2038: here we come

It depends how long after it’s removed from the reactor

Nearly all of nuclear waste does not come from inside reactors.

Nearly all of nuclear waste does not come from inside reactors.

Um. Even hospital nuclear waste sometimes comes from inside reactors, although it may have been created in accelerators instead. But the quantity of radioactive isotopes produced in accelerators is a minuscule fraction of the total. There’s a fair amount of radioactivity in tailings from uranium mines, but that’s not usually described as nuclear waste. There’s some produced by neutron activation in the ancilliary equipment around nuclear reactors, but the rest, the vast bulk, comes from inside reactors. It may then pass on to a reprocessing facility, and you could say that’s where the nuclear waste comes from, but it came from the reactor before that.

By the time it leaves the reprocessing facility it’s quite a while since it left the reactor (it will have been in cooling ponds for weeks or months before going for reprocessing) so most of will indeed be non-radioactive by then. But what’s left…

By bulk, that hospital waste is big – but very low activity compared with the reactor waste, and while bulky, much less so than the reactor waste.

Here you go – here’s the fission products yields: http://web.archive.org/web/20150206140041/http://ie.lbl.gov:80/fission.html

(This link is on my “no valid scientific references” website, of course. Always was.)

Thermal neutron induced fission of U235 is the one you’re mostly interested in, but some of the rest are of significance, too – if anyone’s interested I can explain exactly which and how.

And remember, unlike plutonium (and other actinides produced in reactors) fission products cannot be destroyed in fast breeder reactors – in fact they produce more of them.

An apology. I have equated nuclear waste with radioactive waste, which is quite common; all radioactivity comes from the nucleus. However this muddies the water when discussing nuclear power. ;-(

Even hospital nuclear waste sometimes comes from inside reactors, although it may have been created in accelerators instead.

Even you have made the same equation here. ;-)

Oh, nuclear waste is usually regarded as including radioactive hospital waste – a substantial bulk, but in terms of radioactivity not in the same league as used fuel or the fission products separated from used fuel during reprocessing. And in particular, relatively short lived compared with a significant proportion of the fission products.

Some of the radioactive material in hospital waste is isotopes that are present in significant quantities in fission products, but separating them is too difficult and dangerous, and that’s not where they get them. Some of them are produced in reactors, by neutron activation of non-radioactive materials inserted into the neutron flux in the reactor, but of course kept out of contact with the fuel with all the associated fission products; and some are produced in accelerators by bombardment of targets with ions, or possibly with neutrons produced by the bombardment of a primary target.

so we have until the year 2176 (approximately)

More accurately:

   10REM >TimeTest
   20:
   30REM Unix &7FFFFFFFFF
   40DIM tem% 40
   50!tem%=&FFFFFFFF
   60tem%!4=&7F
   70SYS "Territory_SetTime",tem%
   80*TIME
   90!tem%=&FFFFFFFF
  100tem%!4=&FE
  110SYS "Territory_SetTime",tem%
  120*TIME
  130END
>RUN
Sun,18 Mar 2074.03:28:58
Sat,23 Jan 2247.04:30:04

So Unix has until 2074 and RISC OS has until 2247. QED

How do you work that out, given that traditional Unix¹ uses a signed 32 bit integer counting seconds since 1970, CLib uses an unsigned 32 bit integer counting seconds since 1970, and RISC OS uses an unsigned five byte value counting centiseconds since 1900…

¹ I believe Linux etc now use 64 bit values.

How do you work that out,

Someone on this thread said Unix uses a 5 byte signed integer presumably counting centiseconds since 1900.

If only it were possible to have a rolling float to store timestamps. The longer ago the file was the less the accuracy matters. But obviously that can’t work.

Someone on this thread said Unix uses a 5 byte signed integer presumably counting centiseconds since 1900.

It doesn’t. As Rick says, it’s either a 32-bit or – on 64-bit systems – a 64-bit integer counting seconds from 1970.

Understood. Thanks.