Printer Friendly

Enriched uranium technology at the Sillamae oil shale processing plant.


As a logical sequel to dictyonema shale processing, work with enriched uranium [sup.235]U power reactor fuel and the thorium-rich loparite from Kola Peninsula [1] followed. Both projects ran concurrently in the eighties and ended in early nineties.

The production of up to 180 tons per annum of the metal in the 2.0%- and 3.6%-enriched [sup.235]U[O.sub.2] uranium dioxide nuclear power reactor fuel at the Oil Shale Processing Plant at Sillamae (then P.O.B. P-6685) was decided on March 24, 1980. The factory had to recondition for high temperature operation and bring to the proper isotopic composition enriched [U.sub.3][O.sub.8] uranium oxide from the Electrostal factory (A-7340), located some 50 km eastwards from Moscow.

Altogether 1 360 654 kilograms of enriched uranium in the dioxide were produced from 1983 to 1990, when all uranium-related activities ceased due to political developments that ended the Soviet system, see Table 1. Before that, the factory formed an integral part of the large Soviet nuclear complex, employing more than 100 000 people and serving both military and civilian needs with conjoined fuel cycles. Even electricity was mostly produced in the thermally unstable military-style, originally plutonium-producing light water graphite-moderated RBMK reactors with positive void (vapor bubble) feedback coefficients, well known from Chernobyl, Ignalina and Sosnovy Bor. The main partners in the nuclear fuel fabrication work were the two large fuel fabrication plants at Electrostal (A-7340), at Ust'-Kamenogorsk (B-8857) and the Glazov metallurgical plant M-5057 at Chepetsk that fabricated the zircaloy cladding for fuel rods and assemblies. The two fuel factories received irradiated uranium and chopped scrap from the plutonium and tritium production sites and reactor fuel producers, recovered the uranium as [U.sub.3][O.sub.8] and fabricated the dioxide U[O.sub.2] into both low-enriched (LEU) and highly enriched (HEU) uranium fuel rods and assemblies for military, civilian, naval and special reactors. The RBMK graphite-moderated reactors use 2%-enriched LEU fuel, the pressurized water reactors 2.4%- to 4.4%-enriched LEU, naval reactors, such as the two PWR/VM-A and PWR-VM4 70-MW and 90-MW submarine reactors at the former Paldiski naval base, use 21%-HEU. Very powerful special reactors use anything up to the full bomb quality of >90%.

[sup.235]U enrichment is not the only criterion. The usual [U.sub.3][O.sub.8] oxide is low-melting and cannot be used as reactor fuel. U[O.sub.2] can be thermally stable, but must be prepared as a well-sinterable variety. The oxides are actually groups of more than 40 [2] ill-defined compounds, phases, allotropic forms and solid solutions around [U.sub.3][O.sub.<8] and U[O.sub.>2].

Production of low-enriched [sup.235]U[O.sub.2] (LEU)

Enriched uranium is recovered at the A-7340 and B-8857 fuel fabrication factories as [U.sub.3][O.sub.8] from the sheared and chopped fuel pins. This oxide must be isotopically reconditioned and converted into the proper modification of U[O.sub.2] for use in new fuel rods. Chemical procedures allow to remove all unwanted components except [sup.236]U, formed in the plutonium production reactor through neutron capture to [sup.235]U. This isotope is a non-fissile neutron-absorbing reactor poison that is practically absent in natural uranium ores (relative abundance [10.sup.-10]), but HEU reprocessed from nuclear weapons production material may contain up to 25% of it. The recovered LEU shipments to Sillamae were therefore individually marked with the exact percentages of [sup.238]U, [sup.236]U and [sup.235]U and were additionally measured by mass spectrometry. The [sup.236]U content in recovered [U.sub.3][O.sub.8] was highest (up to 0.43%) in shipments from B-8857 at Ust'-Kamenogorsk, but it never even approached the theoretical 25% limit because the concurrent [sup.240]Pu formation from 239Pu must be kept below 7% in order to avoid premature chain reaction during weapon implosion. The Electrostal product was mostly clean (0.01 to 0.02% [sup.236]U) and that from Glazov M-5057 even pristine.

The [sup.236]U percentage in the regenerated U[O.sub.2] product was generally held close to 0.1% and the [sup.235]U enrichment slightly higher than nominal to compensate for the neutron capture by [sup.236]U.

