The effects on research specimens and museum collection items from electron beam irradiation of mail by the US Postal Service

1. Proposed radiation technology to be used in sterilization of mail sent by USPS

The US Postal Service intends to irradiate selected mail to sterilize it from possible anthrax contaminations, using high energy electron irradiation technology. Irradiation facilities are to be established at a number of major mail handling centers. At this time it is unclear whether irradiation will be restricted to first class mail or whether package mail is also targeted for this treatment. Some information on the methodology was obtained from Mr. Jeff Boeger of the SureBeam Corporation, which manufactures the 150 kW linear electron accelerators that will be used to irradiate mail by the USPS. This equipment produces electrons with an energy of 10 MeV. Electrons of such high energy have a relatively high penetrating power; for example their range in aluminum is about 2 cm. In organic materials, mainly composed of elements with a much lower Z value, this range will be appreciably longer, and for mail (paper, cardboard) it will be approximately 30 cm. The technology is used in the irradiation of food, for purposes of pathogen sterilization, where the packaged food is transported on a conveyor belt through the radiation field. This same technique would be used for the application to USPS mail.

Besides the biocidal application, radiation is also used industrially to initiate specific chemical reactions (such as polymerization of synthetics) or, through the chemical effects, to affect the properties of materials. A few examples of the latter: the plastic coating on virtually all electrical wire is irradiated to render it more resistant to weathering; heat-shrink tubing is irradiated polypropylene; and the non-stick coating of cookware is irradiated Teflon -- where the non-reactivity of Teflon is modified so that it will adhere to the metal cookware surface, while still retaining most of its non-stick characteristics.

2. Interaction of radiation and materials

High energy irradiation causes the deposition of large quantities of energy in the irradiated material, and this in turn causes chemical reactions that are responsible for the desired as well as undesirable effects. The irradiated matter absorbs the energy through ionization. Thus, an electron from the accelerator can hit an atom in the target material, and in this process the former may lose all or part of its energy. The amount of absorbed energy minus that needed to induce the ionization is transmitted to the ionization products; the electron(s) emitted from the atom will move at high energy through the material and induce secondary ionizations. If the original electron did not transfer all its energy in the first interaction with a target atom, the process is repeated, until all its energy has been transferred. The secondary ionization process will continue till all the kinetic energy of the accelerator particle has been used up in ionization processes.

The ions that are formed will eventually recombine with free electrons, but these recombined atoms (and molecules) will be highly energetic and chemically reactive. Free radicals are formed, and many of these may have life times well beyond the time span involved in the initial processes. The number of chemical reactions of a given type that take place will depend on the total amount of energy deposited. This is represented in the concept of radiation "dose", expressed in Grays, notation Gy. The latter unit represents the deposition of 1 Joule of energy per kilogram of irradiated matter. Similarly, the dose rate, a measure for the rate at which the energy is deposited, can be expressed in Gy/min. Obviously, the total dose is a product of the dose rate and the time of irradiation, and the dose rate, in this case of high energy electron irradiation, is a function of the particle energy and the beam intensity. It is of interest to note that the yield rate of a radiation induced chemical reaction often is itself somewhat dependent on the dose rate, but for the purposes of this discussion paper that effect has no great significance.

The induced chemical reactions are the basis of all practical applications of radiation technology. In the biocidal applications, they cause damage that leads to the demise of the organism. The dose needed to induce sufficient damage depends very much on the type of organism. Generally, the lethal dose is inversely related to the complexity of the organism. For example, in highly developed life forms such as humans, a dose of around 1 Gy to the whole body will kill enough cells in vital organs to cause death. For insect control, doses of around 500 Gy are needed for a satisfactory kill rate. Microorganisms, which have no vulnerable cell structures, are killed by major destruction of their DNA, requiring much higher radiation doses for eradication, in the order of tens of kGys. Food irradiations are typically performed with doses of 1.5 - 3 kGy, while eradication of fungal spores requires doses of around 10 kGy and higher.

The dose to be applied in the USPS mail irradiation for the protection against anthrax spores appears still to be a matter of discussion. Yet, it can safely be assumed to well exceed the 10 kGy level. Dr. Burrell Smittle of the Florida State Department of Agriculture expressed the opinion that levels of about 25 kGy would be used, while Dr. Donald Thayer of the USDA Research Service and Dr. Steven Seltzer of NIST indicated an anticipated use of significantly higher doses in the order of 50 - 60 kGy. It was also noted that, if mail were to be irradiated from both sides, this dose would be doubled. These very high doses are needed to obtain the sought after "kill ratio" which is in the order of 12-14 decades (in other words, the fraction of surviving spores is intended to be only in the order of 10-11 to 10-13 ).

Ultimately, the deposited energy will be converted to thermal energy, causing a rise in temperature of the irradiated material. For the conditions considered for use by USPS, this effect could amount to a temperature raise in the order of 5 degrees centigrade.

3. Radiation effects on materials

As mentioned above, the large quantities of energy deposited during irradiation in the target materials leads through the formation of ions, activated atoms and molecules, and free radicals, to a complex series of chemical reactions, and these can have a very significant effect on the chemical and physical properties of the irradiated compounds. These effects can be even more enhanced if the irradiation takes place in a regular atmospheric environment when reactive species such as ozone, O* and OH* radicals are formed.

The number of occurrences of a given reaction depends on the dose. Thus, the amount of induced change in material properties can, like the biocidal efficacy, be controlled by the size of the administered dose. Yet, the amount of change that is permissible (or desirable), depends on the nature and use of the irradiated materials and objects: what may be regarded as trivial effects in the context of industrial applications can be unacceptable in the case of museological and archival collection holdings. It is this latter context with which we are concerned here, and the following discussion pertains to material effects on a scale of magnitude that might compromise the value of such collection materials.

The reactions that we are concerned with include the destruction of existing molecules (chain scission and depolymerization, removal of functional groups as in deamination, decarboxylation etc., and oxidation) as well as the formation of new ones (through recombinations and cross-linkages.) While inorganic materials are not immune to radiation induced effects (and later we will discuss some of these as they are of concern), it is the organic materials that are most vulnerable to significant damage. Literature data on damage rates in the ranges that are of concern to museum and archival collections are limited, but a certain amount of work has been done in order to assess the applicability of radiation technology for biodeterioration control in collections. In 1995, SCMRE was the organizing host to a expert consultants meeting on that subject sponsored by the International Atomic Energy Agency (IAEA), attended by experts from Europe and the USA. Generally, most of the information presented at that meeting still represents the current level of knowledge, since in subsequent years the development of alternative, far less aggressive methods for effective biodeterioration control in a museum collections context have made the application of radiation technology for the purpose something of limited, and at best occasional, utility. Additional information can be gathered especially from literature concerned with the sterilization of food, medical supplies and various other industrial commodities.

Of first concern are the polymeric materials, both natural and synthetic. The natural polymers are more vulnerable to significant change than their synthetic counterparts, and of the natural polymers, cellulose is the most vulnerable. The reactions of concern are chain scission, cross linkage, and oxidation. The effects of these various reactions are depolymerization, loss of strength, embrittlement, acidification and discolorations, and a greatly enhanced rate of subsequent aging deterioration. Quite a lot of experimental work has been done on radiation induced damage to cellulosic materials, since it was hoped that this technology could be used to address one of the major problems in the library and archives field i.e. mold growth in collections that have been exposed to water, for instance during the dousing of a fire. Work done in collaborations between the Centre d’études nucléaires de Grenoble and the Central Laboratory of the Netherlands Institute for Cultural Heritage, indicates that, in order to avoid an unacceptable amount of damage to paper, the dose has to be kept below 2 kGy, well below the level necessary for effective microorganism control. At dose levels of around 4 kGy, serious degradation was observed, and at 7 kGy these researchers recorded extensive oxidation and depolymerization . Other cellulosic materials, especially the fibers including cotton, bast fibers, etc., tend to be equally sensitive. Studies on cotton by a team of Scottish and Greek textile scientists indicated, for example, an exponential reduction of tensile fiber strength with dose, where this strength was reduced by ca. 50% at 100 kGy, while early work at Cornell University recorded a 27% reduction in degree of polymerization in cotton cellulose at 6 kGy. While cellulose in wood must be expected to undergo comparable changes, significant mechanical damage to wood, such as investigated in the studies on waterlogged wood from the Mary Rose shipwreck, requires quite high doses, in the order of 100 kGy. Industrial sources tend to regard damages at doses up to 10 kGy as "somewhat trivial" though they concede that color changes occur quite readily at these levels.

The other major group of natural polymers, the proteinaceous ones, tends to be less susceptible to radiation damage than the cellulosic materials. The major effects result from reactions involving individual amino acids, including deamination, and total disconnection of an individual amino acid from the polymer. There appears to be less susceptibility to chain scission, nor evidence of cross linkage, at the dose levels of the published experiments (up to about 250 kGy). Research on the effects to wool fibers, for instance, showed a loss of about 10% in tensile strength for wool fibers exposed to doses of 20 kGy of gamma radiation, while exposure to accelerated electrons only showed perceptible damage at the 50 kGy level. However, such damage cannot be overlooked when assessing its admissibility in the context of museum collection items, especially since the doses anticipated to be used by the USPS are in this same range.

Synthetic polymers are generally less vulnerable to radiation damage than their natural counterparts. The most sensitive is Teflon, which is reported to show significant effects at dose levels of 10 kGy. Canadian textile researchers at the University of Manitoba studied radiation effects on a number of nylon fibers, where they found that doses of 10 kGy resulted in about a 5% loss of tensile strength, while 15 kGy induced losses of 10-20%.

A special case is that of DNA molecules. The relatively large size of the DNA molecule results in a high probability of it being hit by one or more radiation particles. It is worth noting that the primary mode of radiation induced eradication of micro-organisms is major destruction of the DNA. Hence, irradiation at the levels intended for anthrax spore extermination will also induce major damage to DNA in research specimens. These effects will include fragmentation of the molecule and, through recombinations, formation of mutations. The mutagenic properties of ionizing radiation are, of course, well known, and result from these recombination reactions. While a significant fraction of the original DNA of the specimen irradiated at dose levels of 10-50 kGy may be preserved, the question, which arguably can only be addressed on a case-by-case basis, is to what extent the research value of the specimen is compromised, for the intended or future studies, by the large scale destruction of the specimen’s DNA, and the formation of significant quantities of mutated varieties and of major concentrations of fragmented DNA.

A class of organic molecules that is especially vulnerable to radiation induced damages is that of the dyestuffs. Complete removal of functional groups such as in the azo-dyes, and the destruction of conjugated double bonds, will result in major fading and color changes at dose levels well below those required for the damages discussed above. In fact, these radiation induced color changes in dyes have been used for purposes of radiation dosimetry. It is not unreasonable to predict that, at the dose levels anticipated to be used by the USPS, dyes in textiles, lake pigments, various ethnographic objects, and scientific specimens (e.g. microscopic slide specimens, flowers, elytra) could undergo extensive color changes or fading. The same holds for color photographs; both color slides and prints should be expected to fade and show color shifts.

In addition to these visual effects that result from destruction of dye molecules, one also must anticipate the occurrence of significant color changes, be it blanching or discolorations, in other organic materials, when chromophore sites are created or destroyed. Color changes are also to be expected in a number of inorganic materials, especially glasses and minerals/gemstones. These effects are due to the population of localized, metastable electron traps at lattice imperfections, by free electrons, resulting from ionizations. Glass can acquire a purple color, while various gemstones acquire a variety of colors. These effects can generally be mitigated by annealing the specimen; however, in the case such as a microscope slide with balsam mounted specimen, heating is not a viable option, and it is not necessarily a recommended practice for mineral specimens either.

It is worth noting that irradiation with 10 MeV electrons will not cause the formation of radioactive isotopes through nuclear reactions, since the thresholds for such reactions would require higher electron energies. In organic materials, with relatively low Z elements, the electrons from the accelerator will lose their energy through interaction with the electrons of the target material’s atoms, causing ionizations; with the nuclei their interactions are mainly constricted to scattering, with no consequences for the target material. If the accelerated electrons would hit targets consisting of elements with much higher Z values, increasing amounts of their energy would be lost in the form of "bremsstrahlung", i.e. high energy photons. High energy photons are able to interact with atomic nuclei and induce nuclear reactions; however, the bremsstrahlung from 10 MeV electrons would be below the energy threshold for such reactions.

