# Level of uranium enrichment

## Main Question or Discussion Point

Forgive me if this question has been asked before or is a FAQ.

Can somebody tell me to what percentage uranium has to be enriched in order to build a bomb? Believe me, I'm not trying to build one!! I'm just trying to inject some physics into the current Israel/Iran debate. (Let's confine the discussion to physics, not politics!)

Also, what percentage enrichment is necessary for commercial reactors, and is that a percentage of U-235 or U-238?

Thanks.

Related High Energy, Nuclear, Particle Physics News on Phys.org
For weapons, about 90%. For commercial reactors, about 4%. As far as I understand the tough bit is going up to ~20%, from 20% to 90% it is supposed to be easier. But I'm far from an expert. Look at http://en.wikipedia.org/wiki/Enriched_uranium

>what percentage enrichment is necessary for commercial reactors.

That expression is a bit ambiguous. The canadian use natural uranium with no enrichment by dint of using heavy water (dueterium oxide). Most commercial reactors in the world including the Iranian's use natural water and uranium enriched to 3-4%.

mfb
Mentor
and is that a percentage of U-235 or U-238?
Those values are always the fraction of U-235.

kloptrok said:
As far as I understand the tough bit is going up to ~20%, from 20% to 90% it is supposed to be easier.
Easier in which way? 20% -> 90% needs less additional enrichment than 0.7% -> 20%; but if it is just that, the statement would be similar to "the tough part of the uniform 100km-track are the first 80km, the remaining 20km are easier".

QuantumPion
Gold Member
The difficulty in enriching depends on the relative concentrations. It is difficult to enrich the first ~10% or the last ~10% but it's relatively easier to go from 20% to 80%.

Another common approach is use graphite moderator. No isotope separation involved at all.

I'm wondering whether this is a well-posed problem. Suppose it takes a given amount of time, T, to enrich natural uranium (0.71% U-235) into a fixed amount of reactor-grade uranium (4.5%). Using the same enrichment facilities, how much time would it take to further enrich the reactor-grade uranium into highly-enriched uranium (20%), and then to enrich it even further to weapons-grade (90%)?

A second question would be what commercial (non-military) uses (if any) are there for highly-enriched uranium (20%)?

According to a news report, Iran has offered to convert any highly-enriched uranium it may produce in the future into a powdered form, which it says cannot be further enriched. Is this technically feasible?

Once again, let's just talk about the physics, leaving policy and politics out of it.

Thank you.

QuantumPion
Gold Member

I'm wondering whether this is a well-posed problem. Suppose it takes a given amount of time, T, to enrich natural uranium (0.71% U-235) into a fixed amount of reactor-grade uranium (4.5%). Using the same enrichment facilities, how much time would it take to further enrich the reactor-grade uranium into highly-enriched uranium (20%), and then to enrich it even further to weapons-grade (90%)?

A second question would be what commercial (non-military) uses (if any) are there for highly-enriched uranium (20%)?

According to a news report, Iran has offered to convert any highly-enriched uranium it may produce in the future into a powdered form, which it says cannot be further enriched. Is this technically feasible?

Once again, let's just talk about the physics, leaving policy and politics out of it.

Thank you.
Fission chamber neutron detectors use high enriched uranium. But only a minute quantity, and they last a while. But there would be no reason to build an entire enrichment plant, under enormous international sanctions, just to manufacture neutron detectors which are available on the commercial market. The other potential use is in research reactors, the purpose of which would be for breeding weapons-grade plutonium. So even if Iran's enrichment operation were blown up, if they had produced enough medium-enriched Uranium to get started, they could continue to make weapons-grade plutonium in a small hidden facility or underground.

mfb
Mentor
I'm wondering whether this is a well-posed problem. Suppose it takes a given amount of time, T, to enrich natural uranium (0.71% U-235) into a fixed amount of reactor-grade uranium (4.5%). Using the same enrichment facilities, how much time would it take to further enrich the reactor-grade uranium into highly-enriched uranium (20%), and then to enrich it even further to weapons-grade (90%)?
Well, you have to keep track of the quantity.

Assuming perfect separation (which is not possible), you need 6 tons of natural uranium to get 1 ton of 4.5%-enriched uranium, but 28 tons to get 1 ton of 20%-enriched uranium (with ~4.4 tons of 4.5% as intermediate step). In other words, most of your material processing happens at low enrichment levels.

With realistic setups, separation is not perfect. Depleted uranium is removed with ~0.3% U-235 (number from Wikipedia), so about half of the U-235 is wasted. Therefore, more natural uranium is required (compared to the amount of enriched uranium), this gives even more steps with low enrichment levels.

I would expect that 5 tons 1% -> 1 ton 4% (or whatever) and 5 tons 4% => 1 ton 16% require a similar time. But you have to do the first step 5 times.

A second question would be what commercial (non-military) uses (if any) are there for highly-enriched uranium (20%)?
Apart from some scientific experiments... mix it with natural uranium and use it in power plants?

According to a news report, Iran has offered to convert any highly-enriched uranium it may produce in the future into a powdered form, which it says cannot be further enriched. Is this technically feasible?
I doubt that there is any chemical way to prevent further enrichment . You need UF6 for enrichment, but that has to be created independent of the uranium source.