Re: Gy and Sv = J/kg

[Jack Couch <jgcouc@planetx.bloomu.edu>]

At 03:48 AM 10/5/96 -0500, Jim Muckerheide wrote:

>Jack Couch missed the point:
 <snip>
>
>Jack Couch had written: 
>> There are other examples, such as Circumference = 2 pi Radius.<snip>  
>
Jim wrote:
>This "conversion" doesn't change the _meter_! It does not say 6.28=1. Totally
>unlike saying 20=1 for Gy-Sv "conversion" (as opposed to an actual change in a 
>total value of 20 vs 1 as measuring different quantities). R and C are
>different vectors, they have a different number of meters. They are not
>converting meters to equivalent-meters!  
>
My intent was to agree that when using SI units in connection with H and 
D the numbers of J/kg's don't necessarily equate. Perhaps I could have 
chosen a more suitable parallel and been more clear.

F. Attix said it much better: "Thus where the dimensionless quality factor 
Q has a value of, say, 2, the dose equivalent in tissue would be 2 Sv (or 2
J/kg). 
This does not mean that 2 J/kg = 1 J/kg, any more than saying that if a room 
has a length of 5 m and a width of 3 meters, then 5 m must equal 3 m." 
[Health Physics Vol. 41, pp. 210-213, 1981]

T. R. Crites, in his lament concerning the dose equivalent unit, says: "Deposit 
1 J/kg of 2.5 MeV neutrons for a dose of 1 [Gy] in a body and you'll have 
something like 9 [Sv] of dose equivalent, but you really don't have 9 J/kg 
of energy deposited anywhere." [Health Physics Vol. 38, pp. 431-432, 1980]

>> David Golnick* offers a nice solution to those (including my students) 
>> who agonize over this point. He endows Q with the "units" Sv/Gy...This 
>> of course breaks both with tradition and the formal definition of Q, 
>> where Q carries no units.
>
>This is good; its true. But it still doesn't provide a "solution" to J/kg, nor 
>help the student figure it out. 
>
Yes, it hides the issue. And at some point everyone, including 
students, must be let in on our sticky little secret. Why should we be the 
only ones lying awake at night agonizing over it. :-)

Apparently, the idea of assigning Sv/Gy to Q has been around awhile. The 
earliest citation I can find is by J. Sabol, et al. in 1977 (I can provide a 
reference if anyone is interested.) 

>Louis had it right.  The conversion factor _changes the units_!   The  wrong.
>With Gy in J/kg, Sv is in "units" of "the biological damage equivalent to the
>damage of J/kg of low LET radiation" (J-eq-damage/kg?)   
>
This "conversion factor" doesn't convert units. Its dimensionless. 
The J/kg remains a J/kg. The problem can't be solved by adding camouflage 
to the SI units of H. I do agree that H is not D, and Q is a dimensionless 
weighting factor with an assigned value based on comparative bio-damage.
>
As most list members know, this saga (Ralph Thomas called it a soap 
opera) over dose equivalent units has been ongoing since its inception 
by Cantril and Parker in 1945. 

One of the clearest, most sensible statements I've read concerning the 
Sv is, again, by F. Attix in his letter to the editor.[Health Physics 
Vol. 46, p. 479, 1984] A couple of ideas included in his letter are well 
worth passing along.

"…there is nothing wrong with the present definition…"  "The same units 
(e.g. J/kg) certainly can be applied to different co-existing quantities, 
such as [D] and [H] without being illogical." (Attix explains his position 
with reference to ICRU Report 33.)

Paraphrasing Attix: An alternate, equally valid definition [other than 
leaving Q unitless and defining H with SI (J/kg) units as we have it today] 
could have been reached by ICRU at the outset by acknowledging that 
H *is not a physical quantity*. Therefore it need not be measured in terms 
of an SI-derived unit. H could have been defined solely by H = QD in 
which D is a physical quantity but Q is a coefficient based on biological 
response information. 

He continues by saying that it would have been perfectly reasonable 
(had the ICRU chosen to do so) to define the sievert as the (one and 
only) unit of dose equivalent, that the Sv would not be equated to 
1 J/kg, and it would follow that Q would have units of Sv/Gy. 

Attix then states what perhaps many of us feel.  "I think this alternative 
approach, if adopted initially, would have been more readily accepted 
by the health physics community. However changing to it now probably 
would not be worth the extra confusion attending any such change."

Jack Couch
Bloomsburg University