The enriched U[O.sub.2] production begins with pulsed mixing dissolution of the recovered [U.sub.3][O.sub.8] in 30% nitric acid at 60-70[degrees]C for 2 hours, followed by selective pulsed column 12-contact extraction of uranium with 25-30% tributylphosphate in aromatics-free kerosene (n-dodecane), re-extraction of uranium with water at pH 1 to 2, 60[degrees]C in a 1:1 volume ratio, 8-contact pulsed column, precipitation of ammonium diuranate yellow cake at 60[degrees]C pH 6.8 to 8.0, its thermal decomposition to U[O.sub.3] and thereafter to [U.sub.3][O.sub.8] at 350 to 450[degrees]C and finally reduction to U[O.sub.2] at 650 to 720[degrees]C in a 60 to 70 L/min hydrogen flow, using 245 [m.sup.3] [H.sub.2] for a not quite complete (84 to 91%) reduction of 1 ton of uranium in the dioxide U[O.sub.2.08]. The incomplete reduction and even yellow cake decomposition procedure details are critical for good thermal stability of the dioxide.

The enriched uranium yield from raw [U.sub.3][O.sub.8] to pure U[O.sub.2] was fixed in 1981 for Sillamae to equal the surprisingly low yield practiced for enriched LEU uranium at Electrostal A-7340, see Table 2.

In 1987 the official natural unenriched uranium yield from the concentrates was 99.84% at the Sillamae Oil Shale Processing Plant and the planned share of loss as material unaccounted for (MUF) was just 0.008%. This very high yield was not quite achieved, but the 0.013% unaccounted for loss is also a very good result. For the [sup.235]U-enriched uranium the situation was very different. Instead of the quite realistic 0.008% MUF, 0.730% was officially planned for the unknown losses, and even more (0.746%) was actually reported. The whole uranium project ended at Sillamae in December 1989 due to imminent political changes in the Baltic States. All uranium stocks, materials and equipment were immediately removed to Electrostal and the enriched uranium MUF fell immediately to the very reasonable 0.004% level just before final shutdown without any preceding technological improvements. Why the rather large, about 10 tons at Sillamae alone with possibly much more at Electrostal, where the same excessive losses were enforced, officially nonexistent enriched uranium stocks were created in a very non-transparent manner even at the Top Secret level, is unknown. At any rate, it is not in the waste depository at Sillamae. The very careful measurements of Ehdwall et al. at the Swedish Radiation Protection Institute [3] show that the average [sup.238]U/[sup.235]U activity ratio is equal to 20.5 [+ or -] 3.5 in the waste dump borehole core samples, taken at [less than or equal to] 10 m depth from the upper gray layer that accumulated in the eighties.

The ratio expected for natural uranium is 21.5, but the amount of [sup.235]U in MUF, if present, would have nearly halved this number. No [sup.236]U or actinides were found, which is also relevant.

Pebble bed reactor fuel technology at Sillamae

The centerpiece of any truly modern nuclear power plant is some modification of the high temperature pebble bed modular reactor (PBMR) that uses helium gas as coolant, graphite as moderator and no water in the core. It is completely fail-safe and cannot explode or melt. The very high operating temperature (1000[degrees]C) allows electricity generation by Brayton cycle gas turbines or hydrogen production through direct catalytic (sulfuriodine cycle) water decomposition for hydrogen-based economy [4]. It uses 60 mm diameter graphite spheres filled with thousands of tiny, less than 1mm diameter HEU particles coated with a layer of porous pyrolytic carbon providing space for fuel expansion and fission gas formation, isotropic pyrolytic carbon layers for mechanical stability, a dense silicon carbide shell as diffusion barrier for the fission-formed Cs, Sr and Ba, and finally a dense strong pyrocarbon layer that compresses and fixes the carbide shell. The coated particles are compacted with pyrocarbon-bonded or coke-generated graphite into 60 mm diameter fuel spheres that are covered with dense graphite and polished into near-perfect tough black spheres that can roll and fall some hundreds of kilometers without any significant dust formation in the reactor hopper. The modular design separates the reactor hopper from the turbines and/or hydrogen production and is designed to withstand rockets and direct hits by large aircraft.

