inorganic materials. Raman can identify different forms of carbon—i.e., amorphous, graphite, diamond; pigments; and some crystalline materials. Several excellent sources for further reading on these techniques are included in References 1 through 5.


In SEM, an electron beam strikes a sample and generates different kinds of signals, including those that can be used to examine sample morphology and those that can identify and quantitate elements present in the sample. High resolution instruments can provide up to 1,000,000x magnification, while low resolution instruments used for routine analysis can provide up to 300,000x magnification. Images generated using backscattered electrons provide atomic number contrast because heavier atoms backscatter electrons more efficiently than light atoms. Thus, elements with higher atomic numbers appear to be brighter than elements with lower atomic numbers. For example, a polymer or other organic material will appear as a relatively dark object while metallic particles within an organic matrix will appear as bright spots. This capability allows for the determination of elemental homogeneity and metallic or inorganic inclusions in an organic matrix. Elemental analysis, i.e., identification of the elements present plus semi-quantitation of those elements, can be performed with an EDS detector, which measures x-rays ejected by elements following excitation by the SEM electron beam. The ejected x-rays have energies characteristic of the elements present in the sample and allow an elemental profile of the sample to be generated. EDS is useful in identifying metal alloys, inorganic inclusions in organic material such as fillers in paint, and minerals.

Infrared Spectroscopy

Molecules can selectively absorb wavelengths of light that correspond to their vibrational energies, which normally fall within the mid-infrared region (4,000–500 cm-1). The vibrational energies of a molecule depend on the types of functional groups present and their arrangement within the molecule. Spectra are acquired using a broadband infrared source and measuring the light transmitted through or reflected from a sample. A plot of the light that is transmitted or absorbed creates a spectral signature that can be matched to a database. Infrared is sensitive to polar functional groups such as OH, CH, carbonyls, etc., and is widely used to identify organic and many inorganic compounds. When coupled to a microscope equipped with the appropriate optics, particles as small as 20 µm can be analyzed.

Raman Spectroscopy

Raman also provides information about vibrational energies but operates in a different fashion from infrared spectroscopy. Thus, it is sensitive to different aspects of molecular structure. In Raman spectroscopy, monochromatic light—usually a visible or near-infrared laser—is focused on a sample. While most of the light is scattered elastically, i.e., it has the same wavelength as the incident light, a small amount of the light, perhaps 1 in 106 photons, is scattered inelastically. The difference in wavelengths between the incident light and the inelastically scattered light constitutes the Raman effect.

The Raman effect corresponds to the vibrational energies of the molecules present but it is sensitive to nonpolar groups such as the carbon-carbon backbones of polymer chains. Raman spectroscopy is less widely used than infrared spectroscopy for identifying unknown organic compounds but it is useful for many inorganic compounds, particularly carbon materials including graphite and diamonds, and pigments.

Contaminant Case Studies

The following case studies represent real samples in the cosmetics industry. Some details have been changed in order to protect client confidentiality.

Study 1—blue paint chip: Paint chips are a common type of contaminant. A blue paint chip such as the one shown in Figure 1 can be identified using a combination of infrared spectroscopy, Raman spectroscopy and SEM/EDS. Use of all three techniques allows one to obtain a full profile of the composition of the sample, which in turn allows for the assessment of safety concerns and to make comparisons to a suspect source for the contaminant, such as peeling paint from a drum. An infrared spectrum of the blue paint identified it as being based on vinyl acetate with a calcium carbonate filler, and likely talc (see Figure 2). The bands in the spectrum near 1,738 cm-1 and 1,242 cm-1 are indicative of vinyl acetate, while the broad band near 1,434 cm-1 and sharp band near 877 cm-1 indicate calcium carbonate. A band near 1,020 cm-1, together with a weak band near 3,676 cm-1, indicates that talc, a common filler material, is also present.

