P-5-307

Planning for the Decommissioning of the ASTRA-Reactor

Konrad Mück, Jörg Casta
Austrian Research Center Seibersdorf

Introduction
The ASTRA Reactor, a 10 MW multipurpose MTR research reactor at the Austrian Research Center
Seibersdorf is to be decommissioned after 39 years of successful operation. The reactor had reached first
criticality in September 1960 and was operated at a power level of 100 kW until April 1962. In May 1962 the
power level was raised to 1 MW and in August 1962 to 5 MW. In 1969 the power level was further increased to
6 MW and three years later to 7 MW. Since January 1975 the reactor has been operating on a maximum power
level of 8 MW and in the last decade of 9.5 MW. The average operating period per year was 500 - 900
MWd/year since 1966. From this rather intensive utilization an activity inventory is to be expected which
requires a thorough planning of the decommissioning procedure and stages.
To perform a safe and environmentally compatible decommissioning of the reactor, the possible options
and the required stages for decommissioning and removal of the radioactive components were evaluated in a
decommissioning study. It was decided to follow the stages proposed by IAEA (1). One of the major questions to
be answered in the study was the optimum time scale of reaching each stage and the optimum intervals between
each stage.
To support the decisions at each stage, an estimate of the activity inventory in the various parts of the
reactor and the waste volume to be expected was performed. Measurements of various materials as far as
accessible and numerical evaluations where not accessible were carried out.
Another important issue with regard to dismantling and demolitioning are the methods and procedures
to be employed to result in a minimal radiation exposure of the employed staff, minimize the arising amounts of
waste and minimize the impact on the environment (2). The choice and deployment of most of these methods is
based on long experience at the ASTRA-reactor with the removal of highly-radioactive structures, experimental
facilities and beam tubes. As far as foreseeable at present, they will be described.

Characteristics of reactor
The ASTRA-Reactor is a 10 MW multipurpose research reactor. A view of the reactor is given in fig. 1.
The reactor was designed for irradiation of samples from the top through the pool water, by irradiation loops and
rabbit systems, by irradiation devices adjacent to the core and by a thermal column and beam tubes. Some of the
irradiation facilities and devices were installed at a later phase. This was possible due to the high flexibility of
the design of the reactor. These later installed features include a fast neutron irradiation facility for seed
irradiation, large volume irradiation chambers and irradiation facilities for NTD-silicon production.

Table 1 Relevant parameters for ASTRA-Reactor (3)

type of reactor adapted pool type reactor, light water cooled

and moderated
fuel UxSiy-Al
enrichment 19.85 %
fuel type MTR fuel element with Al-cladding
23 fuel plates in standard element (412 g 235U)
17 fuel plates in control element (304 g 235U)
reflector Beryllium and water reflected
reflector consisting of Beryllium-elements
thermal power 10 MW
neutron fluence, thermal 1,0 . 1014 n/cm²s
fast 0,7. 1014 n/cm²s
coolant flow 840 m3 h-1 (velocity: 3.5 m s-1)
primary system, volume 190 m3
material aluminum

1
P-5-307

1. inner liner
2. biological shield
3. reactor core
4. absorber rods
5. plenum
8. lower hot cell
9. upper hot cell
10. connecting channel
11. thermal column

Figure 1 Cross section view of ASTRA reactor

The reactor originally possessed 10 horizontal beam tubes of which 7 were removed 15 years ago and
the rest 3 years ago (4,5). From this manipulation of higher radioactive parts (in the range of some TBq) with
structure difficult to handle, some of the experience of the staff in manipulating higher radioactive, bulky parts
results.
The reactor is a light water moderated and light water cooled, pool-type reactor. Its core consists of
typically 20 - 24 MTR fuel elements of high U-content (412 g 235U per fuel element). The neutron balance and
thus the burn-up of the core is optimized by a Be-reflector around the core which consists of 25 beryllium
elements of similar size as the fuel elements. More details are given in table 1.

ESTIMATE OF ACTIVITY INVENTORY AND WASTE VOLUMES

For an estimate of the total activity content and the waste volumes to be expected an evaluation of the
activity and waste volume had to be performed on a component by component basis. Each component, device or
structural element was surveyed, its chemical composition estimated and the approximate total neutron fluence
at its site in the reactor reconstructed. Of particular importance in this context were trace elements with high
neutron activation cross sections. From long-term experience with most of these materials, from activation
analysis performed before some of the materials had been deployed in the reactor and from measurements
performed in the context of the study in a few cases where no data were available, the activity of these
components was estimated. Where no measurements were available, activities were calculated by activation
tables as given by Erdtmann et al. (6). This is particularly true for the main chemical elements of a specific
device.
For the experimental facilities in the ASTRA-reactor this is given in table 2. Experimental facilities in

2
P-5-307

this context mean all facilities which had been used for irradiation purposes and are rather easily removable. Due
to their utilization in the reactor, they were exposed to the highest irradiation fluences and therefore have the
highest specific activities.
Table 2 Estimated activity inventory in the experimental facilities

Radio- Half-life Radiation, maximum specific Material

nuclide γ-energy [MeV] activity [Bq/g]
3
H 12.3 y β, no γ 1 105 beryllium
14
C 5736 y β, no γ 2 graphite
51 10
Cr 21.7 d ε, γ (0.32) 1 10 stainless steel
46
Sc 83.3 d β, γ (0.89,1.12) 2 105 aluminium
54
Mn 312.5 d β, γ (0.83) 2 107 steel
55 9
Fe 2.7 y ε, no γ 5 10 steel
59
Fe 44.6 d β, γ (1.09,1.29) 2 108 steel
58 9
Co 70.8 d ε,β , γ (0.51,0.81)
+
1 10 nickel, stainless steel
60
Co 5.27 y β, γ (1.17,1.33) 1 108 Be, Ni, Al, stainless steel
59 6
Ni 75000 y ε,β+, no γ 5 10 nickel
63
Ni 100 y β, no γ 6 108 nickel

The major activity in these components is 51Cr which is activated in stainless steel. With a half-life of
21.7 d, however, its activity is virtually gone one-year after final shutdown. Among the longer-lived
radionuclides only 60Co with a maximum specific activity of 1.1011 Bq kg-1 is of significance. This nuclide is also
of relevance with regard to its penetrating gamma-radiation and requires special precaution and protection in
dismantling procedures. Another major radionuclide is 63Ni in the nickel-liners of irradiation devices for Si-
doping. But since 63Ni is a pure emitter, the major contribution to the exposure of the staff during
decommissioning comes also in these irradiation facilities from 60Co.
The amount of waste to be expected from these experimental facilities is listed in table 3. Although
efforts had been undertaken throughout the history of the reactor to deploy aluminum wherever possible, a major
fraction of this waste is stainless steel. This comes mainly from beam tube experiments. Aluminum parts are not
only easier to disintegrate, they also contain usually much less activity. The majority of induced activities have
short half-lives and are sufficiently decayed until the begin of dismantling. However, in aluminum also 58Co and
60
Co is produced from impurities. Therefore, aluminum from experimental facilities near the core usually
contains too much activity to be classified as low-level-waste.

Table 3 Estimated waste amount from experimental facilities

Material waste volume [kg]

aluminium 2153
graphite 283
lead 180
beryllium 50
nickel 38
stainless steel 2528
boron carbide 2
sum 5234

3
P-5-307

Table 4 shows the estimated maximum specific activity in the various devices installed in the pool. This
comprises the grid plate, the plenum, beam tubes, the thermal column, control rods, the inner liner of the pool
and the lead plates to shield the concrete pool wall against gamma radiation during the life-time of the reactor.
The major activity contributor is 175Hf with a half-life of 70 days. This activity is to 100 % contained in the
Hafnium absorber rods which were used in the reactor in the last 15 years. The rather short half-life of 70 days
and the low specific activity of 178Hf results in an easier handling when the core is dismantled after a one-year
cooling-down period for the fuel elements. Apart from the absorber rods, the most important activities are 51Cr
and the Co-isotopes, similar to the experimental facilities.
Another major radionuclide in that context is the 63Ni in the silicon irradiation facilities and in stainless
steel parts of some devices. Not very relevant to the total activity in the waste, but relevant with regard to
amount, lead which was used on the outer side of the pool to protect the concrete from heating and irradiation

