Accelerator Section Logo

Invited Paper: The Implications of ICRP Publication 92 for Neutron Dosimetry

Ralph H. Thomas

(To be presented at the Health Physics Society Accelerator Section Meeting, Washington, D.C., July 13, 2004.)


Three years ago the International Commission on Radiological Protection (ICRP) announced a review of its current recommendations published as Publication 60, with a view to issuing a revision next year (ICRP 1991, 2001).

Since 1991 the scientific literature has revealed concerns with some aspects of ICRP Publication 60, particularly by dosimetrists interested in the measurement of high-energy and high-linear-energy-transfer (LET) radiations in general and in the measurement of neutrons in particular (for bibliographies see ICRP 1997; ICRU 1998; Thomas 2001, 2003). Recently, Roger Clarke, chairman of ICRP, has agreed that "there have been some persistent difficulties with, and misunderstandings of, the definitions of the Commission's dosimetric quantities. The Commission will remove these by clarifying its definitions and specifying their application" (Clarke 2003).

Toward the end of 2003 as part of its review, ICRP issued Publication 92 entitled "Relative Biological Effectiveness (RBE), Quality Factor (Q), and Radiation Weighting Factor (wR)" (ICRP 2003). This is clearly an important document and will be extremely influential in formulating the commission's final recommendations. Indeed, it may be a foreshadowing of those recommendations.

The writer well understands that the logical formulation of radiation protection quantities is scarcely the most stimulating topic for already overworked operational accelerator health physicists to add to their schedules. Nevertheless, the current ICRP discussions will have important consequences for the workplace and the environment of high-energy accelerator facilities. It would be a good investment in time for members of the accelerator community to understand the impact of the issues raised in Publication 92 and the solutions proposed. It is urged that those interested in neutron dosimetry should read this important report. Space does not permit an extensive review of the entire document here and only some brief comments are possible. Emphasis is placed on the important issue of radiation weighting. At a later date it is hoped that a detailed, paragraph-by-paragraph commentary on much of the report can be made available to the accelerator community by the author.

The Contents of ICRP Publication 92

The report is long: some 107 pages of text of which about 70% is taken up with a summary of the basic radiobiology and the difficulties met with its interpretation, so that guidance can be provided that will facilitate the development of a radiation-weighting system for high-LET radiations. Some 35 pages are likely to be of direct interest to high-energy dosimetrists, of which a dozen pages, mainly in Section 4 entitled "Weighting Factors for Radiation Quality," present the radiation-weighting issue as seen from the perspective of ICRP.

The report provides a good, if somewhat disjointed, summary of the problems and issues and appears to validate many of the criticisms of wR that have appeared in the scientific literature. For example, the report acknowledges that there are "uncertainties in ascribing appropriate wR values for neutrons of high energy (>20 MeV)" but suggests "this is a significant issue primarily [sic] at high altitudes. We anticipate that the current ICRP Committee 4 Task Group on Radiological Protection in Space Flight will advise on these matters" (the emphases are the author's). The author's opinion as to the success or failure of the document in resolving issues is given below in the section entitled "Analysis."

Significant proposals and recommendations of interest to high-energy dosimetrists made in the report include the following:


Publication 92 has much to commend it. The report accepts that change is necessary in the radiation-weighting system and shows itself willing to make some changes, provided they represent "minimal departures from the present system." This analysis must be brief and thus, alas, dwells on the faults of the report.

High-Energy Radiations. A major deficiency of Publication 92 is that there is no significant discussion of the dosimetric issues for particle accelerator radiation environments. The increasing exposure to mixed radiation fields, often not in radiation equilibrium, from particle accelerators with thin shields as well as from commercial air travel presents novel dosimetric problems created by Publication 60 that require attention in ICRP's revised recommendations.

It might be of assistance to ICRP in forming its final recommendations if it were made aware of the needs of, and the body of relevant information largely developed by, the accelerator community.

The wR(En) Relationship. If the primary radiation-weighting generator (model), the Q(L)-L relationship, is "correct," in the sense that it represents best current judgment, then the laws of physics and mathematical logic will inevitably generate "correct" values of mean Q, qE, or wR under the irradiation conditions specified. It would then be expected that, if defined in an anthropomorphic phantom, that qE should be identical to wR:

Indeed, calculations for low and intermediate neutron energies show them to have values of qE near 2, which are judged to be consistent with the relevant relative biological effectiveness (RBE), and consequently appropriate reductions are recommended from the Publication 60 value of 5. However, Publication 92 would describe eq. (3) as a "radical simplification" because near 1 MeV the calculated value of qE is about 13, considerably lower than the values of Q*(10) and of wR approximately equal to 20 recommended in Publication 60. Apparently, this factor of approximately 1.6 at this energy is thought to be unacceptable to ICRP and the additional constraint

was introduced to develop eq. (2).

This additional constraint is justified by the argument that "radical simplification of wR seems impracticable for the reason that it would tend to force tightening of the dose limits in general. If the current value of 20 for fission neutrons were reduced to 10, this would decrease the numerical value of the effective dose from exposure to fission neutrons by a factor of 2. This would amount to a relaxation of limits for neutron exposure, which may meet strong objections and would almost certainly generate pressure to offset the change by a decrease of the effective dose limits, which would then apply to all radiations including photons." This argument seems to be clearly inconsistent with the actions recommended at lower neutron energies. One wonders why a proposed reduction in wR by a factor of 2 for 20-keV neutrons is acceptable but is so egregious for fission neutrons.

If ICRP is to avoid ambiguity, two determiners of radiation weighting cannot be permitted. If ICRP selects the presently recommended Q(L)-L relationship to determine radiation-weighting factors, a value of qE and wR approximately equal to 13 at 1 MeV must be accepted. (We can do arithmetic and physics quite well these days.) Conversely, if ICRP wishes a value of qE = 20 at 1 MeV, it must modify its currently recommended (Publication 60) Q(L)-L relationship. The degree to which the radiobiology permits this to be done is a matter for the judgment of ICRP. It may be significant that Publication 92 makes no such recommendation.

Before reaching firm conclusions it might be prudent to extend the work reported in Publication 92 by studying the dependence of qE on phantom specification. Sufficient databases probably already exist for such a study to be carried out quite easily.

Both refined understanding and additional calculations may help to resolve these issues. It is possible that the two influences of dose distribution and organ-weighting averaging were not fully understood at the time of their introduction (ICRP 1977, 1980). However the value of qE at 1 MeV is no great surprise having been foreshadowed in ICRP Publication 51 (ICRP 1987).

The Relevance of wR at High Energies. In Publication 92 a major effort is directed toward the development of an average radiation-weighting factor applicable to the whole body. Even if it is accepted that Publication 92 achieves this goal, the practical utility of this factor at high energies is nevertheless not clear. The author suggests that greater emphasis needs to be placed on an acceptable definition of the Q(L)-L relationship, from which values of not only wR but also, and perhaps more importantly, other parameters that facilitate the determination of effective dose may be derived.

Although some 25 years ago average radiation-weighting factors were of great value to the "critical organ-MADE" system of protection for neutrons, they are of much less interest at high energies or in more sophisticated detriment-models, such as effective dose equivalent, HwE, and effective dose, E, which specify tissues and organs that facilitate the use of anthropomorphic phantoms. At high energies, and particularly at accelerator laboratories, there is more interest in using conversion coefficients that relate field quantities (e.g., fluence) to determine the radiological protection quantities.

Duality and Operational Quantities. There is general agreement that protection quantities such as effective dose equivalent, HwE, and effective dose, E, may be determined, though some would not agree that they may be measured. When a quantity may be determined it may be sufficient to leave it to the inventiveness of the dosimetrists to discover the precise means with which to do so.

It would be helpful if the ICRP could be persuaded to avoid further controversy over the dual system of radiation protection quantities by softening its apparent imprimatur of the ambient dose equivalent. Although the ambient dose equivalent is of great utility in some aspects of dosimetry, there are significant problems with its application to neutron dosimetry, particularly at high energies (Ferrari 1998, 2004; ICRP 1997; ICRU 1998; Pelliccioni 2004). It may not be wise to endorse ambient dose equivalent to the exclusion of alternatives.

The ICRP might wish to consider ambient dose equivalent to be one weapon of an armorarium that consists of many alternatives, any of which may be used as an operational quantity. Other techniques might, for example, include LET-spectrometry (paragraph 278) and neutron and charged-particle spectrometry (Thomas 2003).

Of course, the simplest option would be to abandon the dual concept of protection and operational quantities altogether and define only protection quantities.

Universality. It is disappointing that the authors of the guest editorial seem to dismiss any possibility of a universal dosimetric scheme that encompasses both internal and external exposure by the statement that "we believe that ICRP should continue the use of wR values that relate, for external radiation, to the incident field. For radionuclide intakes, wR values should relate to the internal fields that cause the absorbed dose to organs and tissues [sic]." Here is an undesirable and unnecessary source of ambiguity.


In retrospect it becomes clear that the calculations described in ICRP 75/ICRU 57 should have preceded the publication of, and been used as a guide to shape, the recommendations of ICRP Publication 60. Had this been done perhaps many of the inconsistencies discussed in ICRP Publication 92 might have been avoided. It is to be hoped that the mistakes of the past will not be repeated in 2005.

There are probably now sufficient data available to enable wise judgments to be made that will facilitate the development of a logical and coherent system of radiation-weighting dosimetry that will be of utility at all energies.

A shift in thinking might be helpful. The telescope should be reversed. Approximations to the simple can only be made if the complex is understood. Attempts to move in the opposite direction can often produce Rube Goldberg (translation to English: Heath-Robinson) contraptions. If a dosimetric system is devised that adequately serves the needs of high-energy, high-LET radiations, then the approximations that may be made to satisfy the simpler demands of low-energy, low-LET radiations will be facile.

Time is surely of the essence. There will inevitably be great pressure on ICRP to complete its revisions as quickly as possible. Nevertheless it is more important to be correct and honest in one's science than to hurry to a timetable - whatever the external pressure. Radiation protection cannot continue to change its vocabulary every decade or so. Any changes to be recommended by ICRP in 2005 must be based on an adequate database to permit wise decisions and the author believes that sufficient information is available to resolve the issues raised in this brief review.

The scientific input to the standards-setting process is necessarily a delicate balance between the ivory tower of academia and the practical concerns of the workplace. Many, including this writer, believe that it is the latter that should lead in the development of radiation protection recommendations and standards policy.


Clarke, R.H. "Radiation Protection for the 21st Century," Radiation Protection Dosimetry 105, 25-28 (2003).

Ferrari, A., and M. Pelliccioni. "Fluence to Effective Dose Conversion Coefficients for Neutrons up to 10 TeV," Radiation Protection Dosimetry 76(4), 215-224 (1998).

Ferrari, A., and M. Pelliccioni. "Fluence to Effective Dose Conversion Data and Effective Quality Factors for High Energy Neutrons," Radiation Protection Dosimetry 76(4), 215-224 (1998).

Ferrari, A., M. Pelliccioni, and R. Villari. "Evaluation of the Influence of Aircraft Shielding on the Aircrew Exposure through an Aircraft Mathematical Model," Radiation Protection Dosimetry 108, 91-105 (2004).

International Commission on Radiological Protection. Recommendations of the International Commission on Radiological Protection, Publication 26. Annals of the ICRP, 1(3). Oxford: Pergamon Press (1977, reprinted with revisions, 1981).

International Commission on Radiological Protection. Statement and Recommendations of the International Commission on Radiological Protection from its 1980 Stockholm Meeting, Publication 30, Part 2. Annals of the ICRP, 4(3/4). Oxford: Pergamon Press (1980).

International Commission on Radiological Protection. Data for Protection against Ionizing Radiation from External Sources, Publication 51. Annals of the ICRP, 17(2/3). Oxford: Pergamon Press (1987).

International Commission on Radiological Protection. Recommendations of the International Commission on Radiological Protection, Publication 60. Annals of the ICRP 21(1-3). Oxford: Pergamon Press (1991).

International Commission on Radiological Protection. Conversion Coefficients for Use in Radiological Protection against External Radiation, Publication 74. Annals of the ICRP, 26 (3/4). Oxford: Pergamon Press (1997).

International Commission on Radiological Protection. "A Report on Progress towards New Recommendations: A Communication from the International Commission on Radiological Protection," Journal of Radiological Protection 21, 113-123 (2001).

International Commission on Radiological Protection. Relative Biological Effectiveness (RBE), Quality Factor (Q), and Radiation Weighting Factor (wR), Publication 92. Annals of the ICRP 33(4). Oxford: Permagon Press (2003).

International Commission on Radiation Units and Measurements. Conversion Coefficients for Use in Radiological Protection against External Radiation, Report 57. Bethesda, Maryland: ICRU (1998).

Pelliccioni, M. "The Impact of ICRP Publication 92 on the Conversion Coefficients in Use for Cosmic Ray Dosimetry," Radiation Protection Dosimetry (in print, 2004).

Thomas, R.H. "The Impact of ICRP/ICRU Quantities on High-Energy Neutron Dosimetry - A Review," Radiation Protection Dosimetry 96, 407-422 (2001).

Thomas, R.H. "Morgan Lecture: Accelerator Radiological Protection-A Personal and Privileged Odyssey," available at Health Physics Society website (2003). See also "Accelerator Radiological Protection - A Personal and Privileged Odyssey," parts 1-3, NVS Nieuws (Dutch Journal for Radiological Protection) (in print, 2004).