Morphology of actinide-rich particles released from the BOMARC accident and collected from soil post remediation



The physical, chemical, and radiological characteristics of material released to the environment from accidents involving nuclear weapon components are dependent upon many factors, especially the manner in which the material is released and delivered to the environment. These characteristics will also be influenced by physical and chemical effects associated with weathering if the material remains exposed to the environment for a long period of time. This study evaluates the morphological characteristics of particles released to the environment as a result of the 1960 BOMARC incident and compares these characteristics to those described following similar incidents at Thule, Greenland (1968) and Palomares, Spain (1966). Each of these incidents involved unique circumstances and conditions that distributed actinide-rich particles to the environment with a range of distinctive morphological characteristics. Morphological and surface elemental analyses were conducted on a set of discrete particles isolated from samples of post-remediated soil collected at McGuire Air Force Base, the site of the BOMARC incident. Scanning electron microscopy and complimentary energy dispersive X-ray spectroscopy were used to perform the analyses. Non-destructive analysis of uranium and plutonium contained in each particle was measured using high-resolution gamma spectrometry. Unique characteristics of the BOMARC samples include some particles exhibiting a smooth, crystalline texture and varying elemental surface distribution of uranium and plutonium, dependent on the particle’s morphology.


BOMARC Plutonium Uranium Scanning electron microscopy Soil 


Broken arrow is a euphemism that describes an accidental event associated with a nuclear weapon, warhead, or component that may involve a non-nuclear explosion or fire resulting in the spread of radioactive contamination in the environment. The release of nuclear and radiological material into the environment creates a potential health risk for humans depending upon the fate of the material in the environment. Contamination may become available for environmental transport by rain, through ground water, or resuspension in air by wind. Depending upon the physicochemical characteristics of contamination, a fraction of the actinide-rich radioactive material may be leached or dissolved in time from particles as a result of exposure to environmental conditions. The characteristics of particles released as a result of two well-known broken arrow incidents, the 1966 accident over Palomares, Spain, and the 1968 Thule, Greenland, accident, have been well documented in the open literature [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]. The 1960 BOMARC incident, in which a BOMARC missile equipped with a nuclear warhead exploded, and distributed uranium and plutonium particles to the environment, has not been investigated as thoroughly.

Nuclear material from two damaged nuclear weapons was released in a midair refueling accident in 1966 that contaminated local farmland near Palomares, Spain. Four nuclear weapons were destroyed in a 1968 accident near Thule Air Base in Greenland which contaminated sea ice and sediment. The actinide-rich particles released to the environment at Palomares [1, 2, 3, 4, 5, 6, 7] and Thule [5, 6, 8, 9, 10, 11] have been well characterized.

The BOMARC accident occurred at McGuire Air Force Base in New Jersey on June 7, 1960, and involved an explosion of a helium tank in a missile shelter causing a liquid fueled BOMARC missile to burn uncontrolled for approximately 30 min before arrival of firefighters. Water sprayed on the fire for 15 h transported contamination from the shelter to the surrounding area and deposited debris and particles in the soil and on the roads between the shelters. Airborne transport of contamination after the explosion was unlikely due to the quantity of water sprayed on the fire. Post accident remediation removed a significant fraction of the radiological contamination released by the fire, although some residual contamination from weapons grade plutonium (WGP), highly-enriched uranium (HEU) and depleted uranium (DU) remained in the soil. Further remediation was performed between 2002 and 2004 to achieve a soil clean-up criterion of 8 pCi 238, 239Pu per gram (0.296 Bq/g) [12].

The objective of this research was to evaluate the physical and chemical characteristics of actinide-rich particles from the BOMARC accident isolated from samples of soil collected approximately 50 years after the incident and compare these observations with those characteristics reported for the particles collected after the Palomares and Thule accidents. Analysis results obtained for the BOMARC particles include activity of Pu and Am in each particle, size and morphological features, and the elemental surface composition.


Seven discrete actinide-rich particles produced by the 1960 BOMARC incident were isolated from samples of remediated soil. The method to isolate particles first required identifying soil samples likely to contain particles by partitioning approximately 60–80 g of soil into aliquots of 7–10 g and screening with an inverted sodium iodide (NaI) detector to determine the aliquot with the hot particle. After finding the hot aliquot, the soil was scanned in a serpentine pattern using a well-collimated cadmium telluride (CdTe) detector. The collimation of the detector allowed identification of the particle when it was directly underneath the detector. When located, the soil surrounding the detector was cleared away and the portion remaining spread out. The process was repeated until only a tiny fraction was left behind and the particle could be removed for analysis.

Particles were prepared for scanning electron microscopy (SEM) by using double sided conductive carbon tape attached to an aluminum pedestal (12.7 mm diameter, Ted Pella, Inc.) to remove the small sample of soil containing the particle from the Petri dish. The activity of uranium and plutonium was measured by gamma spectrometry using a 3000 mm2 thin window, high purity germanium detector to detect 241Am (59.5 keV), Np L X-rays associated with the decay of Pu, and 235U (185 keV). Weapons grade plutonium contains 241Pu, a relatively short-lived beta emitter, which decays into 241Am that emits a photon with energy equal to 59.5 keV. The pedestal containing the particle attached to the carbon tape is placed under the face of the detector using a repeatable geometry. Detector efficiency measurements were performed using 241Am and 238Pu electroplated standard sources from the National Institute for Standards and Technology (NIST) (SRM’s 4904 s-G-26 and 4906-B-87 respectively). Particle imaging was performed using SEM and EDS. The SEM uses a 30 kV electron beam generated by a Philips 30XL ESEM-FEG with EDAX Genesis version 4.6 EDS. The EDS has a 10 mm2 Si(Li) detector to measure X-ray spectra and create elemental maps of the particle’s surface.


The estimated activity and dimensions of seven selected particles isolated from samples of BOMARC remediated soil are listed in Table 1. Figures 1 and 2 contain, respectively, gamma spectra from the summation of six BOMARC particles and a 239Pu source. Figure 3 contains backscatter X-ray images of each particle. Elemental maps of uranium and plutonium found for select particles are shown in Figs. 4 and 5. The major actinide present in all seven particles was plutonium. Uranium and americium were also identified by EDS and gamma spectrometry respectively.
Table 1

Estimated activity and morphological characteristics of Pu particles









239+240Pu activity (Bq)








241Am activity (Bq)








Length (μm)








Width (μm)
























Fig. 1

Sum of high resolution spectra obtained from measurement of six BOMARC particles: a energy region from 0–65 keV, b energy region from 45–55 keV to detect low yield 239Pu

Fig. 2

High resolution gamma spectrum obtained using an electroplated 239Pu source (UC #5291) a energy region from 0 to 65 keV, b energy region from 45–55 keV to detect low yield 239Pu

Fig. 3

Backscatter electron images of a 4029A, b 4503C, c 7379B, d 7586A, e 7586B-1, f 7586B-2, and g 8615E. The box on Fig. 1a indicates the area analyzed by EDS

Fig. 4

Elemental maps for BOMARC crystalline particle #4503C a uranium and b plutonium

Fig. 5

Elemental maps for BOMARC amorphous particle #8615E a uranium and b plutonium


Plutonium was identified in each of the particles by the detection of 241Am, the decay product of 241Pu. The photon energy of the 241Am photon is only 59.5 keV and is likely to result in a biased estimate of plutonium in the particle due to photon absorption within the particle. Although the isotopic composition of weapons-grade plutonium is available in the literature, it is undesirable to measure 241Am to determine 241Pu as a means to extrapolate the content of 239/240Pu in the particle since the result will be highly uncertain. The values given in Table 1 for 239+240Pu were determined by decay correcting the previously reported 239+240Pu to 241Am activity ratio of 5.4 determined 1997 [13]. Alternatively, 239Pu emits a 56.021 keV photon with very low yield (viz. 0.021 %) that can be seen in the spectrum obtained using the electroplated 239Pu source (Fig. 2b). However, the presence of a large quantity of 241Am in the particle generates a 49.7 keV escape peak in the germanium detector that interferes with the low yield photon from 239Pu (Fig. 1b). Thus, the low yield 56.021 keV peak from 239Pu cannot be used to quantitate plutonium for particles in which a significant quantity of 241Am has grown in from the decay of 241Pu.

The texture of many of the BOMARC particles is smooth and almost crystalline. By contrast, the particles from Palomares have been described as having a granulated texture [1] and angular or rounded shapes [1, 2]. The morphology of the Thule particles were reported as a fluffy amorphous or agglomerated grains [9] or, more simply, as popcorn or spongy [11]. Eriksson [8] has shown some Thule particles as a larger piece of soil containing many small fragments of actinide-rich material. BOMARC particle 4029A is unique among all these particles, as it appears to have punctured a larger soil particle much like a splinter of plutonium. Only two of seven particles (viz., 7586A and 8615E) analyzed for this study exhibit an amorphous or craggy-like texture. Particle 7586A is interesting because it has a smooth, crystalline area in the vicinity of the amorphous region. BOMARC particles 7586B-1 and 7586B-2 were found in the same aliquot of soil and may have separated at some time as evidenced by the fracture lines on 7586B-2. The texture of BOMARC particle 4503C appears crystalline and rippled, not quite smooth; no pitting or erosion is noticeable.

Results of energy dispersive X-ray spectroscopy reveal the presence of plutonium in all seven particles. However, the uranium content is highly variable from essentially none in particle 4029A (no L X-rays could be identified for U) to an almost equal amount in particles 7586B-1 and 7586B-2 (Pu/U Lα X-rays were slightly greater than 2 for each). Other trace elements identified include C, O, Si, Al, and Fe, which are expected to be present in samples of soil. Gallium, an alloy used in some nuclear components, was not identified by EDS in these particles. Previous analyses of Palomares particles indicate the similar quantities of U and Pu according to ratios of their respective Lα energies [2, 3, 4, 6]. Particles from the Thule incident reveal a greater abundance of U than Pu [5, 6, 9, 10, 11].

The EDS provided the elemental surface composition and distribution of uranium and plutonium on the particles. Uranium and plutonium were homogeneously distributed on the smooth, crystalline particles (Fig. 4) whereas the amorphous, craggy particles exhibited a heterogeneous distribution of U and Pu (Fig. 5). Elemental maps for BOMARC particle 4029A were excluded since the EDS spectrum revealed very little uranium. Particle 7586A, which exhibits both crystalline and amorphous qualities, has a higher intensity of U in the crystalline portion than the weathered area, which is predominantly Pu. Reports describing particles from the Palomares and Thule incidents reveal both homogeneous [2, 9] and heterogeneous [11] distributions of uranium and plutonium.


Particles released to the environment from the BOMARC broken arrow incident exhibit a smooth and crystalline structure that is different than the characteristics reported for particles following the Palomares or Thule incidents. Energy dispersive X-ray analysis shows that BOMARC particles have more plutonium than uranium on the surface, which is in contrast with particles reported from Palomares (U and Pu about equal) and Thule (greater U than Pu). The creation of actinide-rich particles resulting from a broken arrow incident depend upon the type of weapon and its nuclear material, mechanisms associated with the release and dispersion of contamination, and the conditions and duration of environmental exposure. BOMARC particles 7586B-1 and 7586B-2 were found in the soil sample aliquot and have similar plutonium and uranium composition. They were likely formed as a single particle that fractured after collection and may signify that particles are fragile. Crystalline particles exhibit a homogeneous distribution of U and Pu whereas the distribution of uranium and plutonium on weathered particles was more heterogeneous. This heterogeneity may be due to uranium being more soluble and mobile (as well as its ability to form carbonates) than plutonium when exposed to environmental conditions. Uranium on the surface of the particles may have leached from the surface of particles with weathered, amorphous surfaces.



This research is supported by the UC Education and Research Center, Safety and Health Engineering Program funded by the National Institute for Occupational Safety & Health and by a grant from the U.S. Department of Energy Nuclear Forensics Academic Education Program.


  1. 1.
    Aragon A, Espinosa A, de la Cruz B, Fernandez JA (2008) J Environ Radioact 99:1061–1067CrossRefGoogle Scholar
  2. 2.
    Lopez Garcia (2007) Nucl Instrum Methods Phys Res B 260:343–348CrossRefGoogle Scholar
  3. 3.
    Jimenez-Ramos MC, Garcia-Tenorio R, Vioque I, Manjon G, Garcia-Leon M (2006) Environ Pollut 142:487–492CrossRefGoogle Scholar
  4. 4.
    Jimenez-Ramos MC (2009) Radioprot 44:345–350CrossRefGoogle Scholar
  5. 5.
    Jimenez-Ramos MC, Eriksson M, Garcia-Lopez J, Ranebo Y, Garcia-Tenorio R, Betti M, Holm E (2010) Spectrochim Acta, Part B 65:823–829CrossRefGoogle Scholar
  6. 6.
    Lind OC, Salbu B, Janssens K, Proost K, Garcia-Leon M, Garcia-Tenorio R (2007) Sci Total Environ 376:294–305CrossRefGoogle Scholar
  7. 7.
    Pollanen R, Ketterer ME, Lehto S, Hokkanen M, Ikaheimonen TK, Siiskonen T, Moring M (2006) J Environ Radioact 90:15–28CrossRefGoogle Scholar
  8. 8.
    Erikkson M, Osan J, Jernstrom J, Wegrzynek D, Simon R, Chinea-Cano E, Markowicz A, Bamford S, Tamborini G, Torok S, Falkenberg G, Alsecz A, Dahlgaard H, Wobrauschek P, Streli C, Zoeger N, Betti M (2005) Spectrochim Acta, Part B 60:455–469CrossRefGoogle Scholar
  9. 9.
    Lind OC, Salbu B, Janssens K, Proost K, Dahlgaard H (2005) J Environ Radioact 81:21–32CrossRefGoogle Scholar
  10. 10.
    Moring M, Ikaheimonen TK, Pollanen R, Ilus E, Klemola S, Juhanoja J, Eriksson M (2001) J Radioanal Nucl Chem 3:623–627CrossRefGoogle Scholar
  11. 11.
    Ranebo Y, Eriksson M, Tamborini G, Niagolova N, Bildstein O, Betti M (2007) Microsc Microanal. doi:10.1017/S1431927607070353 Google Scholar
  12. 12.
    Rademacher SE, Hubbell JL (2007) United States Air Force. Report IOH-SD-BR-SR-2007-0001. Accessed 6 Aug 2012
  13. 13.
    Rademacher SE (1999) United States Air Force. Report AFSC-TR-1999-0002. Accessed 6 Aug 2012

Copyright information

© Akadémiai Kiadó, Budapest, Hungary 2012

Authors and Affiliations

  1. 1.Nuclear & Radiological Engineering Program, School of Dynamic SystemsUniversity of CincinnatiCincinnatiUSA

Personalised recommendations