The production of 0.7 mm diameter 21%-enriched HEU microspheres began in 1982 using agglomeration with 17 to 20% hydrocarbon binder (paraffin 65%, petrolatum 30%, oleic acid 5%). The mechanically formed raw microspheres were sorted for perfect sphericity on a 2D-vibrating sorting plate, interspersed at first with [Al.sub.2][O.sub.3] [5] which introduced unwanted aluminum, and thereafter with NH4Cl that sublimes at 337.6[degrees]C after driving off all the organics, and heated for several hours in a [10.sup.-1] to [10.sup.-3] mm Hg vacuum at first at 90[degrees]C, then 300[degrees]C and finally 950[degrees]C to remove all organic matter without harming the perfectly spherical shape. The microspheres were further calcined and sintered for 5 to 7 hours at 1650[degrees]C in a [10.sup.-2] to [10.sup.-5] mm Hg vacuum. Treatment at 1650[degrees]C in a reducing atmosphere (4% [H.sub.2] in argon) for 2 hours may follow. A 80 to 85% (9.0 g/[cm.sup.3]) density was achieved and the spheres were thoroughly detergent-washed to remove any adhering uranium oxide dust.

As the next step a triple coating was deposited in a preheated fluidized bed reactor, flushed with pure argon gas. The first 35 to 40 [micro]m thick low density (1 to 1.15 g/[cm.sup.3]) porous layer was formed in 15 min at 1300[degrees]C in a 4% propane-butane argon mixture. The second 10 to 15 [micro]m thick medium density (1.3 to 1.5 g/[cm.sup.3]) fixing layer was formed in 2 hours at 1500[degrees]C in a 0.8% propane-butane argon mixture. The third 50 to 70 [micro]m thick high density (1.8 to 1.9 g/[cm.sup.3]) layer was formed in 14 hours at 1500[degrees]C in a 0.4% propane-butane argon mixture. Thereafter the coated microspheres were treated for 48 hours at 90[degrees]C with 8N nitric acid and checked for surface [alpha]-activity. The perfectly coated spheres could be separated by flotation in bromoform, thoroughly washed and dried. The 90 to 110 [micro]m thick silicon carbide shell (density 3.15 to 3.5 g/[cm.sup.3]) was made in 16 hours at 1500[degrees]C in argon with additional hydrogen gas (250 to 400 L/h), saturated at 25[degrees]C with 4% methyltrichlorosilane that decomposes into SiC and HCl. The tightness of carbide shell was checked by heating the microspheres in air. This treatment destroys the microspheres with damaged carbide coating. Finally a dense (5 to 10 [micro]m thick) pyrolytic carbon coating is deposited upon the carbide shell at 1300[degrees]C in a 0.4% propane-butane argon mixture.

The fuel spheres, about 60 mm in diameter are formed from a graphite (84%) glycerine (16%) mixture that binds the central 40 mm diameter core of HEU fuel microspheres in a graphite outer shell of 10 mm thickness. All this is pressed together at 60 kg/[cm.sup.2] for the inner core and 30 kg/[cm.sup.2] for the outer shell. The fuel spheres are thermally compacted with pyrolytic carbon formed from methane (preferred) or propane-butane in argon at 860 to 1000[degrees]C, 0.3 to 0.7 ata pressure for 80 hours. Direct coking without the use of pyrolytic carbon requires a very high 1800[degrees]C temperature for direct graphitization that can harm the fuel microspheres. The compacted spheres are finally polished to 60 +0.65/-0.05 mm diameter.

The first model-making experiments were carried out with natural uranium dioxide or even without a fuel particle. Altogether 3600 such models were produced from 1982 to 1986, but one hundred 21%-enriched HEU fuel spheres were also produced. Preparations were carried out to produce 30 000 HEU fuel spheres per year, to be extended to 200 000 per year.

However, political climate was changing. All development and production activities with HEU were abruptly terminated already in early 1987, or three years before complete closure of the uranium production in Estonia.

Presented by A. Raukas Received May 24, 2006


[1.] Lippmaa, E., Maremae, E., Rummel, A., Trummal, A. Tantalum, niobium and thorium cake production at the Sillamae Oil Shale Processing Plant // Oil Shale. 2006, Vol. 23, P. 281-285

[2.] Eyring, L. Refractory oxides of the lanthanide and actinide elements // High-Temp. Oxides. 1970. Vol. 5 II, P. 41-97.

