Lev Kochetkov: from mercury to sodium, from BR-1 to BN-600

Whether at once appeared sodium coolant in Russian fast reactors, or sodium has had rivals? What types of fuel have been regarded as a fuel for the fast reactor program, and do we need such a high BG? These and some other questions web-site AtomInfo.Ru addressed to the IPPE directory adviser Lev Kochetkov.

With Obninsk for a lifetime

For the first time I came to the Laboratory "B" (nowadays IPPE) in 1951 to undergo practical training. In a year I came back here to undergo pregraduation practical training. From April 1952 till now I have been working in IPPE, as a directory advisor recently.

I began working in the laboratory of Mikhail Egorovich Minashin on the problem of thermal reactors, was making calculations of the AM reactor for the First NPP. Then I participated in reactors research for Beloyarskaya NPP, transportable TES-3, naval propulsion, Bilibino NPP. Took part in the start up of the Siberian NPP (Tomsk-7), two blocks of Beloyarskaya NPP, NPP TES-3, NPP in Czechoslovakia A-1.

From 1969 I started working on the problem of fast neutron reactors (fast reactors). So I know about the first period of fast reactor researches only from the words of my colleges.

At first I had to get into the swing of things and then to participate in designing and launching of experimental-industrial reactor BN-350 in the town Shevchenko (Aktau) in Kazakhstan. After that appeared the projects of more powerful reactor facilities BN-600, BN-800, BN-1600, among that projects only BN-600 has been realized.

BN-600 reactor came out of the BN-350 project. BN-600 was designed as a modification of BN-350. Capacity of a new reactor was supposed to be 500 megawatts (el.) at first, figure of 600 megawatts appeared later, during elaboration and after making numerous changes in the project, taking into account collected domestic and foreign experience of fast reactors exploitation.

Capacity of 600 megawatts was a compromise between two desires - increase the capacity without significant departing from the conventional design, applied to the BN-350 project. As a result BN-600 had certain constructive reserves that gave us an opportunity to design BN-800 reactor with the same vessel. Excessive reserve proved to be in sodium-sodium heat exchangers, we knew about it and have used it in BN-800 project.

As a result, we managed to achieve 800 megawatts capacity in the same reactor vessel as of BN-600, now we can speak even about 880 megawatts.

Don't you know that sodium was not the first candidate for the role of a coolant for fast reactors? There were other candidates that seemed to be more promising than sodium.

Fast mercury reactor

In our country IPPE has made the most significant contribution to the development of fast reactors, certainly, by the great support from other domestic organizations. But fast reactors research has been started in the USA. Most important priority at first was to find the method of plutonium enrichment for military purpose.

First we had to guess, that fast neutron spectrum is capable of improving reproduction characteristics! A crucial role must have played Enrico Fermi.

Generally speaking, history of nuclear energy has examples of genius inspirations. I always remember the words of our compatriot academician V.I.Vernadsky, pronounced in 1922 about the prospects of nuclear energy. How could he at that time foresee the appearance of NPPs - it is mind-boggling, as all main discoveries that paved the way for using nuclear energy have been made later.

After Fermi guessed about the advantages of fast neutron spectrum, in the USA in 26th of April 1944 took place an important conference "Discussion of reproduction problems". There Fermi said about possible designs of fast reactors and presented results of his first calculations, specifying the reproduction coefficient range from 1.37 - 1.57.

One of the participants of that memorable conference (Morrison) recollected later, that all participants were very excited after they realized that in fast reactors can be collected more fissionable substance then used in chain reaction. They realized that mankind will receive source of energy for thousands of years.

In 1946 in the United States was launched the first fast reactor CLEMENTINE with mercury coolant. In 1951 followed EBR-1 with sodium-potassium coolant. Four years later the first fast reactor of zero capacity (BR-1) appeared in the USSR, and in 1956 was constructed BR-2 reactor with the capacity 100 kilowatts with mercury coolant. Approximately at that time in Great Britain were launched ZEPHYR and ZEUS of zero capacity. In 1959 we have launched reactor BR-5 with sodium in the first circuit and with sodium-potassium in the second one. Then France, Germany, and Japan joined to the scientific-technological competition. All first research fast reactors in our country were designed and created under the scientific direction of IPPE.

#
Reactor
Year
Country
Power
Coolant
1
CLEMENTINE
1946
USA
20 kilowatts
Mercury
2
EBR-1
1951
USA
1,4 Megawatts
Na-K
3
ZEPHYR
1954
Great Britain
0
no
4
BR-1
1955
USSR
0
no
5
ZEUS
1955
Great Britain
0
no
6
BR-2
1956
USSR
100 kilowatts
Mercury
7
BR-5
1959
USSR
5 Megawatts
Na, Na-K
8
DFR
1959
Great Britain
60 Megawatts
Na-K
9
EBR-2
1962
USA
62 Megawatts
Na
10
Enrico Fermi
1963
USA
300 Megawatts
60 Megawatts(el.)
Na
11
RAPSODIE
1967
France
20-40 Megawatts
Na
12
BOR-60
1969
USSR
60 Megawatts
Na
13
BR-10
1973
USSR
8 Megawatts
Na
14
BN-350
1973
USSR
1000 Megawatts
250 Megawatts(el.)+water
Na
15
PHENIX
1973
France
250 Megawatts(el.)
Na
16
PFR
1974
Great Britain
250 Megawatts(el.)
Na
17
JOYO
1977
Japan
140 Megawatts
Na
18
KNK-II
1978
Germany
20 Megawatts(el.)
Na
19
BN-600
1980
USSR
600 Megawatts(el.)
Na
20
FFTF
1980
USA
400 Megawatts
Na
21
SNR-300
1985
Germany
300 Megawatts(el.)
Na
22
SUPERPHENIX
1986
France
1200 Megawatts(el.)
Na
23
MONJU
1995
Japan
250 Megawatts(el.)
Na

How have been fast reactors developed in our country? In the end of 40th's Alexander Il'ich Leipunsky formulated main concept of fast reactors - their possible advantages and main technologic decisions. It was necessary to choose a coolant, fuel and construction. IPPE started large-scale researches of different types of coolants - mercury, sodium, eutectic compositions sodium-potassium and lead-bismuth, lithium. As a result in IPPE was created one of the best schools (may be the best in the world) of liquid-metal coolants.

For the first soviet fast reactor choice was set on mercury. But we haven't managed to work on this coolant more then a year. Experience showed that mercury was unpromising coolant. More over, dangerous for the man's health. Mercury produced a significant corrosive impact on the materials of the construction that, finally, was the main reason for aborting the use of mercury in fast reactors.

You will ask - why it happened so, that we and Americans have chosen mercury as coolant for our first fast reactors? You know, I have heard about mercury during my university studies. It was proved then, that from the point of view of thermodynamic, on the mercury vapor could be organized a power supplier with a fine coefficient of efficiency by relatively low temperatures. Yes. You have understood correctly - mercury vapor should have been directed right to the turbine. This way we have been told about it. Certainly, these had been only theoretical calculations, that haven't been realized in thermal power engineering later.

Why was mercury so interesting for the fast reactors? The fact is that coolant in fast reactors should not only absorb a small quantity of neutrons, but also moderate them insignificantly. Mercury is a heavy metal that practically doesn't moderate and absorb neutrons. From the point of view of physics, mercury was at first an ideal choice. Spectrum in a mercury reactor remained a fission spectrum. Incidentally, in sodium reactors situation is not the same - sodium atomic weight is relatively low and partially it works like a moderator. In this respect, by the way, lead and lead-bismuth also have an advantage over sodium.

From the point of view of physics thermodynamics mercury seemed to be the best candidate. Unfortunately, mercury appeared to have a lot of drawbacks and as a result we had to refuse from it.

We have returned to mercury once again. Originally in BR-5 project was a nuclear steam generator. Second circuit with sodium-potassium was separated from the water coolant by the surface of heat transfer that consisted of two diploblastic tubes in the nuclear steam generator. And a small spacing between these two tubes was filled, again, with mercury. But it has done a bad turn! Corrosive problems appeared again, and as result mercury penetrated into sodium-potassium. Till nowadays we store in tanks sodium-potassium with mercury that remained from BR-5 reactor. We will still have to reprocess and render harmless this cursed alloy that consists of sodium, potassium and mercury. (Lev Alekseevich is laughing).

Nevertheless fast sodium

After mercury many turned to sodium coolant. How has it seduced designers and scientists all aver the world? It has perfect thermal physics characteristics - high thermal conductivity, very high heat transfer coefficient. Sodium is relatively light, that is why its hydraulic resistance is smallish. Very important sodium advantage is a relatively good compatibility with many steels.Sodium absorbs neutrons, but rather few of them. But on the other hand boiling temperature is high (700°C), that is why reactor vessel can be designed for the atmospheric pressure without the apprehension of coolant boiling. In our BN reactor projects pressure excess doesn't exceed one atmosphere. You can compare the thicknesses of water cooled reactors that work by the coolant pressure 160 at. and the thickness of BN reactors - you will feel the difference. Thickness of a fast reactor vessel is some centimeters!

Sodium's rival was also sodium-potassium - a very seducing coolant. Please, recall - Americans after "mercury reactor" launched precisely "sodium-potassium" one - EBR-1. In Great Britain and in our country were also keeping an eye on this alternative. Why was it better then sodium? Na-K melting temperature was low. By the room temperature it still remained liquid. It means that there is no need to include in the design of the reactor circuit heating system, as in the case of sodium, whose melting temperature is about 70°C.

Sodium-potassium didn't manage to stand a competition for engineering reasons, exactly because of complexity of maintaining equipment, more exactly because of the low freezing temperature - characteristic, that seemed to be an advantage at first. First, it appeared to be more aggressive then sodium and more flammable. Secondly, all repair work with Na-K turned out to be more complex.

Look here. If we turn off heating system of a sodium reactor, circuit temperature will fall and sodium will freeze. Now we can calmly cut tubes, prepare them for welding, general speaking, do everything that we need for the repair work, without apprehension of fire, because sodium in a circuit is already solid. And now imagine what a headache for operatives might have been Na-K! As a result, considerations of a serving simplicity prevailed and sodium has been chosen. At the same time, choice of Na-K for the space nuclear power systems BUK and TOPAZ, designed under IPPE scientific guidance, was extremely successful.

You know that nowadays other coolants are being explored - lead and lead -bismuth. IPPE has gained significant experience on the issue of lead-bismuth coolant, constructing submarine reactors, and we hope, that experimental fast reactor SVBR-100 with a lead-bismuth coolant will be build. Such coolants have their own advantages - first, they a less aggressive to air and water then sodium. It is also very important that they moderate neutrons insignificantly. But they have problems with corrosive stability. But we have done a lot in IPPE to overcome these drawbacks of lead-bismuth coolant.

There was one more direction of fast reactors research in our country - gas cooled reactors. This idea was supported by Savely Moiseevich Feinberg from Kurchatov Institute. In the end of 40th's, approximately when Leipunsky formulated his conception of fast reactors Feinberg proposed an idea of gas cooled fast reactor (with helium as a coolant). Kurchatov Institute was working over this idea for a long time. Admittedly, they still have great enthusiasts of this direction.

But why is gas so bad, and what drawbacks of such a coolant have been revealed in the first years of the research? At first, we need high pressure. One more drawback - thermal physics characteristics of helium in comparison with sodium are much worse. Problems of reactor shut-down cooling are difficult to be solved. For these and that reasons, fast gas-cooled reactors also haven't been realized.

Nowadays a lot of scientists are trying to find any solution that would combine the advantages of sodium and gas coolants. For example French researchers are proposing to use sodium in the first circuit and gas in the second, CO2 exactly (Brayton cycle). They believe that CO2 can give a good thermodynamic coefficient of efficiency by the temperatures, achieved in reactor with a sodium coolant. Supplementary advantage can be an opportunity to remove intermediate sodium circuit, and it means that sodium-gas alternative may become cheaper then a sodium one with a water coolant - working substance in the third circuit.

I would like to stress, that Frenchmen are researching gas-cooled reactors very intensive. But when we met last time we understood, that they saw significant difficulties with the possible accidents with the tube depressurization in the heat exchanger and subsequent water penetrating into sodium circuit and active zone. Don't forget, that gas pressure considers to be much higher then sodium pressure. And gas in the active zone will cause hardships with reactivity. It seems to me that Frenchmen are gradually coming back now to the sodium reactors.

