Energy From Thorium Discussion Forum

It is currently Nov 22, 2017 5:49 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 63 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next
Author Message
PostPosted: Jul 09, 2015 3:29 am 
Offline

Joined: Apr 19, 2008 1:06 am
Posts: 2248
Quote:
If we assume there has not been a catastrophic steam generator LOCA (and thus the main gas circuit is intact) then we could use a battery pack inside the vault to run the circulators at low power for the first five minutes after shutdown [80% of the excess thermal energy is added in the first five minutes].

In place of battery pack, use a thermo-electric generator powered by decay heat of fuel itself. It will be effective as long as it is necessary.


Top
 Profile  
 
PostPosted: Jul 09, 2015 8:26 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1497
That would require far more heat than we have available as thermo-electric generators are horrendously inefficient.


Top
 Profile  
 
PostPosted: Jul 09, 2015 8:54 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5058
It actually takes more like 50 minutes than 5, to get down to 1.5% decay heat. Plenty of time for the fuel to overheat if you lose the active systems at T=0. (station blackout and such).

NaOH solutions apparently scrub CO2 rather well. Steam from SG LOCA would also be scrubbed, as would any tritium oxide. So, a pressure suppression type containment should work well with this reactor (recall there is little hydrogen to worry about, unlike BWRs).


Top
 Profile  
 
PostPosted: Jul 09, 2015 9:01 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1497
I've struggled to find a proper fitted curve for decay heat after a reactor shutdown.
The only one I could find suggested the ~5 minute figure.
At 50 minutes this curve has a value of 1.1% of pre shutdown load.
It has 1.5% at roughly 12 minutes, however by 5 minutes it is down to ~1.8/1.9% and only a small amount of the excess energy is added after that time. We could run the circulators slower for longer obviously if that was required. I just wanted the active phase short to prevent any further equipment failure from undermining it.

I don't want to fill the vault with sodium hydroxide solution as that would cause problems for equipment submerged in it and indeed for maintenance. (Corrosion and the like). It would prohibit any use of aluminium anywhere in the vault, which might be a problem given its cost and thermal conductivity qualities.
It would also prevent simply dumping water when pumping the level down for refueling.

I think I can keep circulators at low power for a few minutes through a station blackout if I put the batteries and the motor drive controllers in the vault - so the any external factors that damage them would have to shatter the containment to do so.


Top
 Profile  
 
PostPosted: Jul 09, 2015 2:43 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5058
There's no need for the vault to be filled with NaOH solution. You'd have a BWR architecture, with the vault being the "drywell" (I realize its not dry) and a pipe connecting to a "wetwell" area for pressure suppression. That area would not be well outside the core area, could be anywhere you want. Could be as simple as a scrubber tank. CANDUs often use vacuum vaults, connecting the rupture disc of the calandria with an external vacuum building, you could have a similar concept but without the need for vacuum.

Aluminium is not that great an engineering material. Stainless steels, especially the austenitics and duplex grades, have superior engineering properties (strength, ductility, toughness, fatigue resistance...). Its hard to beat the austentics in their balance of materials and cost effectiveness properties, in any high temperature or cryogenic area, in fact... and duplex grades are pretty much unbeatable in the medium temperature service regimes. Thermal conductivity is not an important engineering property, in reality, unless for very special apps such as cooling fins. As an interesting case, copper condensers are routinely being replaced by stainless or titanium condensers, with nil thermal performance penalty... turns out things like fouling are more important.

Batteries aren't passive, so you have to consider what happens if you have a super SBO - loss of DC and AC power. Recent BWR offerings (ESBWR, Kerena) can deal with this scenario rather easily, so you'd have to offer something comparable. Keep in mind also that circulators are no good if you have a LOCA of your primary.

I would also be interested to see what kind of channel power you can get. It would have to be close to a CANDU or the concept is dead in the water. With medium pressure CO2 and zircalloy cladding I'm not so sure you can get anywhere near. Water is such a good coolant, it is underrated :lol:


Top
 Profile  
 
PostPosted: Jul 09, 2015 4:03 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1497
The AGR has a channel power of roughly 5MWt with 352kgU/channel.
CANDU manages a channel power of roughly 5.5MWt with ~240kgU/channel.

The channels in the AGR are bigger but that is not a major issue since it is full of something that doesn't really absorb neutrons (compared to a CANDU channel it's empty space). Using zircaloy instead of stainless steel means we can potentially go for more smaller pins as well to increase the heated surface area. Maybe even push as high as 50 cans in the fuel assembly, especially as this design (with its single element per channel) can do away with some of the extra stringers and the like compared to the 8 assembly channel of the AGR.