Table 4 Estimated activity inventory in the devices fix-installed in the pool

Radio- Half-life Radiation maximum specific Material

nuclide γ-energy [MeV] activity [Bq/g]
3
H 12.3 y β, no γ 1 105 beryllium
14
C 5736 y β, no γ 2 graphite
51
Cr 21.7 d ε, γ (0.32) 3 109 stainless steel
46
Sc 83.3 d β, γ (0.89,1.12) 2 105 aluminum
54
Mn 312.5 d ε, γ (0.83) 5 10 6
stainless steel
55
Fe 2.7 y ε, no γ 2 10 9
stainless steel
59
Fe 44.6 d β, γ (1.09,1.29) 6 107 stainless steel
58
Co 70.8 d ε,β+, γ (0.51,0.81) 1 109 stainless steel
60
Co 5.27 y β, γ (1.17,1.33) 1 108 Be, Al, stainless steel
59
Ni 75000 y ε,β , no γ
+
5 10 6
stainless steel
63
Ni 100 y β, no γ 6 108 stainless steel
175
Hf 70 d β, γ (0.34) 2 1010 hafnium
178
Hf 31 y γ (0.32,0.57) 6 102 hafnium

4
P-5-307

Table 5 Estimated waste amount from devices fix-installed in the pool

Material waste volume [kg] waste volume [kg]

activated contaminated
aluminium 2732 3162
graphite 10300 0
lead 10635 0
concrete 6246 0
stainless steel 722 0
cast iron 708 0
paraffin 765 0
glass 0 461
beryllium 130 0
hafnium 18 0
plastics 0 25
boron 17 0
sum [kg] 32273 3648
brittling, contributes to the waste volume (table 5). This and graphite in the thermal column and
graphite-irradiation channels are the major contributors to the waste volume from fix-installed devices in the
pool.
Another major part of the waste comes from contaminated surfaces in the pool and the primary system
of the reactor. These are basically aluminum pipes which show a rather uniform contamination throughout the
whole primary system. Measurements by shielded gamma-spectroscopy on-site in which background gamma
rays were subtracted, showed a contamination of inner surfaces of the primary system of about 170 Bq cm-2. This
consists of 100 Bq cm-2 124Sb, 30 Bq cm-2 46Sc and 40 Bq cm-2 60Co. While 46Sc and 124Sb have rather short half-
lives of 83 and 60 days respectively, the half-life of 60Co is too long for adequate decay before removal of the
component. After a decay of 14 months (until transport of spent fuel elements) the activity of 124Sb will have
decreased to 2.5 Bq cm-2 and that of 46Sc to 0.2 Bq cm-2, together less than the limit for laboratory surfaces, but
60
Co will have decayed only to 34 Bq cm-2. Solutions for this rather voluminous part of waste (roughly 18 t
aluminum) are currently under investigation.

Table 6 Estimated waste amount and surface areas of materials from the primary system

Material Mass [kg] Surface [m²]

aluminum 17942 1622.05
stainless steel 8212 905.18
steel and cast iron 4832 82.12
PVC 154 27.63
PE 185 44.80
polyester 22 0.47
sum 31347 2682.25

The second largest fraction of the surface contaminated waste is the heat exchanger built of stainless
steel. It contributes about 8 t weight and since its surface amounts to about 34 % of the total inner surface of the
primary loop, it is expected to have about 34 % of the surface activity.
The last feature at the end of the decommissioning usually concerns the biological shield. It consists of