Nitride, carbide and vibropac technology

As you remember, choice of a fast reactor design is not limited by the choice of a coolant. One more question remained - a type of fuel. From the point of view of reproduction, we would like to have fuel with a higher density then dioxide. Dioxide average density range is within 8.5 - 9.0 g/cc. For nitride, carbide or, first of all, metal fuel we can get compositions with higher densities, and it is very profitable from the point of view of BG. Besides, these types of fuel have a higher level of thermal conductivity and it is also an advantage. Alexander Ilyich Leipunsky insisted upon the necessity of researching high-density fuels.

During Leipunsky's time uranium carbide zone for BR-5 reactor was build. But we couldn't achieve enough BG, having stopped on the level within 5-6% that is very low for fast reactors. On oxide fuel, you know, BG reaches 10% and 15% in perspective, on the research reactor BOR-60 some assemblies from the uranium dioxide have had a 30% BG.

Besides a low BG, carbide fuel appeared to have other disadvantages, connected with the exploitation difficulties. Treatment of the defective carbide fuel elements is extremely difficult. It is because carbide fuel is pyrophoric and very volatile in addition. As a result contamination of a pickup mechanism occurs.

Next step in the high density fuels research was creating of two zones from nitride - uranium fuel for BR-10 reactor. Here we also don't have considerable results. In contrast to carbide we haven't managed to increase BG to 8-9%. Further occurred numerous fuel assemblies depressurizations as a result of fuel swelling. Fission products were concentrating in fuel, it began to press on the shell of the reactor and that coursed fuel assembly depressurization with the fission yield to the first circuit. Nitride fuel has one more disadvantage - it collects long-life isotope 14C during its irradiation in reactor.

But we have to admit, that we have been working over nitride not enough. There are some ideas how we can increase BG and cope with its swelling. We need time and money for that. A considerable amount of work over metal fuel was done in the United States and in our country in Research Institute of Atomic Reactors and All-Russian Scientific Research Institute for Inorganic Materials. We don't have reasons now to confess that it is prospective.

You may ask, do we have to fight for a high level of a BG? We have no need to do it now. But some day such a necessity may arise. If we don't have fast reactors-breeders, according different assessments -made in our country and abroad - known uranium reserves will be enough only for some decades. Assessments are different, you are right, a lot will depend on a conjuncture. At a certain oil-price it can be profitable to gain uranium even from a sea-water.

There is no need to have a high BG level, one unit is enough to make a next step in the development of fast reactors. Don't forget that in our country and all over the world a large amount of plutonium is stored, both extracted and contained in the spent fuel. If we use it, we will manage to extend the life-time of a combined system that consists of fast and thermal reactors. But more or less we will have to develop the system in the future, that consists generally of fast reactors, that have a high BG level. I think, that a BG level should be something about 1.25-1.30.

But if we have BG level of one unit, or something more than a unit, we will have a chance to realize one of the most important advantages of fast reactors from the point of view of security.For example, VVER must have a significant reactivity reserve for the BG to work for a campaign. This reactivity reserve creates the potential danger of reactor runaway on instantaneous neutrons, like in a bomb. But if a BG is close to one unit, our burned up fuel will be totally replaced by new fissile materials, produced in the reactor.

In this case we don't need a reactivity reserve. Let us do so, that a reactivity reserve for BG should be smaller, than beta - and the danger of reactor runaway on instantaneous neutrons disappears. It is a very important advantage that we still haven't realized in our fast reactors. Finally some words about vibropac-fuel. Englishmen began to develop this type of fuel, but then refused. Then this idea was taken up by Americans. In due time I visited the facility, that they have constructed in Hanford for producing fuel elements using vibropac technologies for FFTF reactor. But they haven't started doing it. I have asked them: "Why have you refused from this idea?" The answer was: "When using vibropac technologies we can't achieve characteristics, that we have using fuel pellets."

But we have to admit, that Research Institute of Atomic Reactors has reached significant results in recent years in the sphere of vibropac technologies in comparison with what has been done abroad. Patterns of such assemblies were made and tested. But, still, we have to work a lot on it. Just imagine - in one core charge of BN-600 - more then 40 000 fuel assemblies. And we have to achieve a sufficient level of reliability to avoid depressurization in terms of a high BG! Now we are working on a lot it in our country.

Lev Alekseevich, thank you very much for the interview for AtomInfo.Ru web-site.

SOURCE: AtomInfo.Ru

DATE: March 20, 2008

Topics: NPP, Russia, Fast breeders


Rambler's Top100