A Raman spectrum of the paint chip is an excellent match for a reference spectrum of phthalocyanine blue, a common blue pigment (see Figure 3). Note that the Raman spectrum shows only bands due to the blue colorant and not bands due to other components of the paint chip because the pigment is a much stronger Raman scatterer than the other components. EDS identifies the presence of a calcium-rich material, confirming the calcium carbonate observed in the infrared spectrum, and magnesium/silicon-rich particles, which correspond to talc (see Figure 4). In this case, EDS data was used to help confirm the interpretation of the infrared spectrum and also to conclude that the paint was not lead-based and did not contain detectable amounts of any other heavy metals. Study 2–metal particles in an eye care product: At the bottom of a jar of a sample eye care product, a crumbly brown particle, about 2 mm in size, was observed. A photomicrograph of a portion of the particle retrieved from the container is shown in Figure 5. EDSc data indicated primarily iron and oxygen, with trace amounts of chromium and magnesium, corresponding to a heavily oxidized steel flake, possibly from a carbon or low alloy steel with a source of tin and some silicate (see Figure 6). Another jar showed an orange, streaky material that when examined under a stereomicroscope, appeared to emanate from a central black particle about 100 µm in size (see Figure 7). EDS data for the orange streaky material indicated an organic material similar to the normal product, but data for the isolated black particle indicated an iron-rich material, likely iron oxides, intermingled with the organic material. In this case, metal contamination took the form of corroded steel, which resulted in visible particulate contamination and product discoloration. The product discoloration was consistent with the dispersal of sub-micron iron-rich particulate in the product matrix but the possibility of interaction between iron and the product could not be excluded.

Study 3–blue discoloration in a white cream: A white cream was found to have streaks of blue color in it. When examined closely, the blue streaks contained dark blue and gray/brown particles (see Figure 8). EDSc data of the blue and gray/brown particles indicated a high amount of copper in the blue particles (see Figure 9). One of the ingredients in the cream was an acetate ester and it was suspected that copper acetate—a blue/green compound—was formed as the result of copper ions reacting with the acetate compound. The formulation did not include a source for copper, so the copper ions were suspected to have originated from brass corrosion. These ions appeared to react with a product ingredient to form a blue color that bled into the cream matrix.

Study 4–purple discoloration in a pump bottle: A viscous, purple material was found concentrated around the pump in a bottle containing a white product (see Figure 10). A metal spring located within the pump mechanism was examined by SEMc, which indicated it was a type of stainless steel that showed signs of intergranular corrosion (see Figure 11). EDSc data was obtained for non-corroded and corroded areas of the spring. Much higher oxygen levels plus some sulfur and chlorine, which can act as corrosion initiators, were observed for the corroded areas of the spring. The EDS data of the discolored product was similar to that of the normal product but with higher sulfur and chlorine. A laboratory test was thus performed to determine whether iron or chloride ions might have produced the purple discoloration observed in the product.

Normal product was mixed with several different compounds, including iron (III) oxide, iron phosphate, iron (III) chloride, sodium chloride and magnesium chloride. Only the iron (III) chloride produced a deep purple discoloration; the other compounds produced faint discoloration or no discoloration (see Figure 12). It was therefore concluded that corrosion on the metal spring, initiated by sulfur and/or chloride ions, liberated iron ions that reacted with one of the product ingredients to form the purple complex. The product ingredient in question was known to form colored salts with the ferric ion.

Study 5–slippery caps: A batch of metal caps was not performing properly, slipping out of place in a capping operation; thus contamination with a lubricant was suspected. Pieces of several bad caps and control caps were cut out in order to isolate the portions of the caps suspected to be contaminated. Micro-extraction of the metal pieces was performed using chloroform, and the dried extracts were analyzed by infrared spectroscopya. Contrary to expectations, the spectra did not indicate the presence of a lubricant on the bad caps. In fact, the spectra (see Figure 13) appeared to indicate a component in the good caps that was absent in the bad caps. Spectral subtraction (see Figure 14) indicated that the good caps included an epoxy-based component that was not present in the bad caps. It was concluded that the caps that did not perform properly were not contaminated as had been expected, but instead had been manufactured without a key component. In this case, the solution was not related to the cleanliness of the general environment, but was achieved by addressing a specific step in the manufacturing process.


Identification of particulate contaminants can be accomplished by using good particle isolation practices, examining particulates by light microscopy, and analyzing using appropriate instrumental methods. By identifying particulates and discoloration, actions can be taken to prevent further incidents and address safety concerns. Avoiding particulate contamination due to foreign materials is primarily a function of plant hygiene and careful control of raw materials but contamination can never be fully eliminated. Quality control (QC) laboratories generally perform an excellent job of catching visible foreign contaminants before they reach consumers.

In cases where discoloration or particulate is generated from ingredient interaction, the job of QC managers is much more difficult, as this type of contamination may be visible only after a period of time. Discoloration is often due to the presence of metal ions, which may originate from ingredients in the formulation or may be generated from an outside source, such as metal corrosion. Understanding which ingredients may be vulnerable to such interactions is key to their prevention.


  1. D Lin-Vien, NB Colthup, WG Fateley and JG Grasselli, The Handbook of Infrared and Raman Characteristic Frequencies of Organic Molecules, Academic Press Inc, San Diego, USA (1991)
  2. E Smith and G Dent, Modern Raman Spectroscopy—A Practical Approach, John Wiley and Sons Ltd, Chichester, England UK (2005)
  3. G Socrates, Infrared Characteristic Group Frequencies Tables and Charts, Second Edition, John Wiley and Sons Ltd, Chichester, England UK (1994)
  4. IM Watt, The Principles and Practice of Electron Microscopy, Second Edition, Cambridge University Press, Cambridge, England UK (1997)
  5. JI Goldstein et al, Scanning Electron Microscopy and X-Ray Microanalysis, Third Edition, Kluwer Academic/Plenum Publishers, New York, USA (2003)

Related Content



Figure 1. Blue paint chip

Figure 1. Blue paint chip

Figure 2. Infrared spectrum of blue paint chip

Figure 2. Infrared spectrum of blue paint chip (top) and library spectrum of vinyl acetate latex with calcium carbonate (bottom)

Figure 3. Raman spectrum of blue paint chip

Figure 3. Raman spectrum of blue paint chip (top) and library spectrum of phthalocyanine blue pigment (bottom)

Figure 4. EDS spectrum for blue paint chip

Figure 4. EDS spectrum for blue paint chip; EDS indicates calcium-rich particles, titanium-rich particles, and magnesium/silicon-rich particles

Figure 5. Brown particle found in an eye care product

Figure 5. Brown particle found in an eye care product

Figure 6. EDS spectrum for brown particle found in an eye care product

Figure 6. EDS spectrum for brown particle found in an eye care product

Figure 7. Orange particle with dark brown central particle

Figure 7. Orange particle with dark brown central particle, smeared onto a glass slide

Figure 8. White cream with blue discoloration and dark particles

Figure 8. White cream with blue discoloration and dark particles

Figure 9. EDS spectrum for blue particles in the discolored cream

Figure 9. EDS spectrum for blue particles in the discolored cream

Figure 10. Purple discoloration in a pump product

Figure 10. Purple discoloration in a pump product

Figure 11. SEM images a) of base metal and b) corroded area on spring

Figure 11. SEM images a) of base metal and b) corroded area on spring

Figure 12. Formation of deep purple color when FeCl3 is added

Figure 12. Formation of deep purple color when FeCl3 is added to the product

Figure 13. Overlay of infrared spectra of chloroform

Figure 13. Overlay of infrared spectra of chloroform extracts of control caps (blue) and problem caps (red)

Figure 14. Subtraction spectrum of control caps

Figure 14. Subtraction spectrum of control caps minus problem caps (top) and library spectrum of an epoxy resin (bottom)

Next image >