4. Consequences of the planned electron beam irradiation of USPS mail for the mailing of museum and archival collection items

Preliminary information suggests that at this time only irradiation of mail, not of packages, is planned, and this may reduce the concerns to a small fraction of all specimen exchange. Should this situation change, and should USPS start to irradiate packages too, it may be more effective to switch from electron to gamma irradiation, presumably applying a similar size dose. For the purposes of the concerns addressed here, the effects would be largely the same, both qualitative and quantitative.

Summarizing the information as it pertains to typical collection specimens exchanged by museums and research laboratories and transported by mail, the following concerns emerge.

  • Living specimens (seeds, cuttings, etc.) will be killed by this irradiation.
  • Materials of cellulosic composition, especially plant fibers and paper, will be quite seriously affected. They will loose significant tensile strength and will become more brittle, while the induced chemical changes, chain scission and oxidation, will accelerate their aging processes. Discoloration is also to be expected. Oxidation also will result from interaction with ozone formed in air during the irradiation; while one may expect efficient ventilation at the radiation equipment, ozone also will be formed within the enclosures of the mailed materials, where the concentration could range in the tens of ppm.
  • Materials of proteinaceous composition, while less vulnerable than the cellulosic ones, still can be expected to be affected at the proposed dose levels in terms of physical changes (embrittlement of skin products, loss of fiber strength in wool and hair samples), and in terms of accelerated aging. Again, discolorations are to be expected. Again, ozone production is an additional factor.
  • Samples of interest because of their genetic information can be compromised, to an extent depending on the type of questions being addressed by the research in which they are to be used, because of large scale destruction of DNA molecules, accompanied by recombinations.
  • Dyestuffs will fade, resulting in fading and color shifts in textiles, stained specimens, and color photographs. The same effect may result in shifts and fading of the natural colors of specimens.
  • Glass can undergo blue/purple discolorations; this may affect the research value of microscopic slide specimens. While this discoloration of the glass can be removed through annealing, this would not likely be a viable option for mounted specimens because of the effects of the heating on mounting medium and the specimens themselves.
  • Mineral specimens may develop colors and/or change colors; generally these effects are reversible through annealing, though of course the effects of that heating on the specimen depend on its nature.
  • In the case of specimens under alcohol, there is the potential for some radiolysis of the preservation solution, leading to the formation of various ions and free radicals in the solution. These reactions are very complex and can lead to a wide range of reaction products, but the concentrations of the latter should be in the ppm range and do not form a major concern. Additionally, the temperature raise resulting from thermalization of the electron beam energy would raise the pressure in the container somewhat, but this effect is not likely to be of sufficient magnitude to cause failures of the containers unless the integrity of the latter were already seriously compromised.
  • Rubber and plastic stoppers of bottles and vials may become somewhat embrittled, but not to an extent of losing the closure of the containers.
  • Magnetic media (floppy disks, zip disks, audio and video tape) will probably loose significant information content. Undeveloped photographic film will be exposed.
  • Radiocarbon dates of irradiated samples will not be affected in a significant way, although there is a theoretical possibility for contamination as a result of chemical reactions that involve reactive groups from carbon containing packaging material.
  • Samples intended for thermoluminescence dating will become useless, since this irradiation will deposit a dose that exceeds the "natural" one by orders of magnitude.
  • Since no nuclear reactions are induced under the proposed conditions, generation of radioactivity in the irradiated samples is not a concern.
  • It is not practical to try to mitigate the radiation effects through shielding of the samples, e.g. with lead metal. The weight of the shielding required to stop these high energy electrons would be quite high and make the mailing expensive; moreover, the bremsstrahlung generated by interaction of the electrons with the high Z elements of the shielding could still result in appreciable doses to be administered to the material inside. USPS also might have objections, not only since it presents an attempt to circumvent their preventive actions, but also since this bremsstrahlung could conceivably create other problems at the irradiation facility.

In view of the above it is strongly suggested that mailing through USPS of vulnerable specimens and collection items, as well as important research information on magnetic media or undeveloped film, be avoided unless it can be arranged for these mailings to be exempted from irradiation.

This information has been prepared by the Smithsonian's Museum Conservation Institute (MCI) for your information and as a service to the professional community.
November 5, 2001

Recent Examination of Some Irradiated Mail

USPS Security of the Mail

For information, contact:
Ann N’Gadi
Technical Information Officer
tel. (301) 238-1240 option 2
fax (301) 238-3709