At the HPS Midyear Conference in San Jose I noticed that many American colleagues are not only utterly confused but in addition are disgusted by the way ICRP and ICRU continuously seem to destabilize radiological quantities. Having followed the development in this field over the last 25 years I went through various stages of depression and confusion myself - which in fact rather disqualifies me to write this article.
In order to split up the great confusion into chewable bites let us put some order in the discussion and quickly do away with radiological units. They are really not an important issue. Many colleagues still consider that the rem as the unit of dose equivalent and the curie for activity were more convenient than the SI units and cling to the former, but let's face it: the world has gone SI.
In a time of decreasing doses and dose limits the unit sievert is actually not so bad as a dose rate of 10 µSv/h in a controlled radiation area is about the level where you as an operational physicist like to think about, and an effective dose of 1 mSv is roughly what we all receive from external natural radiation - excluding radon - in a year.
Personally, I prefer the becquerel very much to the curie (thinking of decays per second as something everybody can easily grasp) than relating the physical quantity activity to that of one gram of radium (we had milligrams of that stuff at CERN and are happy that we got rid of it). What was the limit still for milk in those days of Chernobyl: 800 Bq of Cs-137 in a liter or roughly 1 kBq in a quarter (sic!) so the becquerel is a rather convenient measure for the environment and food stuff.
Switching from units to physical quantities I shall only consider external exposure. The sievert is actually the unit of various physical quantities used in the radiation protection of external radiation. Ambient dose equivalent, personal dose equivalent and effective dose - the latter is not to be confused with the nowadays "obsolete" quantity effective dose equivalent - are the most important ones. The first two, H* and Hp, are operational or secondary quantities whilst the last two, E and Heff, are protection or primary quantities sometimes also called risk quantities. These are actually the quantities in which ICRP recommends dose limits and on which legal limits are based.
Let us discuss operational quantities first. Why do we need them? They are necessary as the protection quantities are by definition not measurable. In fact, the effective dose that builds up in my body moving in an external radiation field is pp (personal and private) and quite distinct from the one that would be present in somebody else's body should he/she move around in the same field. So obviously we need quantities in which our instruments and personal dosimeters are calibrated such that we can measure dose equivalents in any radiation situation and in a defined way. It is clear that the numerical values thus determined should be a good estimator for effective dose that would have been received by people of various sizes and weights moving around in the external radiation field.
The concept of having only one physical quantity for all kinds of various radiation types as the estimator for effective dose is appealing. Hey, wait a moment: there is not one but there are actually two operational quantities in radiation protection: ambient and personal dose equivalent.
Unfortunately, the concept of one or two universal operational quantities in radiation protection went through several iterations that were painfully carved out between 1980 to 1996 in at least four international conferences with the participation of ICRP and ICRU representatives, sometimes just ending up in cul-de-sacs like the index quantities. Historically, ambient dose equivalent for area monitoring, and defined in the ICRU sphere, was the first useful quantity but then ICRU felt the need for an additional quantity to be used in individual monitoring. Their original definition of personal dose equivalent is somewhat weird: "The personal dose equivalent is the dose equivalent in soft tissue below a specified point on the body at an appropriate depth. The calibration of the dosimeter is generally performed under simplified condition (how much?) on an appropriate phantom." Surely the obvious idea for an appropriate phantom was the ICRU sphere itself but that would mean that a personal dosimeter must show an isotropic response. This is surely not the case for a single detector worn on the front of the body hence shielded from the back. Well, ICRU finally surrogated the person by a tissue equivalent slab in which personal dose equivalent is calculated for various external radiation field conditions (energy and angle of incidence) relating the quantity to primary field quantities through conversion factors. For practical calibrations the conditions of the TE slab have been further simplified by introducing the ISO Plexiglas/water phantom. Personal dosimeters shall have a defined angular response which, for strongly penetrating radiation up to angles of 60 degrees, however is not very pronounced.
Although there are subtle differences between the conversion factors for the ICRU sphere and the tissue equivalent slab phantom (already due to their different mass), all in all ambient dose equivalent and personal dose equivalent are both used as estimators for effective dose from external exposure. In fact, if you enter numerical values of either ambient or personal dose equivalent into your dose records and you can show that those values are well below any legal limits you have done a good job (the cream on the coffee is of course that the doses received had been ALARA).
The long and painful birth of the operational quantities in radiation protection appalled many colleagues and some went so far to claim that our trade stepped backward not forward. Would it not be much better to use primary radiation field quantities that could be directly related to (an overestimation of) effective dose? This we had already and here we would just retrograde. Remember MADE, the word which to a person of German mother tongue is not really appetizing? Well, at CERN we used the MAximum Dose Equivalent a lot and somehow our rem (sic!) counters are still calibrated in MADE but as it has been pointed out several times the dose value was rather a maximum than a good estimator and in addition MADE is not an additive quantity. Anyway the basic idea that a dose can only be defined in a phantom was there in the field of neutron dosimetry and the photon people surely learned from the neutron dosimetrists.
What we went through with operational quantities is nothing compared to the tale on protection quantities. Ralph Thomas wrote an article in the early 70s: "What dose equivalent?" Rightly so, as at that time a distinction between operational and protection quantities was not made.
Enter ICRP Publication 74:
Let's see what they have to say: We find Physical quantities (fluence, kerma, absorbed dose possibly in tissue), Operational quantities (ambient dose equivalent, directional dose equivalent and personal dose equivalent) and Protection quantities (Effective dose, organ equivalent dose and organ absorbed dose). All is nicely explained and there are conversion factors too. For those of you who like history books ICRP 74 is fascinating reading; however, beware of many trivial and a few less trivial printing errors like the legend in figure 3. As a Freudian slip we read on the ordinate: Quality of Radiation Weighting Factor. Let us forgive the proof readers as "the quality of mercy is not strained" and hope that ICRU in their parallel publication will do a better printing job.
Number 5 is the best chapter with a critical but reasonable discussion on the estimator properties of the operational quantities with respect to the protection quantities. All the problems of this world arise from the fact that ICRP in their wisdom replaced effective dose equivalent (based like ambient dose equivalent on the Q-LET relationship) by the completely different concept of effective dose based on radiation weighting factors. Here the gurus went as far back as ICRP 4 where dose equivalent could be determined by multiplying tissue kerma in air with the quality factor (now weighting factor) of the external radiation field. This in spite of the fact that for intermediate neutrons the quality factor in the body is greatly reduced because of the many accompanying photons.
So there are now radiation situations where ambient or personal dose equivalent are no longer over-estimators for effective dose but these are luckily only a few. With respect to this issue Graham Stevenson immediately reacted as far as the shielding provisions for the LHC are concerned. I recommend that you read: "G. R. Stevenson, The Implications of ICRP Publication 74 for the Design of the LHC shielding, CERN/TIS-RP/97-15, 1997."
Exeunt ICRP Publication 74.
It is quite obvious. ICRP Publication 74 will not be the final chapter of the saga on radiological quantities.
Remember: "Those who don't understand anything else than dosimetry don't even grasp dosimetry properly."
Aphorism copied and slightly modified from the famous messy notebooks of Georg Christoph Lichtenberg (1742-1799), Professor of Physics in Goettingen.
[Back to top]