Skip to main content

Ads and fast Reactor CompArison Study in support of Strategic Research Agenda of SNETP

Final Report Summary - ARCAS (Ads and fast Reactor CompArison Study in support of Strategic Research Agenda of SNETP.)

Executive Summary:

For a PATEROS scenario, where a number of European countries will stagnate or phase out the use of nuclear power (Group A) and another group of countries will continue nuclear and develop Generation IV Fast Neutron Systems, the Minor Actinide stream to be transmuted (and a representative isotopic vector) has been established: reasonable minimum of 2.3 tons/year and reasonable maximum of 6.5 tons/year.
To transmute Minor Actinides in a critical Fast Reactor, two options can be envisaged: homogeneous loading and heterogeneous loading. In the former case, the minor actinides are added to the classical MOX driver fuel of the reactor. In the latter case, dedicated targets are loaded at dedicated positions. Due to the severe impact of minor actinide loading on safety parameters, the amount of Minor Actinides to be loaded in the core is limited. Based on the CP-ESFR design, scoping calculations were done based on a good estimate for these limits. For the homogeneous option, a loading of 2.5% Minor Actinides results in a transmutation rate of 2.8 kg of Minor Actinides per TWhth produced. For the heterogeneous case, a blanket of 84 target assemblies each loaded with 20% Minor Actinides results in a transmutation rate of 0.6 kg/TWhth.
The reference Accelerator Driven System used is the EFIT-400 design (400 MWth) which has an inert matrix fuel (either CERMET or CERCER) with a ratio of about 45% Pu/(Pu+MA). Transmutation studies have shown that this system can transmute about 35 kg/TWhth. Safety studies showed that the global coolant voiding effect is negative but that local coolant voiding could be positive albeit quite small. However, the effect is smaller than the subcriticality margin; hence no run-a-way accident is possible. A study based on the PATEROS scenario mentioned above showed that for the complete transmutation of all Group A Minor Actinide Stock, 7 EFITs are needed. In case of total elimination of the Minor Actinides in Group B as well, 20 units are needed.
Both fuel reprocessing and fuel fabrication have been looked at. Classical LWR fuel reprocessing technology is mature. Fabrication of MOX fast reactor fuel based on powder technology is also mature. However blending reasonable quantities of Minor Actinides in this fuel has not been done industrially, only at lab scale. Fabrication of Minor Actinide targets and ADS fuel having huge quantities of Minor Actinides requires still a lot of research. Especially the presence of Curium will pose shielding and heat problems. Since not all transmutation take place in one reactor/ADS cycle, this fuel needs reprocessing as well. Again here the technology is still very immature. As a consequence costs are very hard to estimate for these technologies.
Two cost models have been applied to estimate the costs of these new fuel cycles and their ingredients: a home-made model by CNRS and the G4ECONS model applied by NRG. The estimations were within a range of 10-12% of each other. The estimates showed that the cost per MWe produced is 4 times higher for an EFIT than for an SFR. However, because of the very high transmutation capacities of EFIT, much less of these installations are needed. Based on a "zero Minor Actinide balance" hypothesis, four scenarios were analyzed. First, a single stratum case where Fast Reactors take all the electricity production. Then, a double stratum where in the first street, electricity is produced with LWR (70% share) and Minor Actinides are transmuted in a Fast Reactor (30% share). Then a double stratum is analyzed using ADSs to transmute the Minor Actinides. In this case, the LWRs take 97% of the electricity production. Combining these ratios with the costs estimated showed that the scenario with the ADS double stratum was the cheapest (72.3€/MWhe) and the scenario single stratum with SFR was the most expensive (94.3€/MWhe).

Project Context and Objectives:

In this project, we would like to compare from a technical and economic point of view the implementation of either ADS’s or dedicated Fast Reactors as minor actinide burning facilities. The approach is to start from two fixed hypotheses: (1) we work in the heterogeneous approach and look only at the “burning” stratum; (2) we assume a certain influx of minor actinide mass per year that needs to be burned. PATEROS results are used for the basis of this comparison avoiding numerous extensive scenario studies. Findings and conclusions of previous partitioning and transmutation research will be analyzed and taken into account in this project.

To have a pertinent comparison, TRU (Pu + minor actinide) mass and composition have to be reasonably estimated for a representative region (a number of European countries). For the double stratum case, scenario 2 from PATEROS is used to assess the ADS or dedicated burners' fractions. For the heterogeneous approach in a single stratum, scenario 3 from PATEROS will be used to assess the number of dedicated MA target assemblies.

