Assessment of bioburden encapsulated in bulk materials

Assessment of bioburden encapsulated in bulk materials

Available online at ScienceDirect Advances in Space Research xxx (2016) xxx–xxx Assessment of biob...

487KB Sizes 31 Downloads 220 Views

Available online at

ScienceDirect Advances in Space Research xxx (2016) xxx–xxx

Assessment of bioburden encapsulated in bulk materials Wayne W. Schubert ⇑, Laura Newlin, Shirley Y. Chung, Raymond Ellyin Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91109, USA Received 8 February 2015; received in revised form 14 February 2016; accepted 18 February 2016

Abstract The National Aeronautics and Space Administration (NASA) imposes bioburden limitations on all spacecraft destined for solar system bodies that might harbor evidence of extant or extinct life. The subset of microorganisms trapped within solid materials during manufacture and assembly is referred to as encapsulated bioburden. In the absence of spacecraft-specific data, NASA relies on specification values to estimate total spacecraft encapsulated bioburden, typically 30 endospores/cm3 or 300 viable cells/cm3 in non-electronic materials. Specification values for endospores have been established conservatively, and represent no less than an order of magnitude greater abundance than that derived from empirical assessments of actual spacecraft materials. The goal of this study was to generate data germane to determining whether revised bulk encapsulated material values (lower than those estimated by historical specifications) tailored specifically to the materials designated in modern-day spacecraft design could be used, on a case-by-case basis, to comply with planetary protection requirements. Organic materials having distinctly different chemical properties and configurations were selected. This required more than one experimental and analytical approach. Filtration was employed for liquid electrolytes, lubricants were suspended in an aqueous solution and solids (wire and epoxy sealant) were cryogenically milled. The final data characteristic for all bioburden estimates was microbial colony formation in rich agar growth medium. To assess survival potential, three non-spore-forming bacterial cell lines were systematically encapsulated in an epoxy matrix, liberated via cryogenic grinding, and cultured. Results suggest that bulk solid materials harbor significantly fewer encapsulated microorganisms than are estimated by specification values. Lithium-ion battery electrolyte reagents housed fewer than 1 CFU/cm3. Results also demonstrated that non-spore-forming microorganisms are capable of surviving encapsulation within, and liberation from, epoxy solids. It must be noted, however, that all purposely spiked experimental solids, resulted in very low recovery (1  10 3–1  10 5 CFU/cm3) of viable organisms. Ó 2016 COSPAR. Published by Elsevier Ltd. All rights reserved.

Keywords: Planetary protection; Encapsulated; Bioburden; Spacecraft; Organic materials

1. Introduction To preserve the scientific integrity of current and future solar system exploration efforts, the National Aeronautics and Space Administration (NASA) imposes cleanliness requirements on all spacecraft intending to land, orbit, or be in the vicinity of any solar system body having the potential to harbor evidence of extant or extinct life. Any ⇑ Corresponding author. Tel.: +1 8183542999.

E-mail address: [email protected] (W.W. Schubert).

mission to a planetary body where water–ice is thought to be present, such as Europa or Enceladus, must also satisfy NASA planetary protection requirements (NPR, specifically 8020.12D) stating that ‘‘the probability of inadvertent contamination of an ocean or other liquid water body (by a viable Earth microorganism) shall be less than 1  10 4 per mission” (NASA, 2011), Section 5.4.1). The calculation of this probability requires conservative estimation of poorly known parameters, and consideration of the following criteria (at a minimum): (i) total bioburden [the total number of contaminant microorganisms (CM) at 0273-1177/Ó 2016 COSPAR. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://


W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

launch], (ii) survival rate of contaminant microorganisms during cruise, (iii) survival rate of CM in the radiation environment adjacent to the target, (iv) probability of CM encountering/landing on the target (including spacecraft reliability), (v) probability of CM surviving landing/impact on the target, (vi) mechanisms and timescale of transport to the subsurface, and (vii) CM survival and proliferation prior to, during, and after subsurface transfer. Calculating this probability can be challenging given the uncertainty associated with each of these factors. The approach generally taken to solve a multivariable problem such as this is to prioritize those factors that are most influential (and controlling) in the probabilistic calculation outcome and focus technical efforts in those areas first. The authors set out to better understand the first criterion, that is total spacecraft bioburden, paying particular regard to the subset referred to as being encapsulated. Encapsulated bioburden represents the sub-population of microorganisms entrained within solid spacecraft materials during manufacture and assembly. Studies performed in the 1960s and 1970s document the numerous challenges encountered in estimating the number of microorganisms trapped within solid materials. The recovery rates for endospores were consistently relatively low, typically less than 3% (Angelotti et al., 1968; Gustan and Olsen, 1971). Due to the uncertainty and difficulty in estimating the number of encapsulated organisms in any solid material, NASA standardized initial bioburden densities for encapsulated microorganisms. The agency did not, however, standardize techniques for obtaining encapsulated bioburden densities, as was done for surface bioburden. In the absence of spacecraft-specific data, NPR 8020.12D (see Appendix D) states that encapsulated bioburden shall be estimated in accordance with the following specification values: 30 spores/cm3 of nonelectronic materials, 130 spores/cm3 of mixed nonmetallic assemblies (inclusive of electronic parts), and 150 spores/cm3 of electronic piece parts. Intuitively, materials of unknown manufacturing origin/history, or those crafted under uncontrolled, non-cleanroom conditions were initially assigned greater bioburden values. Appendix D of NPR 8020.12D also states that for any given surface or solid volume, there are ten-times as many nonsporulating, vegetative cells present as there are endospores. Hence the encapsulated bioburden specification values of 300 vegetative cells/cm3 in non-electronic materials, 1300 vegetative cells/cm3 in mixed non-metallic assemblies (including electronic parts), and 1500 viable cells/cm3 in electronic piece parts. These specification values for encapsulated bioburden, along with scalability factors to account for non-culturable microorganisms, are used to model and estimate the extent of bioburden dispersion and release in the event of spacecraft impact (assuming total pulverization of that hardware). This brutal impact process would release a great deal of energy, which would be manifested in significant temperature increases in resul-