5
P-5-307

baryte concrete and a very dense iron framework. An exact measurement of the activity concentration in the
concrete shield and its depth penetration is only possible when all the fuel elements and the major part of
radioactive material have been removed from the pool and the pool water drained. However, an estimate of the
activity concentration in the baryte concrete was performed by taking small concrete samples of the baryte
concrete at the outside of the biological shield and activating them before final shutdown. In this way the main
activation products in the concrete were determined.
Results are given in table 7. They represent the data determined 83 days after sample irradiation,
characteristic for a shutdown time of 83 days, and after a cooling period of 3 years, a period typical for the
removal of the main radioactive components and possible demolition of the concrete block. The derived results
correspond to a content of 40 % Ba in the concrete which agrees with typical Ba-concentrations in biological
shields. It also
Table 7 Average activity inventory in concrete samples of the biological shield

Radio- Half-life Radiation, activity [kBq/g]

nuclide γ-energy [MeV] decay time: 83 days decay time: 3 years
46
Sc 83.8 d β, γ (0.8, 1.12) 2.2 0.0
54
Mn 312.2 d ε,γ (0.83) 0.2 0.0
59
Fe 44.5 d β, γ (1.09,1.29) 2.9 0.0
60
Co 5.272 y β, γ (1.17, 1.33) 0.3 0.2
131
Ba 11.5 d ε,β+, γ (0.37, 0.49) 8.7 0.0
133
Ba 10.5 y ε, γ (0.3, 0.36) 2.7 2.2
152
Eu 12.7 y ε,β+, γ (0.34, 1.41) 0.2 0.1
agrees well with measurements on bore hole samples taken at the FMRB of Braunschweig, a reactor of 1 MW
power level (7), when upgraded to 9.5 MW power level and the neutron flux at the inner side of the pool liner is
integrated over the life-time of the reactor. From measurements at the FMRB (7) a penetration of thermal
neutrons into the concrete shield by an exponential function e - 0.05 d was assumed. This results in a penetration
depth of ~ 1 m if a an exemption limit of 500 Bq/kg is assumed.
In table 1 - 6 the waste volume and the maximum activity concentration of each waste category are
described. From these data a balance of the average activity concentrations in each waste group and component
type was derived. Summing over all these component categories gives the total amount of waste and the
approximate activity arising. This was grouped according to the usual waste categories of high-, medium- and
low-level waste in table 8.
As known from previous dismantling experience, the by far greatest activity is contained in 300 kg
material (spent fuel elements). After the transport of these spent fuel elements to the reprocessing and storage
facility, the remaining activity is about two orders of magnitude lower. Also here, almost 100 % of the activity is
contained in just about 320 kg of medium-level waste, while the largest part of the waste volume, 99.8 % of the
volume, contains only about 6 GBq of activity. The activity concentration in this later part of the waste is in most
parts very low, and the amount of waste produced will significantly depend on parameters for unrestricted use
given by the licensing body.

Table 8 Waste amount and activity inventory after 1 year decay time

type of waste amount [kg] activity [GBq] relevant nuclide

high-level 308 1 107 137
Cs
medium-level 320 2 105 60
Co
low-level
133
activated 101000 5 Ba
60
contaminated 60000 1 Co

DISMANTLING PROCEDURES AND STRATEGIES

For experimental devices, irradiation capsules and devices in most cases simple cutting and sawing
techniques will be applied before conditioning. There is long experience with cutting procedures with higher

6
P-5-307

radioactive substances in which the exposure of the staff never exceeded very low levels. If required, cutting is
performed under water to adequately shield the radioactive components. Also special shielded configurations
may be applied, e.g. for cutting of the beam tubes and beam tube inserts a special device for transport and
shielding is available by which already in the past several beam tube experiments were removed from their
position to a storage site. With the same device beam tubes were cut and removed from the pool later when these
irradiation facilities were replaced by irradiation facilities for silicon doping (4,5).
For some experimental devices cutting techniques with a special machine designed for cutting the head
and bottom of fuel elements before shipment to reprocessing may be used. As there are several options at the
ASTRA-reactor available, the optimum choice will be feasible according to the device to be removed.
By these techniques all experimental and irradiation devices as well as fixed installed systems such as
the grid plate, the plenum and the protective lead shields will be removed. For the later a precise cutting design
will be decided on at a later stage.
Dismantling procedures for the primary system are simple since the activity levels are low and consist
of only surface contamination at levels slightly above limits for laboratory surfaces. Further treatment of the
pipes, e.g. surface decontamination or bulk pressing, will be discussed further in the future to minimize the waste.
Procedures and techniques to dismantle the biological shield are evaluated at present. A decision on the
optimum procedure will not be taken before the time when such a decision is required. In about three years the
decision is pending and the most advanced techniques available at that time (8,9,10) will be applied.
According to IAEA recommendations (1) a three stage process in dismantling is foreseen. Of the
possible options an immediate dismantling to stage 1 of IAEA technical guide-lines (storage with surveillance)
after the final shipment of spent fuel and complete removal of high-level waste from the site, succeeded by an
immediately following, continued dismantling to stage 2 of these guide-lines (restricted site use) was identified
as the most reasonable and under present auspices optimum choice. The reasons are that the majority of
radionuclides possess either half-lives up to about 80 days which decay sufficiently to permit a continuing
dismantling after stage 1, or of half-lives so long that waiting periods of more than 50 years would be required to
substantially reduce exposure levels. Since these later nuclides show rather low activity levels, the handling of
most contaminated and slightly activated components may be performed without much complication at an early
stage.
CONCLUSIONS
To perform an environmentally safe and to the dismantling staff least exposing decommissioning of the
reactor, the possible options and the required stages for decommissioning and removal of the radioactive
components were evaluated. It was decided to follow the stages proposed by IAEA (1) with virtually no interval
between stage 1 and stage 2. Also stage 3 may follow immediately after stage 2 if adequately prepared.
The methods of choice for cutting and removal of the major radioactive components and devices are
more or less the same as deployed in the past with the removal of experimental facilities not required anymore.
A long-standing experience at the ASTRA-reactor is available for these techniques.
A preliminary evaluation of the arising radioactive waste was performed. The estimates amount to
approximately 320 kg of medium-level radioactive waste and about 60 t of contaminated and 100 t of activated
low level radioactive waste. The activity inventory is roughly 200 TBq in the medium-level waste and 6 GBq in
the low-level waste

REFERENCES
1. IAEA. Planning and Management for the Decommissioning of Research Reactors and Other Small
Nuclear Facilities, Technical Reports Series Nr. 351, Vienna (1993)
2. IAEA. Safety in Decommissioning of Research Reactors, Safety Series Nr.74, Vienna 1986
3. Safety Analysis Report for ASTRA-Reactor, Seibersdorf, Austria (1983)
4. K. Mueck. Verkürzungen der Strahlrohrkammern am ASTRA (in German). Report RBS-32 (1985)
5. J. Casta. Reparatur der Abschlußklappe/Ablauf und Verkürzung der Strahlrohrkammern G und H (in
German). Report RBS-41 of Research Center Seibersdorf (1996)
6. G. Erdtmann. Neutron Activation Tables, Verlag Chemie Weinheim, New York (1976)
7. W. Hajek. Bericht über Kernbohrungen am biologischen Schild des FMRB, PTB 6.301 (1998)
8. W. Hajek. Dismantling and Decommissioning Technologies and Techniques for Research Reactors, 25th
anniversary WM99, 28 Feb. - 4 March 1999, Tucson, Arizona
9. H. Kistmacher, H. Haferkamp, D. Seebaum, A. Feld, Fr.-W. Bach. Einsatzmöglichkeiten des Lasers in
der Stillegungstechnik, Stillegungskolloquium Hannover, Juni 1997
10. IAEA. Decommissioning Techniques for Research Reactors, Technical report series No. 373, Vienna
(1994)