The economic impact of both options should be evaluated for both investment cost as well as operational cost but not the needed R&D cost. The project will also consider technological maturity (as for example using the NASA Technological Readiness Level) and how this can be incorporated in the economic analysis. The associated fuel cycle cost will be evaluated. For the double stratum (reactor system + fuel reprocessing plant + dedicated MA-bearing fuel fabrication) one can assume that the units are located near to each other, to limit transportation cost.

A crucial parameter to be established for both reactor systems is the maximal minor actinide (MA) content in a core loading. This maximal MA value is determined by operational safety criteria to be adhered by the dedicated burner. It is well-known that the delayed neutron fraction as well as the Doppler effect significantly decrease with increasing minor actinide content. An evaluation of a number of safety parameters for the systems will give an upper boundary for the minor actinide mass present in the core. Since for the double strata option, only ADS systems were studied in scenario 2 of PATEROS, there is a need to study the equivalent critical systems.

Since man power is limited in this project, we have attempted to use as much pre-existing information as possible. Especially for the selection of the Fast Reactor system and Accelerator Driven System, we have relied on information from past and current European Framework Programs. For the Fast Reactor system, we have adopted the Sodium Fast Reactor design from the project CP-ESFR. This Sodium Fast Reactor is a 3600 MWth reactor system. For the Accelerator Driven System, we have chosen the EFIT-400 design from the 6th Framework program IP-EUROTRANS. It is a lead-based system driven by a proton accelerator. For more design details, we refer to deliverables 2 and 3 of this project for a short overview and the references therein that point to more detailed information.

The purpose of Work Package 4 was to identify the cost differences associated with the fabrication and reprocessing of FR and ADS fuel. Different fuel fabrication processes can be considered: Powder Metallurgy and Sol Gel.
Separate MA and U/Pu production lines have also been proposed to minimise the impact of infrastructure cost directly associated with the management of MAs. In the case of homogeneous fuel fabrication, the costs are likely to lower since there will be one single process line in comparison to the expected multiple lines for a heterogeneous option.

Arguments will be presented for the reprocessing technology selection. Options are shown to be driven by the chemistry of the fuel and the demands of the fuel cycle, in particular cool down rather than transmutation system or the transmutation scenario; single or double strata.

Pyrochemical reprocessing would be selected when the fuel is insoluble in nitric acid/organic systems or insoluble products might form during aqueous reprocessing; typically, those fuels containing Zr, Y or Mo. Where fuels do not contain these elements and the fuel cycle allows, aqueous reprocessing options would be selected.

Irrespective of the reprocessing or fuel fabrication technologies chosen or identified, it has been argued that the major contributor to cost will be:

• The size of the facility;
• Number of process lines;
• Number of overall unit operations in the selected process, and;
• Number of remote unit operations and complexity to manage MAs.

The choice of technology is assumed to have a relatively small impact on cost.
However, all fuel fabrication processes, with the exception of U and MOX fuel are technically immature and Technology Readiness Levels will be assigned accordingly. This will have a significant impact on the Research & Development effort needed to deploy these technologies.
As these technologies are immature, one expects difficulties in a good cost estimate.

In Work Package 5, a cost estimate was done using two models, one house-made model by CNRS and the G4ECONS model applied by NRG. These costs include the construction, operation & maintenance costs and fuel cycle costs. Research & Development costs will not been considered. Also the transportation issue of "fresh" transmutation fuel and spent transmutation fuel will be analyzed both from the technical side and the cost side.

Project Results:
1. WP 1: reference scenario definition.

The reference scenario considered in the frame of the ARCAS project refers to PATEROS project [1] where a regional scenario, at a European level, was analysed in detail. Scenario 1 was taken into consideration, in which spent UOX and MOX fuel discharged from LWR is reprocessed (mono-recycled) in order to separate TRU from fission products (which, together with reprocessing losses, are sent to a geological repository). Reprocessed Pu and MA are recycled in the regional transmuter facility, which in this case is the ADS-EFIT (Accelerator Driven System – European Facility for Industrial Transmutation) [2] and blended with TRU separated from spent fuel of subcritical transmuter fuel cycle (as soon as available) in subsequent cycle passes (Figure 1).
The final goals of the scenario are:

a) to fully reprocess spent fuel legacy of some European countries (Group A), which are supposed to be in a stagnant or phasing-out scenario: Belgium, Czech Republic, Germany, Spain, Sweden and Switzerland, in order to eliminate all the TRU stocks, before the end of the present century;

b) to store Pu (for a possible future use for the deployment of fast reactors, which were not simulated in this case) and to stabilize the MA inventory in European countries (Group B) pursuing nuclear energy generation: France was considered in this case.

(Fig 1. PATEROS simplified flow scheme; see .pdf file 'Description of main S&T results-foregrounds' enclosed).

According to scenario assumptions ADS-EFIT will be deployed in a regional centre starting from 2045 (this hypothesis should appear rather unrealistic) up to 2090 – then a constant energy production level – i.e. number of transmuters - is assumed, regional fuel cycle facilities such as reprocessing and fuel fabrication plants for innovative fast reactor fuel, and a spent fuel (SF) interim storage are considered. In particular a reprocessing capacity of 850 tonnes/year was assumed for Group A reprocessing plant, while 1700 tonnes/year were necessary in order to stabilize the inventory of Group B.
Calculations were performed with COSI6 – ver. 6.0.1 a code developed by CEA (Cadarache)[3]. In particular simulations addressed the MA streams (and their isotopic composition) evaluation from Group A (i.e. coming from a spent fuel storage after some decay time) and Group B (i.e. coming from a continuous feed from a PWR fleet with a 50,000 MWd/t burn up, fuelled by 90% UOX and 10% MOX, a 5 years cooling time and a total yearly energy production of 430 TWhe). The outcome of the simulations is shown in Table 1: MA composition and a range of minimum and maximum annual values are indicated. It should be noted that the hypothesis that all European nations except France will phase-out during the present century may appear too optimistic from the MA waste stream amount point of view. If we consider that the nuclear power installed in France is today 63,130 MWe, while the total for Europe is 169,932 MWe [4], a factor of ca. 2.7 should be considered (as it appears that a phase-out of nuclear energy in the near future in OECD countries is unlikely, due also to environmental concerns about global warming, as stated by IPCC or IIASA scenarios [5][6]).
(Table 1. Proposed reference MA composition; ; see .pdf file 'Description of main S&T results-foregrounds' enclosed).

It is important to stress however that the composition indicated above is just indicative: in order to adopt an efficient transmutation strategy in fact it is mandatory that fast systems (both critical and subcritical) run a closed cycle, by reprocessing their own fuel and recycling it in their fuel fabrication plants as first choice. This strategy will present some relevant consequences, which should in principle affect fuel cycle costs heavily:
Recycling of transmuters fuel, and then blending it with fresh fissile material in order to balance the fissioned mass, will obviously modify fuel isotopic vector at every new reactor load: this fact will require probably to modify fuel shares (e.g. inert matrix/heavy metal, amount of uranium, etc.), which, in their turn should affect safety coefficients, performance, burning capacity, etc.;
Recycling fuel in a closed cycle fashion will cause probably an accumulation of heavy elements, thus increasing fuel gamma and neutronic emission, such as also decay heat power: as new technologies, making use probably of remote handling and improved shielding issues, should be required this parameters should be taken into consideration accurately in costs evaluation.
Finally it should be taken into consideration in transmuters evaluation that if nations with a phasing-out policy are considered, as in case of PATEROS scenario, plutonium management should be an issue, especially if adopted transmuters are not specifically designed for this goal. Simulations show that if maximum MA stream cited above is adopted (i.e. present European nuclear fleet) a plutonium annual stream of ca. 24 tonnes/year results, which means an accumulation of more than 2100 tonnes of fissile material by the end of the present century. It should be pointed out also that small nations that decide to continue to produce nuclear energy without planning to use produced plutonium in fast reactors (such as France) will have to adopt a proper strategy for its final disposal (regional transmuter design should take this issue properly into account).

2. WP2: Definition of the Fast Reactor System.

In the frame of the EU CP-ESFR project [7], a basic SFR concept was proposed as a 'Working Horse' (WH) design, which was further optimized in an effort to improve the original reactivity coefficients. A short description of the optimized reactor concept is provided in this section, which will be used for analytical estimations in ARCAS. The cross section of the core is depicted in Figure 2.

The 3600 MWth core is composed of 225 sub-assemblies (S/A) in the inner core and 228 S/A in the outer core; 453 S/A in the whole reactor.

The core S/A are MOX type, where the composition is as follows (in weight percentage of the total Heavy Metal):

Inner core, 8 S/A active rows: 85.12% Depleted Uranium, 14.76% Pu, and 0.13% Am (as a result of Pu decay during fabricated fuel storage).

Outer core, 4 S/A active rows: 82.72% Depleted Uranium, 17.15% Pu, and 0.12% Am.

(Fig. 2. Cross section view of CP-ESFR core; ; see .pdf file 'Description of main S&T results-foregrounds' enclosed)

The core optimization is called CONF2 case and the features are:

A lower axial blanket made of depleted Uranium dioxide.

An upper Sodium plenum to enhance neutron leakage in the region in case of plenum voiding.

An upper neutron absorber layer, above the Sodium plenum.

Before the optimization process, the basic ‘Working Horse’ design consisted only in the active length, 1 m high, with no lower axial blanket, sodium plenum and upper absorbing layer. The WH design was intended to be a break-even core.
It is found that MA homogeneous loadings in the reference reactor lead to moderate transmutation values, up to 6.9 kg/TWhth for 4%w loading, and noticeably deteriorates reactivity coefficients (Doppler constant and core void worth). However, the deterioration depends very much on the exact core configuration. Hence, dedicated core design strategies for lowering the MA impact have an important effect for the Doppler constant, whose deterioration may decrease from 40 to 15%. The reduced void worth deterioration is found similar before and after application the optimization guidelines (some 25% in both cases). On the other hand, the extended void worth significantly decreases compared to the reduced void worth, which means that optimization guidelines are promising and should be further pursued, even targeting negative core void worth. In the meanwhile, a combination of lower MA loading, 2.5%, and optimization guidelines seem to be a promising concept, as deterioration will be lower.
Heterogeneous blanket configurations lead in general to low deterioration of safety parameters or even to little improvement when core optimization guidelines are considered. However, concerning transmutation values, virtually no net transmutation is found or just a small net value after optimization guidelines. An interesting case between homogeneous and heterogeneous has been also presented with MA loading in the outer core (together with Pu) leading to medium transmutation values (3.6 kg/TWhth) and no deterioration of reactivity coefficients.
In any case, the impact of MA loading on reactivity coefficients expand over a range of results, which illustrates the necessity for rigorous safety analysis in order to advance the issue of the core feasibility from the point of view of licensing. This is indeed an open field for research, as no fully dynamic safety analysis is yet available. Concerning first European scenario analysis, it is obtained that medium transmutation values (as of 2.8 kg/TWhth) in all reactors could lead to elimination of the neptunium and americium stock, out of the reactor site, in the frame of a century, and that the result is compatible with Pu breeding in all reactors. The curium mass stock, however, is not eliminated but just somewhat decreased. Such an objective would imply the fabrication of a very large number of MA bearing fuel assemblies at low contents, 2.5%w. Also, the Pu amount involved in fuel fabrication would be very large.

3. WP3: Definition of the Accelerator Driven System.

The aim of Work Package was to select and characterize the reference Accelerator Driven System to be used in the ARCAS project. As, in Europe, there is only one design for an industrial transmutation facility available, the choice of the reference system was rather easy: the EFIT, European Facility for Industrial Transmutation, as designed in the 6th European Framework programme IP-EUROTRANS [8].
The accelerator foreseen in the EFIT design is an 800 MeV proton accelerator delivering 20 mA of current. This beam impinges on a windowless spallation target, where the induced spallation reactions produced the required source neutrons. The 19 central positions of the hexagonal core lattice house the spallation target which is surrounded by fuel assemblies. The number of fuel assemblies is such that the core, by design, will not become critical (even in accidental conditions). The reactor core is cooled by pure lead (as opposed to lead-bismuth eutectic as foreseen in the experimental facility XT-ADS). This allows a high inlet and outlet temperature (400°C and 480°C respectively) and as a consequence a rather high thermodynamic efficiency of 40% .
For the fuel one opts for uranium-free fuel since this avoids extra build-up of plutonium (by capture in U-238). Because there is a relationship between the energy produced and the material destroyed by fission (one fission produces about 200 MeV of energy), the final balance is always a loss of 42 kg/TWhth [9]. The design goal of EFIT was hence to have a loss of minor actinides as close to 42 kg/TWhth as possible and a loss of plutonium close to 0 kg/TWhth [10][11]. The second goal for the design of the core was to have a reactivity swing as close to zero as possible, reducing the power fluctuations during the cycle without the need to compensate for this using the proton accelerator.

Two types of advanced fuels have been analysed in the EUROTRANS project: the CERCER option and the CERMET option. The former uses a MgO matrix, the latter a Mo matrix. For the CERMET, two sub-options have been analysed: a matrix with natural Mo and a matrix enriched in the lighter isotopes of Mo, avoiding excessive neutron capture.
The ADS EFIT core used for this study is based on the one defined in the deliverables D3.2 and D3.3 from the AFTRA (Advanced Fuels for TRAnsmutation systems) domain (DM3) within the EURATOM FP6 integrated project (IP) EUROTRANS [12][13]. For this comparative study, we used EFIT-400 (AFTRA) core with one zone configuration, a thermal power of 400 MW and with the two selected fuels for EFIT core: CERCER with MgO matrix and CERMET with Mo matrix enriched in 92Mo. As shown in Figure 3, the core contains 6 rings of fuel assemblies (FAs), surrounded by 2 rings of reflector ones and a cylindrical core barrel with 30 mm as thickness. The spallation target and the surrounding region (containing mainly lead) occupy the space created by the withdrawal of 19 fuel assemblies from the central region proposed in the EFIT reference design [14].

(Fig. 3. Single-zone reference core model of the EFIT-400: radial layout; see .pdf file 'Description of main S&T results-foregrounds' enclosed).

The results of MA transmutation rate are calculated using the ALEPH code (SCK•CEN home-made code) [15]. The ALEPH code is designed to combine a Monte Carlo codes (MCNP or MCNPX) for spectral calculations with a modified version of ORIGEN-2.2 code [16] for evolution calculation. The nuclear data used are based on the JEFF-3.1 library [17].
As shown in Table 2 and Table 3, in terms of total MA transmutation rate, for both ARCAS and EFIT vectors, the transmutation performances are the same: reaching values of 39kg/TWh and 36kg/TWh for EFIT-400 CERCER fuel and EFIT-400 CERMET fuel respectively.

Table 2. MA Transmutation rates (kg/TWh) for EFIT-400 with CERCER fuel; see .pdf file 'Description of main S&T results-foregrounds' enclosed).
Table 3. MA Transmutation rates (kg/TWh) for EFIT-400 with CERMET fuel; see .pdf file 'Description of main S&T results-foregrounds' enclosed).

WP4: Definition of the fuel reprocessing and fuel fabrication facilities.

The objective of ARCAS WP4 is to define a fuel fabrication plant and a reprocessing plant for fast reactor (FR) and ADS fuels in order to compare costs. Clearly, these plants may be different for each neutron system, but that difference will only depend on the fuel types and their irradiation conditions.
The output from WP4 comprises baseline information which identifies process differences upon which an economic assessment of heterogeneous fuel fabrication plant and spent nuclear fuel reprocessing plant for FR and ADS can be made in WP5. Inert matrices of Mo or MgO (Yttria Stabilised Zirconia) and pure MgO have been selected as model fuels for ADS and FR systems respectively. The case for reprocessing of ADS fuel using pyrochemical technology and the fabrication of ADS fuel using Sol Gel is made and appropriate high level case studies completed. Similarly, the case for reprocessing FR fuel using aqueous technology and the use of powder metallurgy as the preferred fuel fabrication route for UO2 blanket and U/Pu oxide core fuel is made. Sol Gel is the preferred route for minor actinide (MA) fuel fabrication.
Reprocessing options are expected to fall into two “camps”.
These are:

Materials well suited to existing fuel fabrication processes and compatible with nitric acid/organic phase, PUREX or GANEX type, separation processes and;

Those that are not where non-aqueous process routes (i.e. pyrochemical) are most likely to be deployed.

The following assumptions are made for the fast reactor fuel

Heterogeneous actinide and MA fuel pins are U/Pu and inert matrix (MgO)/MA;

Based on an oxide system;

High Pu content in FR core;

MgO is soluble and easily diverted within an aqueous reprocessing option, therefore only aqueous reprocessing options are required for this scenario;

Oxidation and dissolution of high Pu content fuel and MA is possible;

Aqueous reprocessing solvents are sufficiently stable to very high burn-up fuel, however, in extreme cases, the effect of radioactive content in aqueous reprocessing can reduce the effect of solvent extraction dramatically. This is especially true in the first stages of these processes where the organic solvent is in contact with fission product activity in the aqueous solution;

An organic phase clean-up and recovery step will be included in any reprocessing scheme to maintain process efficiency, and;

Aqueous processes will be assessed in conjunction with an appropriate scenario (e.g. sufficient cool-down time).

and for the Accelerator Driven System fuel

Fuels are based on an actinide oxide dispersed in an, inert matrix;

Heterogeneous actinide and MA fuel pins – Mo or MgO/PuO2 and Mo or MgO/MAO2;

Provided Mo is recovered and recycled at an early stage of reprocessing, then aqueous processes should be considered viable, if not;

Pyro-processing due to the potential issues of CaesiumPhosphoMolybdate (CPM) and insoluble product formation in aqueous processes is proposed;

Pyro-processing is used for spent fuel when fabrication techniques have required a ZrO2 or Yttria Stabilised Zirconia (YSZ) in the MgO matrix;

Where aqueous reprocessing is selected, the assumptions shown under FRs above also apply here ;

All fuels will be subject to very high burn-up.

Carbide and nitride systems have been discounted from this work due to their very low technological readiness levels in comparison to oxide and metal fuels.
Both aqueous and non-aqueous process steps have been investigated and various conclusions drawn. Two baseline processes for fuel fabrication have been considered Powder Metallurgy and Sol Gel. MOX fuel production is, of course, based on powder metallurgy and is well established at industrial scale however, the production of separate MA oxide fuels is not.
Preparation and production of heterogeneous oxide fuels for the double strata advanced fuel cycle using either FR or ADS is extremely challenging due to high alpha, high decay heat, high neutron emission and high gamma activity. Any fuel processing facility will therefore necessarily need to have very high integrity containment to prevent the spread of highly mobile alpha activity, include heavy shielding for the penetrating radiation and almost certainly require the deployment of remote engineering technology for some plant operations / plant maintenance purposes. No preferred technology for fuel fabrication was identified due to the low Technology Readiness Levels, however, with dedicated production lines for each fuel type, the technology of choice can be selected when suitable technological maturity is obtained. Costs were expected to be closely related to the number of unit operations rather than technology selection.
It should be noted, the ARCAS study has been bounded to include Cm heterogeneous targets, and therefore fuel fabrication plants are required to include heavy neutron shielding. Should the decision be taken to sentence Cm to a dedicated decay store, then shielding requirements become less demanding for fuel fabrication. This scenario however, is outside the scope of ARCAS.
For reprocessing plants, an analysis of the different unit operations for both aqueous and pyrochemical options was completed. A basic gap analysis highlighted the technical immaturity of both technology options and, as expected, they were found to have very low TRLs of 2-3.
All fuel fabrication processes, with the exception of U and MOX fuel are technically immature and assigned low Technology Readiness Levels of 2-3. Costs were expected to be related to the number of unit operations, shielding requirements and remote technology deployment rather than the technology selection, per se.

5. WP5: Economical comparison.
The last work package of the ARCAS project is to gather all information from the other work packages in order to able to present a comparison between the two options of fast reactors or accelerator driven systems. An economic analysis and a business case description are being prepared for the EFR and EFIT nuclear plant designs with transmutation capabilities.
Two methods are used to determine the cost structures for the two options. One is to calculate the cost per kilowatt-hour electric and per metric ton actinide waste destroyed for each design separately. The GIF tool G4Econs is being used for this. The other method is to calculate these costs for certain defined scenarios of reactor parks. Three scenarios are being considered: only Fast Reactors with heterogeneous targets, double strata with Fast Reactor burners, and double strata with ADS transmuters. (The double-strata nuclear fuel cycle consists of the commercial reactor fuel cycle (the 1st stratum cycle) and nuclear transmutation fuel cycle (the 2nd stratum cycle) based on FR or ADS that transmute the minor actinides generated in the 1st stratum.)
With the number of units needed per GWe of LWR installed and the investment cost of a transmutation unit, the investment cost per GWe is determined. For selected nuclear evolution scenarios, the total investment cost needed for transmutation can be determined. Also, the total generating costs are compared, giving an answer to the question on how much the MA transmutation would add to the cost of kWh. These costs would include both the investment, operational and fuel cycle costs. The fuel cycle costs consist of all the parts of the closed cycle, including reprocessing and fuel/target fabrication.
Both FR and ADS have transmutation capabilities. As expected from their fuel loadings and spectra, the project work packages 2 and 3 have demonstrated that ADS have very superior capability for transmutation compared to FR. Also the required transportation of nuclear spent fuel and dedicated burner fuel can be limited because of the high concentration of minor actinides in ADS fuel. The challenging question is whether these advantages could compensate for the extra difficulties and then costs of building these facilities. Table 4 shows the cost advantages and disadvantages for the three reactor systems considered. Given the extra complexity of its design (need for a reliable powerful accelerator), ADS most probably have a higher LCOE (€/MWhe) than FR, who in turn have a higher LCOE than LWR. If transmutation is not needed, if Pu is not managed separately, then utilities using LWR have no incentive to pay the extra costs of MA transmutation and LWR are by far the best and probably only choice. If however utilities would be obliged legally to manage their heavy nuclides and in particular the remaining MA after Pu removal, then a market could emerge for transmutation. Calculating the electricity costs for a nuclear park consisting of LWR and transmutation facilities, the higher costs of electricity produced by ADS may then be balanced by its limited share in the energy mix and the bigger share of lower cost kWh produced by LWR.

Table 4. Comparison of reactor costs of both models (CNRS and NRG)
(see .pdf file 'Description of main S&T results-foregrounds' enclosed).

Here it should be noted, that particularly ADS is not designed for electricity production. Nevertheless electricity cost is the only parameter allowing for a clear comparison between the different scenarios. Stated differently, one can consider the extra electricity cost for MA burning as the ‘price’ to be paid for minor actinide recycling and transmutation. The over-cost can then be viewed within the advantage of added sustainability of the closed fuel cycle, that recycles all its minor actinides, as well as from the viewpoint of reduction of long-lived nuclear waste.
In the case of ADS, although the reactor costs and, subsequently electricity production costs, are very high, the efficiency of burning minor actinides is as well. In the scenarios this leads to a small fraction of ADS needed. The high electricity costs for ADS are then compensated by a large share of LWR in the first stratum with relatively low electricity costs.
In Table 5, the reactor system costs from Table 4 are combined with the scenario results on installed capacity in order to compare the scenario electricity costs. From this table, we can conclude that the influence of the first stratum (i.e. the use of EPR) equalises the total costs. All scenarios are comparable in terms of costs, particularly in view of the large uncertainties. Comparing with a single stratum and the use of only EPR’s, so without recycling of minor actinides, the double stratum scenarios typically add 15 – 30% to the costs. One can view this as the ‘price’ of increased sustainability of a fully closed fuel cycle and a significant reduction of the long-lived radioactive waste. A single stratum with only fast reactors is more expensive.

Table 5: Comparison of scenario electricity costs (see .pdf file 'Description of main S&T results-foregrounds' enclosed).

A more detailed follow-up study should be conducted to address in more detail the differences in the various models and scenarios. Follow-up study can also help in fine-tuning the costs, and reduce uncertainties.
As a last point for a follow-up study, plutonium management can be addressed in more detail. Within ARCAS, plutonium management is not a specific goal. Nevertheless it is important when considering a fully closed and sustainable fuel cycle with recycling of all transuranic elements.
Next, the first steps into a business plan for a transmutation facility for long-lived nuclear waste have been investigated. As no such facility is yet under construction or even planned, a detailed business plan, even preliminary, is not possible, but some conclusions could already be drawn in this stage. Most important of these is that no transmutation of long-lived nuclear waste will take place without government obligation and legislation. The existence and direction of a country nuclear program is important. If this direction is into long-term use of nuclear energy, fast reactors will probably be part of the program for reasons of fuel security of supply. The fast reactor technology developed for this program may be used than for transmutation purposes as well, causing ADS not to be used in this country. The use of ADS technology would then be limited to those countries phasing out nuclear energy or using it temporarily as a ‘bridging’ technology. It remains to be seen whether these countries would make the investment for development of ADS technology to an industrial scale for the limited amount of minor actinides to be transmuted, also since the ADS facility will not eliminate the need for a final repository of highly radioactive nuclear waste.

The business strategy investigated here would be that electric utilities owning the nuclear plants will pay a fee for the transmutation of their minor actinide waste, and will add this to the electricity price. This does not exclude other strategies, e.g. sell the reprocessed plutonium and pay the minor actinide transmutation with this.

As there are no industrial transmutation facilities planned yet, three projects with some relationship with the transmutation issue have been investigated: 1) MYRRHA, an ADS research facility for actinide transmutation; 2) Pallas, a dual-purpose thermal research reactor for medical isotope production and nuclear materials research, including transmutation fuels; and 3) PRISM, a fast reactor nuclear power plant for plutonium destruction and electricity production. Still other projects could be investigated in this way, like the French Jules Horowitz research reactor and the French fast reactor research facility ASTRID. Although the three projects already exist for several years and are in various stages of development, financing is still an issue. For MYRRHA, the Belgian government has offered a partial financing, and for the other two projects all financing should come from the market.

The dependence of the economic performance of a transmutation facility from the electricity price has been investigated. If the electricity price is low, the economic performance of ADS-EFIT and EFR are comparable only for very good EFR transmutation performances, while for high electricity prices EFR is more convenient than ADS-EFIT. In case standard values are considered there is no net economical convenience in the adoption of one particular system.

When looking at the costs of electricity nuclear power plant fleets including FR and ADS respectively, the results of the comparison of these costs depend strongly on their relative costs. The increased costs of electricity produced by ADS may be balanced by its limited share in the energy mix and the bigger share of lower cost kWh produced by LWR. The discussion about the break-even price of ADS that makes the ADS scenario more competitive than the FR scenario is very difficult. Given the very low levels of readiness of most of the technologies involved in this study, whether for the reactor but also for the fuel cycle, the cost models cannot be expected to be very representative of future technology costs.


[1] PATEROS, PArtitioning and Transmutation European ROadmap for Sustainable nuclear energy, European Commission - Sixth Framework Programme.

[2] M. Sarotto, C. Artioli, and V. Peluso, "Preliminary Neutronic Analysis of the three zones EFIT-MgO/Pb core," ENEA, FPN-P9EH-002, 2007.

[3] L. Boucher and et al., "COSI: the complete renewal of the simulation software for the fuel cycle analysis," in ICONE 14, Miami, Florida, US, 2006.

[4] European Nuclear Society. (Nov 2010) Euronuclear. [Online].

[5] IIASA, Global Energy Perspectives., 1998.

[6] IPCC, Special report on emissions scenarios., 2000.