tant debris. In addition, specification values may be overestimated because the chemical makeup of many spacecraft materials is known or suspected to be toxic (e.g., phenolic resins) to microorganisms. Finally, many solid materials are manufactured and/or cured at temperatures greater than ambient, albeit lower than that required to achieve effective microbial reduction (e.g., cure of two-part epoxies). Previously, laboratory approaches such as chemical dissolution, grinding with a kitchen blender, and grating soft pliable materials with a cheese grater were tested with varying success (Benardini et al., 2012, 2013). The authors quickly learned that no one technique could be universally applied to all materials. For example, most cured epoxies are insoluble in solvents tolerated by bacterial spores (Stam et al., 2012). Furthermore, while the authors have successfully recovered bacterial endospores from poly (methyl methacrylate) (PMMA) via dissolution in organic solvents (Mohapatra and La Duc, 2012a,b; Stam et al., 2012), vegetative cell membranes are lysed and cell components are inactivated by these harsh chemicals. Grinding with a conventional food blender can effectively reduce the size of solid materials to fine fragments, but extreme care must be taken to avoid overheating due to friction and thus cell death. Other grinding techniques have been employed by other investigators (Gustan and Olsen, 1971), none of which reported significant endospore recovery rates. Citing reported problems with viable microbial recovery from solids, Bauermeister et al. (2014), reported similar low level polymer bioburden recoveries of <0.1– 2.5 CFU/cm3. They also studied direct microscopic observation and molecular approaches. Based on our laboratory’s previous success in pulverizing spacecraft solids into fine powders, thereby facilitating the recovery of endospores from solid epoxy, Lucite, silicone elastomer coatings, and electronic parts (e.g., integrated circuit chips, resistors), a cryogenic milling technique was used to grind the solid samples created and analyzed in this study. Upon release via cryogenic grinding, encapsulated cells and spores are released from the solid matrices via physical impact, where the solid material easily fractures at liquid nitrogen temperatures. Once liberated, cells and spores are capable of germinating and proliferating into detectable colonies on rich agar growth medium. While this technique may not be ideal, after having conducted a rather thorough learn-as-you-go survey of available approaches, the authors are confident in advocating cryogenic milling as the best method for viable microorganism detection available at this time. Indeed, cryogenic grinding techniques have facilitated rapid and reproducible sample processing while minimizing the heat stress typically associated with grinding. All of this aside, cryogenic grinding remains a conservative approach, as it is a much gentler process than total spacecraft pulverization due to impact. The difference in encapsulated bioburden between specification and ‘‘actual” values remains poorly understood and is likely to be material-specific. In many cases, the

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://

W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

‘‘actual” encapsulated bioburden (assayed empirically) associated with spacecraft materials may be much lower than specification values suggest. Encapsulated bioassay studies performed in support of the Mars Science Laboratory mission reported the presence of 2.16 spores/cm3 of thermal paint, 2.37 spores/cm3 of carbon fiber tank overwrap, and 0.13 spores/cm3 of cured adhesive. All of these results are at least an order of magnitude lower than the 30 spores/cm3 specification value. Even though ca. 62% of the total volume of the Mars Science Laboratory was subjected to one form of microbial reduction or another, and ultimately was extremely biologically clean (Benardini et al., 2014), it was necessary to obtain empirical encapsulated bioburden estimates to ensure that the mission maintained compliance with PP requirements. In the end, if these evaluations had not been performed on these materials, the mission would have had no choice but to apply more penalizing, overly conservative specification values to these volumes. This would have resulted in the mission exceeding its 500,000 total spore bioburden limit, thereby demonstrating non-compliance with NASA requirements, and ultimately precluding permission to launch. Bioburden estimates for future missions indicate that encapsulated bioburden will be a driving factor in considering planetary protection countermeasures for missions to any planetary body suspected of having an ocean or liquid water. In such a case, the probability of impact of a portion or all of the spacecraft must also be considered. To this end, experiments included in the current study were designed to better understand the likelihood of terrestrial microorganisms surviving impact on the target body. The goal of this study was to generate data germane to determining whether revised bulk encapsulated material values (lower than those of historical specifications) tailored specifically to the materials designated in modern spacecraft design could be used, on a case-by-case basis, to demonstrate adequate compliance with planetary protection requirements. The materials selected, tested, and having showed the most promise in the current investigation can then be recommended for future experiments and perturbations aimed at ascertaining their actual, empirically-determined bioburden densities. To achieve the aforementioned goal, two key objectives were delineated: (1) determine the extent of resolvable native bioburden background encapsulated in missionrelevant bulk organic materials, particularly those contributing most largely to the total encapsulated bioburden accounted for the spacecraft, and (2) determine the recovery rates for three predetermined non-sporulating bacterial cell lines upon encapsulation within, and release from, a representative bulk material. The latter geared towards establishing a baseline understanding of the robustness of non-spore formers through encapsulation and release, compared to previously generated endospore data. The results of this experiment promote more accurate estimations of the fraction of the starting population able to actually survive such a perturbation, and be recovered and detected subsequently.


2. Materials and methods The research described here was intended to bolster the current understanding of the extent of encapsulated bioburden typically encountered in select spacecraft materials. A prerequisite of all techniques to be employed was amenability to biology. The methods used in this study were chosen because of their ability to liberate microorganisms from an encapsulated state in various solid materials and transition them to an environment promoting germination, outgrowth, and proliferation. In such an environment (i.e., nutrient-rich agar) the incidence of surviving spores and/or cells could easily be established by simple enumeration of resulting colony forming units (CFU). 2.1. Microbiological assays The NASA standard assay (NASA, 2010) is predicated upon collecting samples onto cotton swab heads, and subsequently liberating biological materials in an aqueous saltbuffered solution. This suspension is then rapidly heated to 80 °C and incubated at this temperature for 15 min, thereby selecting for bacterial endospores. The procedure can also be performed sans heat-shock, thus permitting the survival and detection of both endospores and vegetative cells. The latter version of the assay was most appropriate for this study, as the detection of total viable microorganisms, irrespective of morphology or taxonomy, was the criterion under investigation. After having been mixed vigorously, aliquots of the previously pulverized sample-laden suspension were placed in sterile petri dishes and sterile Tryptic Soy Agar (TSA) was added, mixed, and allowed to solidify. The plates were incubated at 32 °C for up to 168 h and resulting microbial colonies were enumerated colony-forming units (CFU). Though slightly modified to suit the purposes of this study, the NASA standard assay is the standard method by which flight projects are required to assess the extent of microbial burden on spacecraft hardware. The enumeration of CFU as far out as day seven was a deviation from the NASA standard assay (typically 72 h max), one that permitted the detection of slow-growing microorganisms. This delayed evaluation of CFU was also valuable in identifying small colonies in the debris field of the ground materials. Absent undesirable spreading of colonies leading to undercounts, a longer incubation time is conservative. Throughout this study, all test materials were handled aseptically and all experiments were conducted using presterilized reagents, utensils, and equipment. 2.2. Cryogenic grinding to release microbes from a solid material matrix The release of encapsulated microorganisms was accomplished by cryogrinding samples in a Spex Sample Prep Freezer-millÒ model 6770, (Metuchen, NJ). This process was carried out in sterile, sealed 2  10 cm tubes containing a heavy, magnetically driven impacting shuttle that acted as

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://


W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

a hammer. Samples were severed into 4–8 mm pieces, which were then loaded into the cryogrinding tubes. In the Freezermill apparatus, samples were frozen to liquid nitrogen temperatures ( 196 °C) for 10 min prior to grinding. Samples were then cryogenically ground for 9 min at an impact rate of 10 cycles/s. Cryogenically ground solids appear as a fine powder with particle sizes about 25 lM diameter with 80% particles less than 50 lM diameter. After the samples had equilibrated to room temperature, the powder remains of the sample were collected, transferred to test tubes of known mass, and weighed to determine net mass. 2.3. Plating methods specific to ground solids Cryogenically ground solids pose distinct challenges to analysis. To overcome the difficulty of discerning growing colonies from particles in a debris field, the former were enumerated after seven days of incubation, which afforded ample time for proliferation. Highly concentrated suspensions were plated in volumes of 0.5 mL/plate as opposed to the standard 2 mL/plate. There were also issues encountered in working with the standard saline rinse solution, due in large part to the ingredient Tween-80, a non-ionic surfactant. While Tween-80 typically bolsters the release of organisms from surfaces, this surfactant caused undesirable spreading of test microorganisms across entire agar plates. This obscured and confounded the enumeration of resulting CFU on several plates. Consequently, efforts were made to count CFU at earlier time points (e.g., 24 h) to circumvent this phenomenon, with varying success. For all bulk solids analyzed in this study, the total number of microorganisms recovered from all dilutions was summed and reported. 2.4. Materials selection The materials used in this study were selected based on satisfying specific criteria. All of the materials selected were organic compounds. Metallic and ceramic components were not considered as they are typically manufactured at or above sterilizing temperatures. All materials were also deemed acceptable for use in future spacecraft construction. Materials were selected from an approved spacecraft materials list. Emphasis was placed on materials that were likely to have a significant bioburden. Materials used in great abundance with high mass or volume were strongly considered.

Materials likely to have a high temperature treatment step in the manufacturing process were excluded. Materials chosen for testing are listed in Table 1. Having satisfied these criteria, a two-part epoxy adhesive was selected. A shielded, double twisted-wire pair was also selected. As the test wire was dismantled to remove the organic components, the authors discovered that the wire was composed of numerous thick wires, and braded wire shielding, which represented a significantly large surface area per unit length of wire. In the end, the authors decided to consider the wire in its entirety so as to more realistically determine the total (mated surfaces plus encapsulated) bioburden per unit length of wire. Several organic materials that account for large volumes aboard modern-day spacecraft were intentionally omitted from this study. The manufacturing processes for many of these materials (e.g., carbon-fiber laminate, circuit board mounting fiberglass) include high temperature fabrication stages, the ramifications of which can be credited as a form of heat microbial reduction. 2.5. Materials tested The bulk materials tested in this study included: Epoxy Adhesive (Scotch-Weld-3M 2216 B/A Gray), a shielded twisted-wire pair, a typical spacecraft lubricant (Table 1), and the electrolyte components of a lithium-ion battery (Table 2). 2.6. Epoxy A widely used material in the aerospace industry, ScotchWeld-3M 2216 B/A gray adhesive was chosen for bulk encapsulated bioburden analysis. This epoxy adhesive is used as a staking compound to secure integrated chips and other components in the fabrication of electronic circuit boards. The epoxy was mixed according to manufacturer’s Table 2 Selected Li-ion battery electrolyte components for analysis. Chemical name



Diethyl carbonate Dimethyl carbonate Ethyl methyl carbonate Methyl butyrate

105-58-8 616-38-6 623-53-0 623-42-7

Sigma–Aldrich Sigma–Aldrich Sigma–Aldrich Sigma–Aldrich

Table 1 Materials selected for analysis. Material


Spacecraft application

Epoxy adhesive, Scotch-Weld3M 2216 B/A gray Wire, twisted pair with outer teflon cover Cable bundle

3M Corp., St Paul, MN

Staking compound used in electronic board fabrication

Thermax, Carlisle Interconnect Technologies, Nogales, NM JPL special fabrication

Typical spacecraft wire

Braycote 600EF Li-ion battery electrolyte components (see Table 2)

Castrol, BP Lubricants USA Sigma–Aldrich

A typical or representative cable bundle consisting of 10 twisted wire pairs (as above), overwrapped with EMI tape and Kapton tape Lubricant extensively used on mechanisms Lithium-ion batteries

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://

W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

instructions and transferred to sterile 2 mL microcentrifuge tubes. After having cured at ambient temperature for the recommended seven days, the tubes were cut open and the solid epoxy was excised. Epoxy solids were then cryogenically ground and further processed.


electromagnetic insulation (EMI) tape and adhesive Kapton tape. One-cm partitions were cut and subjected to cryogenic grinding, as described above.

2.9. Lubricant 2.7. Wire A candidate wire was selected for encapsulated bioburden analysis. The wire was a double-braided-wire pair, shielded in a braided wire sheath, and covered with an overwrap of Teflon (exterior) and Kapton (interior) (Fig. 1). Specifically described as Wire, Twisted Pair, Shielded, Jacketed, 26 Gauge – Whitmore P/N M27500K26DC2S25. Wire segments ten cm in length were wiped sequentially with acetone, 10% Clorox, and isopropyl alcohol to remove any surface-associated bioburden from the outer surfaces and thus avoid inadvertent contamination. These cleaned segments were then cut into 1 cm partitions, which were cryogenically ground. The encapsulated bioburden contributions from the outer cover and the inner braided wire, including the inside layers, were assessed separately. The cover with the Kapton layer was removed from a 10-cm segment of wire. Then the outer cover/Kapton was cut into 1-cm segments, the surfaces of which were disinfected via an overnight soak in 10% Clorox. These disinfected 1-cm segments were then rinsed in deionized water, air dried, cryogenically ground, and processed further. The inner shielded wire was not disinfected, but proceeded directly to the biological assays. 2.8. Wire cable A wire cable typical of those used in the aerospace industry was subjected to encapsulated bioburden interrogation. The cable consisted of ten braided-pair wires, identical to the wire described above, and an overwrap of

Braycote 600EF is a PTFE (polytetrafluoroethylene) lubricant dispersed in a perfluorinated polyether liquid medium. The abundance of microbes entrained in this lubricant was determined by either of the two following methods: Homogenization. A Waring blender (Stamford, CT) was used to homogenize the lubricant in an equal volume (1:1) of standard planetary protection rinse solution (NASA, 2010). Once homogenized, the entire volume was transferred to petri dishes in 5 mL aliquots. Molten Trypic Soy Agar (42 °C) was then added to each plate and all plates were incubated at 32 °C for seven days. At this time CFU were enumerated and recorded. Distribution atop filter paper. A quantity of lubricant was deposited onto a pre-sterilized Whatman #1 filter paper. Using sterile gloves, the lubricant was actively applied onto a filter paper by wiping the lubricant and filter paper in a sterile petri dish, evenly distributing the material over the paper. The amount of lubricant tested was determined by weighing the paper before, and again after, the lubricant was applied. Seeded filter papers were cultured atop TSA with a double-strength agar overlay applied to cover the entire lubricant-paper sample. Plates were incubated for seven days and resulting CFU were enumerated and recorded. 2.10. Electrolyte components of lithium-ion battery The lithium-ion batteries anticipated for use on future missions have electrolyte formulations that include the four

Fig. 1. The double braded-wire pair used in this study is depicted showing the internal structure. From left to right the wire structure consists of a white outer Teflon polymer, an inner orange Kapton layer, a braded wire shield, the first inner wire with gray polymer cover, the second inner wire showing an orange Kapton layer and the inner multi-strand wire. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://