[3.] Ehdwall, H., Sundblad, B., Nosov, V., Putnik, R. Mustonen, H., Salonen, L., Qvale, H. The content and environmental impact from the waste depository in Sillamae // Swedish Radiation Protection Institute SSI--Rapport 94-08. 1994.

[4.] Freemantle, M. Nuclear power for the future // Chemical&Engineering News. 2004, Vol. 82, No. 37. P. 31-35.

[5.] Maremae, E., Tankler, H., Putnik, H., Maalmann, I. Historical survey of nuclear nonproliferation in Estonia 1946-1995, Kiirguskeskus. 2003. P. 19-46.

E. LIPPMAA, * (a) E. MAREMAE (a), A. TRUMMAL (a), A. RUMMEL (a), J. LIPPMAA (b)

(a) National Institute of Chemical Physics and Biophysics (NICPB), Ravala 10, 10143 Tallinn, Estonia

(b) Conseil Europeen pour la Recherche Nucleaire (CERN), Geneve 22, Switzerland

* Correponding author: e-mail
Table 1. Production of [sup.235]U-enriched U[O.sub.2] in 1983-1989
at the Sillamae Oil Shale Processing Plant

Nominal * Content of uranium metal in
[sup.235]U kilograms in [sup.235]U-enriched
enrichment, U[O.sup.2.08] dioxide **
 1983 1984 1985 1986

 2.0 34 623 79 109 98 606 53 489
 2.4 - - - -
 3.0 13 190 25 504 - -
 3.3 68 025 52 056 71 492 88 342
 3.6 - - - 19 027
 4.4 - - - -
 21.0 - - - -
 45.0 - - - -
 90.0 - - - -
Total 115 838 156 669 170 098 160 858

Nominal * Content of uranium metal in 1989
[sup.235]U kilograms in [sup.235]U-enriched retail
enrichment, U[O.sup.2.08] dioxide ** price,
 % roubles/kg
 1987 1988 1989

 2.0 103 547 119 173 40 119 337
 2.4 78 036 58 679 150 569 422
 3.0 40 347 73 816 66 082 553
 3.3 22 571 4 252 -
 3.6 - - -
 4.4 - - - 840
 21.0 - - - 4080
 45.0 - - - 8900
 90.0 - - - 19800
Total 244 501 255 920 256 770

* The actual [sup.235]U enrichment is slightly larger than nominal due
to [sup.236]U contamination.

** The dioxide is not fully reduced for better thermal stability in the
fuel rod.

Table 2. Total planned and actual LEU reconditioning losses in
1983-1989 at the Sillamae Oil Shale Processing Plant according to
the formerly classified Top Secret monthly production reports

 Year Natural uranium in [U.sub.3][O.sub.g]

 Total loss % with Material
 MUF unaccounted for

 Planned Actual Planned Actual

 1981 0.28 0.28
 1982 0.26 0.25 0.06 0.08
 1983 0.23 0.23 0.06
 1984 0.21 0.21 0.04
 1985 0.20 0.20 0.10 0.04
 1986 0.175 0.174 0.059
 1987 0.16 0.16 0.008 0.013
01. 1988 0.15 0.15
07. 1988 0.15 0.15
12. 1988 0.15 0.15
01. 1989 0.11
07. 1989 0.11
12. 1989 0.11 0.06

 Year Enriched LEU uranium in U[O.sub.2]

 Total loss % with Material
 MUF unaccounted for

 Planned Actual Planned Actual

 1981 0.85 * -
 1982 0.85 -
 1983 0.88 0.88 0.44
 1984 0.86 0.86 0.54
 1985 0.838 0.835 0.60
 1986 0.79 0.79 0.54
 1987 0.750 0.750 0.730 0.746
01. 1988 0.74 0.74
07. 1988 0.68 0.68
12. 1988 0.655 0.655
01. 1989 0.655 0.655
07. 1989 0.600 0.030
12. 1989 0.600 0.004 <0.004

* The total planned LEU reconditioning loss at the
Electrostal A-7340 reactor fuel factory
COPYRIGHT 2006 Estonian Academy Publishers
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lippmaa, E.; Maremae, E.; Trummal, A.; Rummel, A.; Lippmaa, J.
Publication:Oil Shale
Date:Sep 1, 2006
Previous Article:Assessment of remaining life of superheater austenitic steel tubes in oil shale PF boilers.
Next Article:Tantalum, niobium and thorium cake production at the Sillamae oil shale processing plant.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters