This is archived material from the Federal Bureau of Investigation (FBI) website. It may contain outdated information and links may no longer function.

Technical Note - Forensic Science Communications - April 2008

Technical Note - Forensic Science Communications - April 2008

April 2008 - Volume 10 - Number 2


Technical Note

Preliminary Evaluation and Sampling of Some Commonly Used Two-Part Epoxies

Stephanie Kleinjan
Forensic Chemist
Forensic Science Laboratory
Bureau of Alcohol, Tobacco, Firearms and Explosives
Walnut Creek, California

Sarah Walbridge
Forensic Chemist
Forensic Science Laboratory
Bureau of Alcohol, Tobacco, Firearms and Explosives
Walnut Creek, California

Abstract | Introduction | Materials and Methods | Results and Discussion | Conclusions | References


Two-part epoxies are encountered in forensic casework; however, little published work has explained how to characterize these materials. In this study, we examined 14 commercially available two-part epoxies using techniques commonly used by forensic laboratories. We found that the discriminating capability of the techniques employed depends on the type of epoxy. Further, even after all visual, microscopic, and instrumental examinations, some epoxies within a given group could not be differentiated from the others.


A variety of two-part epoxies may be purchased from local home-improvement or hardware stores. These adhesives are inexpensive and easy to use and typically have quick curing times. Although the consumer market for two-part epoxies is directed toward legal, common uses, epoxy products and other types of adhesives often are used in the construction of improvised explosive devices (IEDs). Little published work has addressed the characterization of two-part epoxies for the forensic community; yet they are widely encountered (Bakowski et al. 1985; Blackledge 1992; Curry 1987; Noble et al. 1974; Schlipf and Karpf 1997; Wheals 1980/1981). Identifying the type of epoxy used in the construction of an IED is not typically within the scope of a routine examination. However, in some instances an investigative lead is needed or a comparative analysis needs to be made between an epoxy used in a device and that obtained from a suspect. 

This project was intended to be a preliminary evaluation of two-part epoxies using a variety of instrumental techniques. Two-part epoxies were purchased from The Home Depot and ACE Hardware stores. Fourteen epoxies were placed in three different groups based on physical appearance. In addition to the visual and microscopic differences between groups, the epoxies were analyzed further using Fourier transform infrared spectroscopy (FTIR), X-ray fluorescence (XRF), X-ray diffraction (XRD), polarized light microscopy (PLM), and pyrolysis-gas chromatograph-mass spectrometry (Py-GC-MS). These techniques were chosen for two reasons: first, as an effort to evaluate a potential analytical approach for two-part epoxies and, second, because some or all of these techniques are used in the forensic examination of explosives. The comparative data obtained here could aid an analyst with these available instruments in ascertaining which technique would be the most applicable for a particular type of two-part epoxy. The multi-instrumental data obtained here allowed for the epoxies within the groups to be further subdivided and differentiated.

Materials and Methods

Fourteen two-part epoxies were separated into three distinct groups based on physical appearance:

Group 1

  • Transparent/yellow in color
  • Extra Setting Time Epoxy (Super Glue Corporation/Pacer Technology, Rancho Cucamonga, California)
  • Loctite Quick Set Epoxy (Loctite/Henkel Consumer Adhesives, Avon, Ohio)
  • 5 Minute Epoxy (Devcon/ITW Performance Polymers Consumer Division, Riviera Beach, Florida)
  • Gel Epoxy—Super Strength (ACE Hardware Corporation, Oak Brook, Illinois)
  • ACE Quick Set Epoxy (ACE Hardware)

Group 2

  • Opaque/gray in color, two-part liquid
  • J-B KWIK (J-B WELD Company, Sulfur Springs, Texas)
  • Metal/Concrete Epoxy (Loctite/Henkel Consumer Adhesives)

Group 3

  • Opaque putty/gray in color, two-part putty stick
  • FastSteel Epoxy Putty Stick (Polymeric Systems, Incorporated/Whitford Worldwide, Elverson, Pennsylvania)
  • Loctite Express Repair Epoxy Putty (Loctite/Henkel Consumer Adhesives)
  • RectorSeal EP-200 Epoxy Putty (RectorSeal Corporation, Houston, Texas)
  • Leakstopper Steel Epoxy Putty (Devcon/ITW Performance Polymers Consumer Division, Riviera Beach, Florida)
  • PC-METAL Epoxy Putty (Protective Coating Company, Allentown, Pennsylvania)

Each two-part epoxy was mixed according to the manufacturer’s directions and allowed to cure for a minimum of 24 hours. Solid samples of the epoxies were used for XRF, Py-GC-MS, and FTIR analyses. Another working sample was created by softening a solid sample of the epoxy with acetone using a mortar and pestle. This sample was then ground gently to release and disperse the inorganic components from the resin. The ground samples were then observed under the stereomicroscope and used for PLM, XRD, XRF, and FTIR analyses.

Instrumental Parameters


  • PerkinElmer (Waltham, Massachusetts) Spectrum One with Specac (Smiths Group, Cranston, Rhode Island) Golden Gate Single Reflection Diamond ATR Series MkII accessory: solid sample, no manipulation
  • Nicolet (ThermoFisher Scientific, Waltham, Massachusetts) Magna-IR 560: acetone-softened ground sample, KBr pellet


  • EDAX (EDAX/Ametek, Mahwah, New Jersey) Eagle µ-Probe Spectrometer, Supermax 160-5 Detector, EDAM (EDAX Data Acquisition Module) III Analyzer, Vision Software V3.38
  • Rhodium anode X-ray tube with beryllium window, silicon/lithium detector with beryllium window 30 kV, variable mA to obtain ~2000 counts/second
  • Solid samples placed on carbon stub with Spectro Tab (#16084) or Adhesive Tab (#16079) (Ted Pella, Inc., Redding, California)


  • Philips (PANalytical/Spectris, Almelo, The Netherlands) PW1800/10 with PW1808 sample changer
  • MDI Datascan 3.2, MDI JADE 5.0 software (Materials Data Incorporated, Livermore, California)
  • Copper anode X-ray tube, xenon detector with graphite diffracted-beam monochromator kV = 40, mA = 40, 5–80º, 0.050 step size at 1 second per step
  • Ground sample, affixed to silicon zero-background crystal with ethanol


  • CDS (CDS Analytical Inc., Oxford, Pennsylvania) Pyroprobe 2000/HP 6890 GC/HP 5973 MSD (Agilent Technologies, Santa Clara, California)
  • Pyrolysis conditions: 0–750 ºC, 20 ºC/ms, 15 seconds
  • GC conditions: Rtx-5MS column, 40 ºC for 1 minute, 10 ºC/minute up to 300 ºC, hold 3 minutes (30-minute run time)
  • MS conditions: source—230 ºC, Quads—150 ºC, scan 33–350 mz, 4.49 scans/second
  • Kraton standard (0.05 g/1 mL toluene) at beginning and end of day
  • Solid sample with glass wool plug in CDS Analytical (Oxford, Pennsylvania) fire-polished quartz tubes (#10A1-3007)


  • Olympus (Center Valley, Pennsylvania) BX60 Polarized Light Microscope
  • Samples mounted on clean microscope slides in Permount (Fisher Scientific, Pittsburgh, Pennsylvania) (refractive index = 1.525)


  • Olympus SZX12
  • Dolan-Jenner Industries (Boxborough, Massachusetts) Fiber-Lite Model 180 high-intensity illuminator with ring light attachment or gooseneck oblique lighting

Results and Discussion

Group 1 two-part epoxies were transparent and yellow in color and visually similar in appearance (Figure 1a). No significant differences in spectra were observed with FTIR analysis (Figure 1b). No crystalline material was detected in these epoxies by XRD. XRF could differentiate Quick Set Epoxy (Loctite), Gel Epoxy—Super Strength, and Quick Set Epoxy (ACE) from Extra Setting Time Epoxy and 5 Minute Epoxy by the presence of silicon in the samples. Further differentiation between these three epoxies was observed using Pr-GC-MS. Extra Setting Time Epoxy exhibits a distinguishable pyrogram versus the other four epoxies. This is observed by the presence of an additional peak at approximately 5 minutes. The results are tabulated in Table 1.

Figure 1a: Images of Group 1 two-part epoxies

Figure 1b: FTIR spectra for Group 1 two-part epoxies

Table 1: Results of Group 1 Testing

Group 2 liquid two-part epoxies were all opaque and gray in color (Figure 2a). In this comparison, these epoxies were visibly different from one another. For casework, however, one may be limited in sample size and should not rely on a visual differentiation for these epoxies. It should be noted that the J-B WELD products are very similar except for their luster. High interference from the inorganic portion of these epoxies decreased the ability to distinguish them by FTIR alone. PLM analyses also did not yield discriminating data. However, this group could be subdesignated into two groups using XRD and Py-GC-MS. The three products differ in elemental composition as determined by XRF. The J-B WELD products could be distinguished from the Loctite product by the presence of BaSO4 using XRF and XRD. Furthermore, J-B KWIK displayed a distinguishable pyrogram from the other two products, which is observed by the presence of two additional peaks at approximately 4.5 and 7 minutes (Figure 2b). A National Institute of Standards and Technology (NIST) search of these two peaks suggests possible identities of these two compounds but cannot conclusively identify them without further research and proprietary information about epoxy ingredients. The ground samples for the Group 2 epoxies were visually examined under the stereomicroscope using top and oblique lighting. These epoxies contained magnetic, opaque, silver flakes and flat, rectangular crystals. Colorless, transparent, flat/plate-like particles showing second-order and higher interference colors, characteristic of calcium carbonate, were observed. Observations were made with plain polarized light and crossed polars using a first-order red-plate compensator. The results are tabulated in Table 2.

Figure 2a: Images of Group 2 two-part epoxies

Figure 2b: Pyrograms for Group 2 two-part epoxies

Table 2: Results of Group 2 Testing

Group 3 putty-stick two-part epoxies were all opaque and gray in color (Figure 3a). These epoxies were all similar, if not indistinguishable, in visual appearance. An analyst receiving only a small portion of these epoxies would have difficulty differentiating these by appearance alone. High interference from the inorganic portion of these epoxies decreased the ability to distinguish them by FTIR alone. Py-GC-MS analyses also did not yield discriminating data. However, some of the samples within this group could be differentiated by PLM and XRD. The ground samples for the Group 3 epoxies were visually examined under the stereomicroscope using top and oblique lighting. These epoxies contained magnetic, opaque, silver flakes and magnetic, black agglomerations. Colorless, translucent, fibrous, and flat/plate-like particles with low-order interference colors showing positive sign of elongation (+SOE), characteristic of talc, were observed by PLM. All of the epoxies in this group, with the exception of EP-200, contained glass spheres that were observed by stereomicroscopy and PLM (Figure 3b). Leakstopper Steel Epoxy Putty was distinguishable by XRD because of the presence of elemental iron. EP-200 was distinguishable by PLM and XRD because of the presence of CaMg(CO3)2. The results are tabulated in Table 3.

Figure 3a: Images of Group 3 two-part epoxies

Figure 3b: Photomicrographs showing glass spheres found in five of the six samples from Group 3

Table 3: Results of Group 3 Testing


There are little published data about the discrimination of common two-part epoxies. A preliminary evaluation using an array of instrumental techniques was conducted to see if visually similar epoxies could be differentiated and to what degree. As a result of the data obtained, the epoxies within the groups could be characterized further. This information can provide a forensic examiner with information to aid in the determination of epoxy products used in IEDs. It is noted that with Py-GC-MS, reproducible pyrolysis data depend on sample size and sample placement in the quartz tube, and as such, detailed comparisons with pyrolysis data should be made with extreme caution.

From this preliminary evaluation, Group 1 two-part epoxies would be the most difficult to distinguish without performing XRF and/or Py-GC-MS analyses. Group 2 contained the most visually different two-part epoxies whose characteristics could be further subdivided with XRF, XRD, and/or Py-GC-MS. Group 3 two-part epoxies also were very difficult to differentiate on physical appearance alone and appear to be the most complex in their components. Based on these preliminary results, PLM and XRD techniques appear beneficial.

The data obtained in this project are useful for future research. A method of reproducibility for each technique needs to be assessed. The multi-instrumental approach used in this preliminary research appears promising as the best approach in the characterization of two-part epoxies. Additional two-part epoxies, to include marine and gel epoxies, will be purchased and characterized to create a thorough database of the types of two-part epoxies commonly encountered in the analysis of IEDs.


Bakowski, N. L., Bender, E. C., and Munson, T. O. Comparison and identification of adhesives used in improvised explosive devices by pyrolysis-capillary column gas chromatography-mass spectrometry, Journal of Analytical and Applied Pyrolysis (1985) 8:483–492.

Blackledge, R. D. Application of pyrolysis gas chromatography in forensic science, Forensic Science Review (1992) 4(1):2–15.

Curry, C. J. Pyrolysis-mass spectrometry studies of adhesives, Journal of Analytical and Applied Pyrolysis (1987) 11:213–225.

Noble, W., Wheals, B. B., and Whitehouse, M. J. The characterization of adhesives by pyrolysis gas chromatography and infrared spectroscopy, Forensic Science (1974) 3:163–174.

Schlipf, E. and Karpf, M. Differentiation of two-component epoxy adhesives by pyrolysis-gas chromatography, Archiv fur Kriminologie (1997) 199(5/6):143–151.

Wheals, B. B. Analytical pyrolysis techniques in forensic science, Journal of Analytical and Applied Pyrolysis (1980/1981) 2:277–292.