W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

components tested (Table 2). One additional Li-ion battery compound, lithium hexafluorophosphate, was not tested. This chemical is known to readily form hydrofluoric acid upon contact with water. A previously standardized filtration procedure (Eaton et al., 1998) was used to enumerate microorganisms in each of the four electrolyte solutions. As mixed cellulose acetate membrane materials were incompatible with the solvents under examination, polypropylene membranes were used for three of the four components. Polytetrafluoroethylene (PTFE) filters were used for the fourth component. All filters had pore diameters of 0.45 lM, as this pore size is suitable for capturing the vast majority of bacterial cells and spores. Electrolyte solutions were collected from original containers, transferred to sterile 50 mL centrifuge tubes, and stored under argon gas. Four-mL aliquots of each of these solutions were then filtered through either sterile polypropylene membranes or polytetrafluoroethylene (PTFE) filters. Each membrane or filter was washed three times with 10 mL of phosphate buffered saline (PBS), and was subsequently placed atop a pre-poured tryptic soy agar plate. Agar plates were incubated at 32 °C for seven days, at which time CFU were enumerated and recorded. Per the NASA standard assay, CFU were also evaluated at 72 h. 2.11. Selection of test microbes A suite of non-endospore forming bacteria were selected to be encapsulated in, and liberated and recovered from, epoxy adhesive (Table 3). Candidate organisms were selected based on previously observed or reported characteristics, such as desiccation resistance, radiation tolerance, and/or prior isolation from spacecraft. 2.12. Encapsulating microorganisms Selected bacterial strains were grown on TSA plates into stationary phase. At this time cells were harvested directly from each plate with a sterile loop. Cells were then washed twice in PBS and resuspend in 1 ml of PBS. Aliquots of the final suspension were stored at 4 °C for and later analyzed to quantify the initial liquid-borne microbial population. The 1 mL cell suspension was then thoroughly mixed into cryogenically ground Scotch-Weld-3M 2216 B/A epoxy powder. These cell-infused mixtures were then placed in 100 mL media bottles, which were covered with airpermeable cleanroom wipes and allowed to dry. Once

dry, the mixture was placed in a slowly-rotating mixer (5 rpm) for three to five days, at which time they were incubated in a desiccator until used for encapsulation. Subsamples of this cell-infused epoxy powder were subjected to microbiological assay and resulting population estimates served as the initial (t = 0) population values at the outset of encapsulation into the epoxy. A predetermined amount of the cell-infused powder was mixed together with parts A and B of Scotch-Weld-3M 2216 B/A Gray (in accordance with manufacturer’s instructions). Accurately measured masses of cell-infused powder with known cell bioburden (approximately 1 g per 70 g of epoxy parts A and B) were added to the unmixed epoxy components. The components were aseptically mixed by hand using presterilized wooden spatulas in aluminum dishes until a uniform appearance was achieved. The mixture was then distributed in 1 mL fractions into microcentrifuge tubes and allowed to cure. After having thoroughly cured (ca. 3 days), the cell-infused epoxy solids were cryogenically ground and microbiologically assayed (as described above). Survivor fractions were estimated according to the relative abundance of emerging CFU on tryptic soy agar, a nutrient-rich medium. Microorganisms were separated from material particulates via vortex mixing (1 min) followed by ultrasonic vibration for 2 min. Serial dilutions were prepared for each test coupon and aliquots ranging from 0.5 to 2.0 mL each were distributed to petri dishes containing molten Trypic Soy Agar (as described above). Positive controls of cell-infused powder established the cell population on the desiccated powders prior to encapsulation. Controls of non-inoculated cured epoxy exposed to sterilizing temperatures were included. 3. Results and discussion The extent of bioburden encapsulated within small pieces (ca. 1–2 g) of a two-component epoxy, a standard shielded wire, a lubricant, and the electrolyte components of a lithium-ion battery was assayed. Empirically determined encapsulated bioburden indices were significantly (at least an order of magnitude) lower than corresponding specification values (300 total viable microorganisms/cm3) would estimate for all materials examined. Data resulting from these studies are numerical values of viable microbes and estimated as colony forming units. The values were averaged and reported as the number of viable microbes per volume and or mass. Standard deviations were calculated from replicate samples.

Table 3 Bacterial species encapsulated in epoxy adhesive. Microorganism




Desiccation tolerant?

Gram reaction

Acinetobacter radioresistens Deinococcus radiodurans Staphylococcus xylosus

OB 305 ATCC 13939 OB 205

Viking spacecraft, isolate ATCC, type strain Viking spacecraft, isolate

No No No

Yes Yes Yes

Negative Positive Positive

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://

W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

(21.6 CFU/g or 27.4 CFU/cm3), possibly a consequence of incomplete chemical disinfection or the lower specific gravity of this material compared to the greater density of the metals in the other fractions. Furthermore, the cover itself is a laminate, so it is possible that microorganisms became trapped between the two layers. An attempt was also made to assess the intrinsic encapsulated bioburden of a cable bundle consisting of ten wires identical to the single wire described above. This cable bundle was wrapped by several layers of Kapton adhesive tape, standard for spacecraft hardware. While the cable and its Kapton sheath could be cryogenically ground to a fine powder, upon returning to ambient temperature the adhesive component of the Kapton tape agglutinated the powder particulates into a solid lump that could not be analyzed. Attempts to resolve this issue via cryogenic grinding in the presence of water–ice or powdered talc were unsuccessful.

Table 4 Extent of native background encapsulated bioburden associated with epoxy. Epoxy Scotch-Weld-3M 2216 B/A Gray



Mean of 10 epoxy solids analyzed

3.2 ± 5.5

4.3 ± 7.6


3.1. Determining the extent of background encapsulated bioburden 3.1.1. Epoxy adhesive The results of experiments interrogating the bioburden encapsulated in epoxy adhesive solids were recorded as CFU/gram, as accurate measurements were made of the fine powder analyzed. These values were then converted to volumes (CFU/cm3) in order to compare such results directly with specification values. This was accomplished by accounting for the specific gravity of the epoxy solid. In the end, the resulting value of 4.3 CFU/cm3 (Table 4) was well below the historical specification value of 300 microorganisms/cm3. Cryogenically ground solids appeared as a fine powder. Particle sizes for an analogous epoxy, Hysol EA 9396 that was cryogenically ground for 9 min had an average particle size of ca. 25 lM diameter with 28% less than 10 lM diameter and 80% particles less than 50 lM diameter.

3.1.3. Lubricant The lubricant examined in this study housed a relatively small number of encapsulated microbes (3.02–3.89 CFU/cm3). The two parallel methods of analysis employed yielded similar results (Table 6). As this lubricant is hydrophobic, assaying as much surface area as was feasible was critical to obtaining meaningful results. Both cellulose filters and suspension in standard saline rinse solution appeared to distribute the lubricant effectively. While CFU enumerated after seven days of incubation were relatively low, continued observation of the plates long past the end of the experiment revealed additional colonies and fungal growth.

3.1.2. Wires Considering that any mission will require hundreds of meters of wire, the background encapsulated bioburden associated with this material is of immense consequence. The wire examined, in its entirety having disinfected the outermost shielding surface, was determined to harbor 5.9 CFU/g or 15.9 CFU/cm3 (Table 5). This was significantly lower than the specification value. The inner wire, composed of a braided shield and internal braided wire pairs, housed a mere 0.85 CFU/gram or 3.17 CFU/cm3. The specific gravity of the metallic portion comparatively lowers the CFU/gram values. As this fine braided wire represents a large mated surface area (due to the multiple layers of insulation and numerous wire strands), this is of importance for planetary protection because even low level bioburden on large surface areas is a significant concern. Mated surfaces in this type of configuration cannot be sampled by swab or wipe methods in the same way as are free exposed surfaces. The resulting encapsulated bioburden values for the cover layers of the outer wire, however, were greater per unit mass and volume. The wire cover was burdened with a substantially greater microbial population

3.1.4. Electrolyte components of lithium-ion battery The four electrolyte components of the lithium-ion battery examined were quite clean (Table 7). Dimethyl carbonate (DMC) yielded no detectable CFUs, and the few colonies that arose from the other components were of fungal lineage. The volatile nature of these solvents necessitated filtering in a chemical fume hood rather than a bio-containment cabinet. As such, it is likely that the fungal colonies observed were contaminants. Polypropylene membrane filters ended up being the filter of choice. PTFE filters exhibited a mottled appearance, while standard cellulose nitrate and cellulose acetate membranes dissolved in the presence of these solvents. Further investigation into the suitability of the polypropylene membrane, and per-

Table 5 Intrinsic encapsulated bioburden of a shielded wire pair and its components. Part tested

Number of independent determinations

Total mass tested (g)

Wire (entire wire) Wire-without cover Wire (cover only)

5 3 3

10.4 7.1 1.8

CFU/g 5.9 ± 8.6 0.8 ± 0.8 21.6 ± 5.1

CFU/linear cm


0.6 ± 0.8 0.2 ± 0.2 0.6 ± 0.1

15.9 ± 23.4 3.2 ± 3.2 27.4 ± 6.5

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://


W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

Table 6 Background encapsulated bioburden associated with a common lubricant. Lubricant


Total mass (g)

Number of determinations



Braycote Braycote

Cellulose filter distribution Suspension in PP rinse solution

2.5 1.8

11 1 (34 plates)

2.1 ± 2.1 1.6 ± 5.3

3.9 ± 3.9 3.0 ± 10.0

Table 7 Intrinsic encapsulated bioburden of electrolyte components of a Li-ion battery. Electrolyte components



Dimethyl carbonate Diethyl carbonate Ethyl methyl carbonate Methyl butyrate

0±0 0.2 ± 0.2 0.05 ± 0.1 0.05 ± 0.1

0±0 0.19 ± 0.2 0.05 ± 0.1 0.06 ± 0.1

haps PTFE, filters is warranted should electrolyte components be selected for more rigorous testing in the future. A battery component not examined, lithium hexafluorophosphate, forms hydrofluoric acid in the presence of water (Xu, 2004). Fluorine gas (F2) has been shown to be sporicidal (Schubert et al., 2001) and readily dissociates into hydrofluoric acid. Other halogen solutions, such as chlorine, bromine, and iodine are also known to be effective disinfectants. Experimental results, in addition to the known reactivity of the halogen series, suggest that lithium-ion battery components are themselves selfsterilizing solutions. 3.1.5. Assessing the recovery of viable cells after encapsulating in epoxy An effort was made to assess the extent of encapsulated bioburden having the potential to be transported in bulk organic spacecraft materials. Large populations of viable, non-spore-forming microorganisms were intentionally encapsulated in an epoxy matrix and the percent recovery of survivors was assayed following liberation via cryogenic grinding. Non-spore-forming, vegetative bacterial cells were specifically chosen to evaluate the potential of microbial survival amidst the many challenges that the encapsulation process presents (e.g., chemical exposure, thermal flux of curing epoxy). The numbers of all three strains were significantly reduced by the initial desiccation step (Table 8). The populations of Acinetobacter radioresistens, Deinococcus radiodurans, and Staphylococcus xylosus were reduced by 6-, 8-, and 6-orders of magnitude, respectively

after preparation by desiccation only. Subsequent encapsulation and cryogrinding procedures further reduced these populations by three to five orders of magnitude. Unexpectedly, nearly all D. radiodurans cells were inactivated, with only a few CFU remaining after encapsulation and liberation via cryogenic grinding. While these losses may at first seem immense, it is important to realize that significant numbers of two strains of non-spore-forming bacterial cells survived. These results, for the first time ever, demonstrate that non-spore-forming microorganisms inoculated in high number can survive encapsulation within, and be recovered from, solid organic materials. These findings indicate that, with regard to prolonged survival in an encapsulated state, organisms cannot be excluded from consideration simply because they cannot form endospores. The background encapsulated bioburden of epoxy material received directly from the manufacturer, sans addition of test microorganisms, was 3.2 CFU/g (Table 4). This low number of background encapsulated organisms likely reflects a relatively clean manufacturing environment, as well as the harsh nature of the curing process. In comparison, artificially encapsulated microorganisms prepared in highly concentrated titers were recovered at an incidence ranging from 1.8 to 91,000 CFU/g (Table 8). Means of assessing spacecraft hardware-associated encapsulated bioburden are anticipated to be specific for each mission. As such, the materials to be tested would be dependent on the target bioburden required, the total volume of organic materials, and their contribution to the overall bioburden. Electronic piece parts and conformal coatings, for example, are expected to remain large contributors to the estimated encapsulated bioburden of any future mission. Looking to the future, a more diverse suite of materials, as well as greater quantities (ideally a minimum of 10% of the amount present on the spacecraft) needs to be examined in a manner similar to that described here. Such studies will continue to foster and promote a more thorough understanding of the extent of bioburden

Table 8 Recovery of test microorganisms before and after encapsulation in epoxy adhesive. Microbe


Initial population of cells added to carrier powder

Encapsulated (CFU/g cured dry epoxy powder) starting population

Recovered population (CFU/g epoxy)

Fraction recovered

Recovery (%)

Acinetobacter radioresistens Deinococcus radiodurans Staphylococcus xylosus

Viking Mission OB 305 ATCC 13939 Viking Mission OB 205

1.5  1012 ± 1.6  1011 4.7  1011 ± 8.1  1011 5.6  1010 ± 1.6  109

9.2  106 ± 2.4  106 6.9  103 ± 4.1  103 3.2  107 ± 1.0  107

9.1  102 ± 8.5  102 1.8  100 ± 2.2  100 9.1  104 ± 2.7  104

9.9  10 2.6  10 2.8  10

0.0099 0.027 0.28

5 4 3

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://

W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

intrinsically encapsulated in bulk materials used in spacecraft manufacture and assembly. Similarly, accurately estimating the percent incidence of non-spore forming microbial lineages entrained in such materials could factor significantly into the implementation of current and future planetary protection countermeasures for spacecraft hardware. An improved understanding of the conditions favorable to a low background encapsulated bioburden, including cleanroom conditions for fabrication, aseptic handling, and elevated temperatures during manufacturing, would aid in limiting the number of materials required to undergo destructive bioassay testing. Additional studies must be conducted to validate the cryogenic grinding technique described here. In addition, other strategies must be developed to analyze difficult materials, such as wet lubricants and pressure-sensitive adhesives.


harbored less than 1 CFU/cm3, which is significantly favorable because these batteries cannot be subjected to dry heat microbial reduction processes. This study found that nonspore-forming microorganisms are capable of surviving encapsulation within, and liberation (via cryogenic grinding) from, epoxy sealant solids. While the survival fraction was very low, it does imply the survival potential in materials for microorganisms. The low native background in encapsulated bioburden of the materials examined, in addition to the limited survival fraction of microorganisms purposefully encapsulated into solid matrices, are extremely valuable findings with respect to accurately gauging and constraining the contamination threat from spacecraft bioburden. Applying empirically determined estimates of encapsulated bioburden makes sense, and should prove valuable in lowering the total accountable bioburden for a given spacecraft.

4. Conclusions For each and every material examined, the results of bioburden analyses showed that the incidence of encapsulated microorganisms was considerably lower than specification values would estimate (NASA, 2011). These results are particularly meaningful since none of the samples were subjected to heat-shock (i.e., no selection for endospores), and as such the resulting populations observed represented an aggregate of non-spore-forming organisms and bacterial endospores (although the contribution of each fraction spore-forming vs. non-spore-forming was not determined). Hence, this approach results in empirically-derived estimates of the total microbial population capable of surviving and growing under the test conditions, not just endospores. The lower encapsulated bioburden values observed suggest that subsequent dry heat microbial reduction treatment(s) might be adequate to lower initial encapsulated bioburden to levels approaching sterility. The recovered fraction of non-spore-forming microorganisms intentionally encapsulated within the 3M 2216 epoxy was significantly lower than the initial population inoculated prior to curing and cryogenic grinding. While the encapsulation, liberation, and recovery pipeline did not entirely eliminate the encapsulated non-spore forming microbial populations, the fractions recovered were reduced by three to five orders of magnitude. This is a smaller recovery than the two orders of magnitude reduction typically observed for encapsulated Bacillus atrophaeus endospores subjected to this process. While a small fraction of the non-spore-forming microbes survived the encapsulating and release processes, these vegetative cells strains were by no means as robust as bacterial endospores. The recovery fraction for intentionally encapsulated microbes was not applied to any other experiments described in this study. All spacecraft materials tested yielded encapsulated bioburden of less than 30 CFU/cm3, which is important because it matches currently applied standards. Chemical components of the lithium-ion battery electrolyte reagents

Acknowledgments Copyright 2015, California Institute of Technology. U.S. Government sponsorship acknowledged. The research described in this communication was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. The authors thank Alissa (Tenuto) Korsak of the University of Idaho for desiccation resistant strains and valuable information, Paul Willis for advice regarding materials selection, Subha Comandur for cabling and wire test materials, Steve Dawson, Kumar Bugga, and Charlie Krause for Lithium-ion battery electrolyte components, Eric Oakes for epoxy adhesive, Myron La Duc for reviewing this manuscript, and J. Andy Spry for mission and policy consultation. References Angelotti, R., Maryanski, J.H., Butler, T.F., Peeler, J.T., Campbell, J.E., 1968. Influence of spore moisture content on the dry-heat resistance of Bacillus subtilis var. niger. Appl. Microbiol. 16 (5), 735–745. Bauermeister, A., Mahnert, A., Aurerbach, A., Bo¨ker, A., Flier, N., Weber, C., Probst, A.J., Moissl-Eichinger, C., Haberer, K., 2014. Quantification of encapsulated bioburden in spacecraft polymer materials by cultivation-dependent and molecular methods. PLoS One 9 (4), e94265. Benardini, J.N., La Duc, M.T., Beaudet, R.A., Koukol, R., 2014. Implementing planetary protection measures on the mars science laboratory. Astrobiology 14, 27–32. Benardini, J.N., Morales, F., Schubert, W.W., Kazarians, G.A., Koukol, R.C., 2012. Employing a Grinding Technology to Assess the Microbial Density for Encapsulated Organisms. National Aeronautics and Space Administration, NASA Technical Reports Server (NTRS). Document ID: 20120011889, . Benardini, J.N., Koukol, R.C., Kazarians, G.A., Schubert, W.W., Morales, F., 2013. Using a Blender to Assess the Microbial Density of Encapsulated Organisms. National Aeronautics and Space Administration, NASA Technical Reports Server (NTRS) Document ID: 20130013561. .

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://


W.W. Schubert et al. / Advances in Space Research xxx (2016) xxx–xxx

Eaton, A.D., Clesceri, L.S., Greenberg, A.E., Franson, M.A.H., 1998. Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Washington, DC. Gustan, E.A., Olson, R.L., 1971. Release of Microorganisms from Solid Materials. Technical Report NASA-CR-121707, NASA Technical Reports Server (NTRS). Document ID: 19710024580, California Institute of Technology, Jet Propulsion Laboratory, Pasadena, CA. Mohapatra, B., La Duc, M.T., 2012a. Rapid detection of viable bacillus pumilus SAFR-032 encapsulated spores using novel propidium monoazide-linked fluorescence in-situ hybridization. J. Microbiol. Meth. 90, 15–19. Mohapatra, B., La Duc, M.T., 2012b. Evaluation of fluorescence in-situ hybridization to detect encapsulated Bacillus pumilus SAFR-032 spores released from poly(methylmethacrylate). Microbiol. Immunol. 56, 40–47. NASA, 2010. Handbook for the Microbiological Examination of Space Hardware NASA-HDBK-6022, National Aeronautics and Space Administration, Washington, DC, .

NASA, 2011. Planetary Protection Provisions for Robotic Extraterrestrial Missions. NPR 8020.12D, April 2011, National Aeronautics and Space Administration, Washington, DC. . Schubert, W.W., Yavrouian, A.H., Plett, G.A. 2001. Sterilization of Surfaces by Fluorine Gas. National Aeronautics and Space Administration, NASA Technical Reports Server (NTRS). Document ID: 20060033729, hdl:2014/16475, . Stam, C., Bruckner, J., Spry, A., Venkateswaran, K., La Duc, M.T., 2012. A molecular method to assess bioburden embedded within siliconbased resins used on modern spacecraft materials. Int. J. Astrobiol. 11, 141–145. Xu, K., 2004. Nonaqueous liquid electrolytes for lithium-based rechargeable batteries. Chem. Rev. 104, 4303–4417.

Please cite this article in press as: Schubert, W.W., et al. Assessment of bioburden encapsulated in bulk materials. Adv. Space Res. (2016), http://