The delta-T over the AGR core is 300 Celsius, this design could do slightly better despite its lower output temperature by accepting a lower inlet temperature to compensate. Let's say 590 celsius outlet and 200 celsius inlet, which brings us a delta-T of 390C.
That implies either a lower coolant flow velocity through the channel or a thinner gas space which should reduce pumping power I suppose.


As to a scrubber, would we have a tank of NaOH solution in a containment extension adjacent to the reactor vault, with a pipe leading from the vault and connecting to a perforated tube assembly on the bottom of the tank?


Top
 Profile  
 
PostPosted: Jul 09, 2015 4:54 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5058
Just checked some more AGR facts...

Cladding: stainless steel :mrgreen:
Peak cladding temperature: 882C :shock:

Uses SEU/LEU fuel, so not really fair comparison in channel power. in terms of kW/kgU235, CANDU should be about 3300 kW/kg U235 with AGR would be in the 400-600 ish range.

Scrubber (actually just quench tank, scrubbers are active) can be placed anywhere for this concept. So probably low down, somewhere shielded (not normally radioactive though) & accessible. Preferably close to the reactor vault, maybe just right next to it. Would be just a quench tank with a pipe from the relief valve, terminating in a submerged spider-sparger. Plus another pipe likely from the containment/vault area to quench the steam or CO2 after LOCAs.


Top
 Profile  
 
PostPosted: Jul 09, 2015 5:20 pm 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
I read an old paper ( 1964 ) of the CEA speaking of the concept. It was made during the construction of the EL-4.

The EL-4 was a prototype, they imagined what a standard industrial power plant could have been.
They intented to use a beryllium based cladding and UO2 fuel.

The mean burn-up with natural uranium was forecasted to be 8700 MWd/t which is slightly superior to that of CANDU's.

The out temperature of the CO2 is 510°C and the pressure is 75 bars, the steam was at 490 °C and 87 bars. The total net plant thermal efficiency was 35 % ( 500 MWe on the grid and 1430 MWth in the core )

The power per channel was 3 MWth.

It is interesting to imagine this concept with modern materials ( enriched zirconium, SiC, modern supercritical turbines ).

I guess that there is no problem with positive void ( which should be nearly nil ), the problem is dealing passively with LOCAs, there was a power plant is Switzerland of this concept, which was abandonned following a LOCA.
https://en.wikipedia.org/wiki/Lucens_reactor


Top
 Profile  
 
PostPosted: Jul 09, 2015 6:54 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1497
As I understand it, the problem is the heat movement from the cladding to the gas right?
The cladding tubes are extruded, does this mean we could potentially adopt a non circular outside cross section?
Otherwise my plan is to simply go to thinner tubes to increase the zirconium surface area.

The prototype AGR at Windscale used fuel that had pellets 10.5mm in diameter instead of 14.5mm in diameter. Which, if we assume the heat transfer coefficient remains the same, would increase the surface area by 38%, cutting the delta-T by the same amount.
Peak AGR outlet temperatures is 661C which means the delta-T in the AGR is 221C - which means our adapted rods would have a delta-T of around 160C.

160C over our outlet temperature of 590 Celsius gives us - 750 Celsius.
That is quite a bit colder than the AGR, but probably still on the hot side.
If we push to an 8mm fuel pellet (which is the smallest AREVA commonly make) - then we get to a delta-T of something like 122C - which takes us to 712 Celsius.

Can we extrude a serrated outer pattern?

That does have the nice side effect of reducing peak fuel temperature and hopefully thus reducing the peak fuel temperature in an accident.


Top
 Profile  
 
PostPosted: Jul 09, 2015 8:15 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1497
This patent indicates a Zr-Cu-Cr/Mo alloy that has good corrosion resistance at a temperature of ~700C, if we add in chromium plating then 712C from the very small 8mm fuel elements sounds viable.

Go with Zr-0.5Cu-0.5Mo with Chromium plating using Depleted Molybdenum sound good?
Or maybe just use Zr-0.5Cu-0.5Cr as I originally projected?
It appears performance will be similar.


Top
 Profile  
 
PostPosted: Jul 09, 2015 8:54 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5058
I doubt Zr alloys have acceptable high temperature (ie creep-fatigue) strength. I vaguely recall that spent LWR fuel will fail by creep if held at 700C for more than x days. AGR use stainless for a good reason.

Beryllium cladding, that won't work either. Its the typical, physicists triumphing over engineers, that. It happens far too often in nuclear engineering.


Top
 Profile  
 
PostPosted: Jul 10, 2015 6:38 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1497
AGR fuel has far higher fission gas pressure/creep loadings than what I am proposing.
After all it has to restrain the fuel after the gas plenum is closed (whereas with liquid metal bonding we can potentially size the gap so restraint never occurs) - and it has no fission gas plenum at all because there are a stringer of eight elements in one channel so the plenums would be inside the core which would reduce fuel density too much.
This is one of the reasons that AGR enrichments and burnups have stayed so low compared to the LWR - despite the fact that the reactors are not currently allowed to refuel at full power. It can't be driven into a regime with significant fission gas release because the element pressures will climb too high.

Indeed I imagine we can rig it so the fuel cladding is only ever in compression in the core.
I would imagine the only way to test this would involve a gas tube made of the alloy, filled with helium and put in a furnace for a long period at the correct temperature - unfortunately this is far beyond my capability.
For what it is is worth though that patent does indicate that the alloy has much better high temperature creep performance than Zircaloy-2.


Top
 Profile  
 
PostPosted: Jul 10, 2015 7:54 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1497
So I have decided to consider alternative fuel claddings at this time. The liner tube would remain Zr-Cu-Cr as it will not be under significant long term loadings of the type likely to cause creep, it will also be substantially cooler than the cladding as it will only be at the coolant temperature.

So we have a few choices:

1. Fe-20Cr-25Ni-0.5Nb alloy - this is the alloy used in the AGR, so we know its good. Unfortunately it is neutronically awful, with huge captures from the Nickel and Chromium. Enrichment of Chromium might help but the Nickel is still going to soak up neutrons like a sponge.

2. Fe-20Cr-7.5Al - "Kanthal" - an advanced ferritic stainless steel that has been considered for Accident tolerant LWR fuel. It has the advantage of the Aluminium content which is essentially neutron transparent compared to most of the materials we are considering. The chromium content will still soak up neutrons but possible Chromium enrichment would seriously help here. There is an alternative grade being considered which is Fe-15Cr-15Al, which is obviously superior neutronically. Trying to find more information.

3. Silicon Carbide - better neutronically but it has serious R&D requirements since it is still considered to be at least 10 years away for PWR use. It also has thermal conductivity problems late in the exposure cycle.


If Zirconium is out number 2 currently seems the best option in my opinion.

EDIT:
Further information on the FeCrAl front.
An alloy Fe-12Cr-5Al strengthened by a dispersal of Y+Zr oxide apparently does very well in corrosion resistance to steam at high temperatures while retaining excellent creep resistance at very high temperatures. (It forms a protect alumina scale up to 1200+C apparently).
If we go by atm% the composition is:
Fe - 77.4
Cr - 12%
Al - 9.7%
Y - 0.1%
Zr - 0.17%
O - 0.63%
Impurities will be small amounts of nitrogen and carbon primarily. If we use nuclear grade Zirconium there shouldn't be major problem with neutron absorption from impurities so I will just ignore them..
That gives us a weighted neutron cross section of something like 2.37b, which is big but still far better than the alloy used in the AGR - which has a weighted cross section of 3.13 barns.
I will check the relative strengths but I imagine this alloy will come out stronger, permitting a reduction in clad thickness which will further improve it.
(AGR fuel cladding is ~380 microns thick, whereas it is proposed for LWR use that the ODS FeCrAl alloys be only 300 microns, if we assume that accident tolerant LWR conditions are likely to be similar to GCHWR conditions then it seems likely that 300 microns would be sufficient here as well, reducing the FeCrAl effective capture to roughly ~1.87 barns, or 60% of that in the conventional AGR).


Top
 Profile  
 
PostPosted: Jul 10, 2015 11:13 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5058
No need for high nickel content - just use modified regular austenitics. 316 and all its derivatives are cheap and good. The fast reactor people invariably ended with up with 316 class for their cladding. They ended up modifying for very high fast flux swelling resistance, which involved optimizing minor elements (P, S, etc), tweaking Ni/Cr ratio (1:1 is better than the regular 316 ratio), and stabilizing (Ti,Nb,V, preferably some combination). There's a handful of 15/15 class alloys (D9 from the USA for example) that do well. These are near 15 Ni and 15 Cr, Ti-(Nb-V) stabilized class alloys. Pretty much all current fast reactors use this class of alloys now.

I'm not a big fan of ferritics, they are difficult to weld and work with, and not as tough as the austenitics.


Top
 Profile  
 
PostPosted: Jul 10, 2015 11:55 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1497
The problem with 316 is that it is rather more neutron absorbing than the Kanthal derivative.
The cladding is already going to be soaking up vast numbers of neutrons - don't need to lose more if I want to keep the uranium requirement for the reactor down and keep it anywhere near economic.

Do we need that many welds in a fuel assembly? Surely the cans are extruded in one piece and only need end caps welding on?
How do you normally attach the cans into a fuel assembly?

Also, would using liquid metal bonding to reduce the fuel centreline temperature of the fuel pin make it possible to go to larger fuel pins which would reduce the amount of cladding in the fuel compared to the amount of uranium?
Essentially the opposite of what I was proposing with the zircaloy fuel?


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 63 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group