[7] L. Fiorini and A. Vasile, "European Commission – 7th Framework Programme: The Collaborative Project on European Sodium Fast Reactor (CPESFR)," Nuclear Engineering and Design, vol. 241, no. 9, pp. 3461–3469, 2011.

[8] Sixth Framework Programme of the European Commission, EUROTRANS: EUROpean programme for the TRANSmutation of high level nuclear waste in an accelerator driven system, 2005, FI6W-CT-2004-516520, EURATOM FP6.

[9] Carlo Artioli, Giacomo Grasso, and Carlo Petrovich, "A new paradigm for core design aimed at the sustainability of nuclear energy: The solution of the extended equilibrium state," Annals of Nuclear Energy, vol. 37, no. 7, pp. 915-922, 2010.

[10] Carlo Artioli, Hamid Aït Abderrahim, Georgios Glinatsis, Carlo Petrovich, and Michael Schikorr, "Optimization of the minor actinide transmutation in the ADS: the European Facility for Industrial Transmutation EFIT-Pb concept," in Proc. of AccApp'07, Pocatello, 2007.

[11] Carlo Artioli et al., "Minor actinide transmutation in ADS: the EFIT core design," in PHYSOR2008, Interlaken, Switzerland, 2008.

[12] V. Sobolev and et al., "Neutronic performances of EFIT-400 (AFTRA) reference core at the beginning of operation," FP6 IP-EUROTRANS FI6W-CT-2004-516520 D3.2 (rev.2) 2006.

[13] A. Rineiski and et al., "Comparison of the neutronic performances of the ETD reference core loaded with the selected TRU-fuels from BOL to EOL," Deliverable IP-EUROTRANS FI6W-CT-2004-516520 D3.3 2010.

[14] G. Corsini and M. Petrazzini, "Main components functional sizing of EFIT," Deliverable IP-EUROTRANS FI6W-CT-2004-516520 D1.24 2006.

[15] Wim Haeck and Bernard Verboomen, "An optimum approach to Monte Carlo burn-up," Nuclear Science and Engineering, vol. 156, no. 2, pp. 180-196, 2007.

[16] A.G. Croff, "ORIGEN2: a versatile computer code for calculating the nuclide composition and characteristics of nuclear materials," Nuclear Technology, vol. 62, 1983.

[17] A. Santamaria et al., "The JEFF-3.1.1 Nuclear Data Library - JEFF Report 22," OECD/NEA, ISBN 978-92-64-99074-6, 2009.

Potential Impact:

Expected final results:
At the end of the project, a decision maker should be able to evaluate based on scientific and economic data what option is most favourable for his or her case (NPP park, boundary conditions on national policy, …).
The final results, a scientific/economic comparison of a critical and sub-critical transmutation facility, including spent fuel reprocessing and transmutation fuel fabrication should allow decision makers to evaluate both options for their respective country.
The technical options have been attributed a technological readiness level. This will allow decision makers to take decisions on funding in Research and Development if these decision makers intend to deploy these technologies.

Potential impact:

The final results, a scientific/economic comparison of a critical and sub-critical transmutation facility, including spent fuel reprocessing and transmutation fuel fabrication should allow decision makers to evaluate both options for their respective country.
The outcome of the project also demonstrates that for the majority of the technologies used in the calculations, the Technological Readiness Levels are very low. This implies that a lot of Research and Development is extremely urgently needed in these fields in order to be able to deploy these technologies. It also implies that cost estimates are hard to make. This is again a call for more Research & Development. Cost estimates can only be reliable in case of a certain level of Technological Readiness. Hence, business plans have been hinted at and examples are given in deliverable 7, but the level of uncertainties is so high that no reasonable business plan can be established.
As a CSA, the project was not to perform a lot of technical/scientific work. However, a lot of information has been gathered from past and present projects and collected in a coherent way.
Promotion of the work
Being a short CSA (2 years), the partners have focused on writing clear deliverables extracting crucial information to be presented at scientific conferences. No publication in peer-reviewed journals was envisaged as 2 years' time is short and this is not a scientific/technical development project.
The results of the project have been presented at two nuclear technology conferences:
the 12th International Exchange Meeting on Partitioning and Transmutation (IEMPT12) held in September in Prague in 2012 and the 2013 Fast Reactor Conference (FR13) held in March 2013 in Paris. The presentations and submitted papers have been attached.
The results have also been communicated by the coordinator at Technical Meetings organized by the IAEA in the INPRO/GAINS and INPRO/SYNERGIES projects.
Finally, partial results have been presented in June 2012 by the coordinator at a session of the French Comité National d'Evaluation (CNE2) which has to report to the French government on the scientific progress made on the treatment of nuclear spent fuel.

List of Websites: