Search

Abstract: Lithium in a breeding blanket is burned up through neutron nuclear reactions in fusion DEMO reactors. Effects of decrease of solid breeder materials due to lithium burn-up on tritium breeding ratio (TBR) are not systematically calculated in the past. For the SlimCS blanket design, TBR is calculated taking into account the lithium burn-ups by one dimensional Sn radiation transport calculation code ANISN in this study. The 6Li burn-ups are 8–79% after 10-year operation. TBR due to 6Li decreases to 40% of the initial one in some layer, while it increases in some layers. The TBR integrated over all the blanket decreases to around 96% of the initial one. The study makes it clear that the reduction of the TBR due to the lithium burn-up is not so large. [Copyright &y& Elsevier]

Abstract: An invessel neutron monitor using micro-fission chambers is one of the most important diagnostics to measure the fusion power in ITER. A single micro-fission chamber with 10mg of 235U to be installed in a gap between a shielding blanket module and a vacuum vessel can cover the range of the fusion power from 100kW to 1GW. For the low fusion power operation including D–D operation, a micro-fission chamber bundle consisting of 13 micro-fission chamber units has been designed. Nuclear heating of the micro-fission chamber for high power operations and the micro-fission chamber bundle for low power operation were analyzed by neutron transport calculations using the MCNPX code. Heat transport analyzes for the micro-fission chambers were carried out with a general purpose Finite Element Method code ANSYS 8.1, which indicated that the detector temperature might be kept less than the operational temperature limit by the thermal conduction to the vacuum vessel with the heat transfer coefficient of 100W/m2 K or larger. [Copyright &y& Elsevier]

Abstract: A neutronic study has been carried out for VECTOR, which is a low aspect ratio tokamak with the aspect ratio of 2.3, to achieve simultaneously sufficient shielding for super-conducting toroidal field coils (TFCs) and tritium breeding ratio (TBR) larger than unity, using the three-dimensional Monte Carlo neutron transport code MCNP-4C. The tritium production is mainly made by outboard blankets consisting of the self-cooled lithium lead (LiPb) as tritium breeding and neutron multiplying material, and the reduced activation ferritic steel F82H as structural material. The inboard blanket consisting of vanadium hydrate (VH2) and F82H as structural material plays the role of radiation shield for the TFCs. Adding Be layers to the LiPb layers effectively improves the TBR. By employing a 0.8m thick inboard blanket with a 0.13m thick LiPb layer, sufficient shielding for the inboard TFC and a total TBR of 1.08 are achieved. [Copyright &y& Elsevier]

Abstract: Quartz glasses of KU-1 and KS-4V are candidate window materials for the ITER optical diagnostics in UV and visible range. The temperature dependence of the radiation-induced transmission losses in UV range has been investigated under γ-ray irradiation for the KU-1 and KS-4V quartz glasses up to 10MGy. The KU-1 and KS-4V quartz glass samples were irradiated at room temperature, 100, 200 and 300°C. Dose rate was 0.46–1.1MGy, and 2.6Gy/s from 1.1 to 10.4MGy. It was confirmed that there is no significant loss in wavelength range longer than 350nm for both window materials. Transmission loss in KU-1 is larger than that in KS-4V under temperature below 200°C. KU-1 has large temperature dependence. On the other hand, temperature dependence is not clear in KS-4V above 100°C. Prominent recover of the transmission loss was not observed after irradiation for each sample. From the transmission loss point of view, KS-4V is better window material than KU-1 at temperature below 200°C. KU-1 is available at 300°C. [Copyright &y& Elsevier]

The D–T neutron skyshine experiments have been carried out at the Fusion Neutronics Source (FNS) of JAERI with the neutron yield of ∼1.7×1011 n/s. The concrete thickness of the roof and the wall of a FNS target room are 1.15 and 2 m, respectively. The FNS skyshine port with a size of 0.9×0.9 m2 was open during the experimental period. The radiation dose rate outside the target room was measured a maximum distance of 550 m from the D–T target point with a spherical rem-counter. Secondary gamma-rays were measured with high purity Ge detectors and NaI scintillation counters. The highest neutron dose was about 9×10−22 Sv/(source neutron) at a distance of 30 m from the D–T target point and the dose rate was attenuated to 4×10−24 Sv/(source neutron) at a distance of 550 m. The measured neutron dose distribution was analyzed with Monte Carlo code MCNP-4B and a simple line source model. The MCNP calculation overestimates the neutron dose in the distance range larger than 230 m. The line source model agrees well with the experimental results within the distance of 350 m. [Copyright &y& Elsevier]

Irradiation tests of a mica substrate bolometer were carried out at the Japan Materials Testing Reactor up to the fast neutron fluence of 0.1 dpa. A performance of the bolometer was investigated during the irradiation. Significant increase of the meander resistance from 275 to 446 Ω was observed in the fast neutron fluence of 0.06 dpa. In the post-irradiation examination (PIE), it was confirmed that the gold meander contained 46% mercury which was produced by the nuclear transmutation reaction of 197Au(n, γ) 198Au→198Hg+β−. The bolometer sensitivity was measured by the change of the meander resistance for the input power. The sensitivity decreased slightly at constant reactor power. There was not significant change in the bolometer response time. During the cool down phase of the first cycle all connections went open circuit. Some recovered on start-up of the second cycle, however, all connections went open circuit again near the end of the second cycle. They were not recovered any more at the third cycle. Breaks of the gold meander were observed in the PIE. The use of gold meanders might be problematic in ITER. [Copyright &y& Elsevier]

Abstract: We developed a CAD/MCNP interface program prototype to easily perform Monte Carlo fusion nuclear design calculation with high accuracy using detailed calculation geometry. Usually, there is only CAD data on the solid region. However, CAD void region data must be input to MCNP. Our newly developed CAD/MCNP interface program creates the void region data automatically. The void region data is created by subtracting the solid region data using Boolean operations. The program has also the function of uniformly dividing the overall data into small cubes to simplify the void region data. It has been applied to create 3D CAD data in the ITER benchmark model. It was demonstrated that data of the void region in the ITER model were successfully created using our program. [Copyright &y& Elsevier]

Abstract: The ITER neutron diagnostic system needs to cover both the DD- and DT-phases of ITER experiments. To ensure the required accuracy for diagnosing both phases, it is preferable to calibrate the diagnostic system to both DD and DT neutrons. In this paper, the feasibility of two kinds of calibration measurements has been verified through MCNP simulations for the ITER ex-vessel neutron yield monitor. It is indicated that both calibration procedures using DD and DT neutron generators can be implemented with the required accuracies within a realistic measurement time. In the advanced DD-phase of the ITER experiment with high density D plasma producing high energy tritons, where a number of DT neutrons are produced, the independent calibration to DD or DT neutrons could lead to unacceptable measurement uncertainty. An effective method to reduce the measurement uncertainty for the two-component neutron source is discussed. [Copyright &y& Elsevier]

At Fusion Neutronics Source (FNS) of JAERI, tritium breeding experiments with blanket mock-ups consisting of advanced fusion reactor materials are in progress. The breeding zones are thin layers of lithium titanate which is one of the candidate tritium breeder materials for the DEMO fusion power reactor. It is anticipated that the application of small pellet-shaped lithium titanate detectors manufactured from the same material as the breeding layer will reduce experimental uncertainties arising from necessary corrections due to different isotopic lithium volume concentrations in breeding material and detector. Therefore, a method was developed to measure the local tritium production by means of lithium titanate pellet detectors and a liquid scintillation counting technique. The lithium titanate pellets were dissolved in concentrated hydrochloric acid solution and the resulting acidic solution was neutralized. Two ways of further processing were followed: direct incorporation into a liquid scintillation cocktail and distillation of the solution followed by mixing with liquid scintillator. Two types of lithium titanate pellets were investigated with different 6Li enrichment and manufacturing procedure. It was found that lithium titanate is suitable for tritium production measurements. However some discrepancies in the measurement accuracy remained with one of the investigated pellet detectors when compared with a well-established lithium carbonate measurement technique and this issue needs further investigation. [Copyright &y& Elsevier]

Abstract: This paper summarizes the proposals and findings of the IFMIF Specification Working Group established to update the users requirements and top level specifications for the facility. Special attention is given to the different roadmaps of fusion pathway towards power plants, of materials R&D and of facilities and their interactions. The materials development and validation activities on structural materials, blanket functional materials and non-metallic materials are analyzed and specific objectives and requirements to be implemented in IFMIF are proposed. Emphasis is made in additional potential validation activities that can be developed in IFMIF for ITER TBM qualification as well as for DEMO-oriented mock-up testing. [Copyright &y& Elsevier]

Abstract: For contributing to the ITER project and promoting a possible early realization of DEMO, since 2007 the International Fusion Energy Research Centre (IFERC) shall perform the activities on (1) DEMO Design and R&D Coordination, (2) Computational Simulation Centre, and (3) ITER Remote Experimentation Centre in the framework under the agreement between the Government of Japan and the European Atomic Energy Community for the joint implementation of the Broader Approach Activities in the field of fusion energy. The DEMO Design activity aims at establishing a common basis for DEMO Design, including design features of DEMO, possible common concepts of DEMO, a roadmap for DEMO, and so on. Based on the common interest of EU and Japan toward DEMO, the DEMO R&D activities have been planned and carried out for the areas which are related to DEMO blanket development: SiC/SiC composites, tritium technology, materials engineering, advanced neutron multiplier, and advanced tritium breeders for DEMO blanket. In the activity of the Computational Simulation Centre, the objective is to provide and exploit a supercomputer for large scale simulation activities to analyse experimental data on fusion plasmas, prepare scenarios for ITER operation, predict the performance of the ITER facilities and contribute to the DEMO Design. [ABSTRACT FROM AUTHOR]

Abstract: A series of breeding blanket neutronics experiments were carried out with DT neutrons in the Japanese ITER test blanket module at JAEA Fusion Neutronics Source (FNS), and tritium production rates (TPRs) inside different blanket mock-ups were measured in detail. Most of calculations made using FENDL-2.1 and MCNP agreed well with the measured TPRs. On the other hand, there were overestimations of the calculation for the TPR in the experiment with a reflector and the TPR around the boundary between the rear part of the breeder layer and the beryllium in the pebble bed mock-up experiment. In this study, we investigated the possibility of improvement of the calculation by modifying the part of angular distributions in backward directions in FENDL-2.1. By uniformly decreasing the backward elastic scattering cross-sections of 56Fe by 20% below the original data for incoming neutrons in the energy range from 0.5234MeV up to 15.0MeV, the overestimation of the calculations for the TPR in the enriched breeding blanket experiment with a reflector was reduced. Also, by uniformly decreasing the backward elastic scattering cross-sections of 9Be by 20% and 30% below the original data for the incoming neutrons in the energy range from 0.93MeV up to 14.94MeV, and by 20% for those from 0.62MeV up to 14.94MeV, the overestimation of the calculations for the TPR around the boundary between the rear part of the breeder layer and the beryllium was lessened. [Copyright &y& Elsevier]

Abstract: There are very thin and long slits at boundaries between vacuum vessel and port plugs in ITER. The slits are 0.5–2cm in width and their length amount about 200cm in depth. Therefore, it is a key issue to evaluate neutron-shielding effects of assembly with such thin and long slits. In order to verify them, our group has started to conduct the D-T neutronics experiment with an experimental iron assembly at the Fusion Neutronics Source (FNS) in Japan Atomic Energy Agency (JAEA) under the ITER/ITA Task 73–10. Fission rates among a slit in the assembly were measured with U-238 and U-235 micro-fission chambers. From our experiment, the following facts were found: (1) the C/E value of U-238 fission rate is within 0.9–1.1 at the most measured points and (2) the C/Es of the U-235 fission rate were almost closed to the unity within 0.9–1.2. However, they tended to decrease for the depth slightly. [Copyright &y& Elsevier]

Abstract: Double-differential cross-section (DDX) for emitted charged particles is necessary to estimate material damage, gas production and nuclear heating in a fusion reactor. Detailed measurements of the cross-sections for beryllium, carbon and fluorine, which are among the composition materials of expected fusion blankets and first walls, were carried out with a charged-particle spectrometer using a pencil-beam DT neutron source. As verification of the cross-sections evaluated in three nuclear libraries (JENDL-3.3, ENDF/B-VI and JEFF-3.1), our measured data were compared with the data evaluated in the libraries. From the comparison, the following problems were pointed out: Beryllium: Remarkable differences in energy and angular distribution for α-particles were observed between the measured data and the libraries. The estimated total cross-section for α-particle production well agreed with the libraries. Carbon: There was a discrepancy of about 20% between JENDL-3.3 and ENDF/B-VI (JEFF-3.1) for α-particle production cross-section, and no DDX for α-particles is given in the libraries. Our obtained total cross-section for α-particle production was rather consistent with ENDF/B-VI (JEFF-3.1), and the value evaluated in JENDL-3.3 seemed too large. Fluorine: The remarkable differences for DDX of protons and α-particles were observed between the obtained result and JENDL-3.3, although detailed DDX was stored only in JENDL. The obtained total cross-sections mostly supported the evaluation of ENDF/B-VI (JEFF-3.1). [Copyright &y& Elsevier]

In order to evaluate neutron activation of materials relevant to the liquid blankets such as Lithium/vanadium-alloy and Flibe/vanadium-alloy blankets, irradiation experiments have been performed using the fusion neutronics source (FNS) at JAEA. Specimens of Er, Teflon (CF2=CF2) and NIFS-HEAT-2 (V–4Cr–4Ti) were selected in the experiments for the evaluations of the activation of Er in the MHD coating (Er2O3), F in Flibe (LiF–BeF2) and V, Cr, Ti in the structural materials, respectively. The irradiation experiments were performed with two kinds of neutron field: D–T neutrons and Be moderated neutrons for the purpose of reaction with high and low energy neutrons, respectively. The neutron spectra calculations were carried out using Monte Carlo transport codes MCNP-4C and JENDL3.3 files. The neutron-induced activity measured with a high purity Ge detector was compared with calculations carried out by the FISPACT-2001 codes and EAF-2001 files. The results showed that the calculated activities of most of products through reaction with high energy neutrons are in agreement with the experiments within 20% uncertainty. [Copyright &y& Elsevier]

Abstract: The possibility of tritium self-sufficiency in a low aspect ratio (low-A) tokamak reactor with the outboard blanket only has been studied to simplify the inboard structure located in a high magnetic field. The tritium breeding ratio (TBR) of the outboard blanket increases by applying a neutron multiplier such as lead and beryllium to the inboard reflector. The effect in local TBR of a 40cm thick inboard reflector of lead is about 40% that of the inboard breeding blanket. The local TBR of a reactor with both the lead inboard reflector and outboard blanket using a Li2O breeder and a beryllium multiplier can be larger than 1.35 for aspect ratios less than 2.9. The local TBR of 1.35 satisfies the net TBR requirement of 1.05 where there is 78% coverage of the breeding zone over the total plasma-facing surface area. The result indicates there is a design solution achieving tritium self-sufficiency even in a low-A tokamak reactor with the outboard breeding blanket only, using an appropriate inboard reflector. Be12Ti and Li2TiO3 are recommended from the viewpoint of safety. In the combination, the aspect ratio is required to be less than 1.35 to achieve local TBR of 1.35. [Copyright &y& Elsevier]

Abstract: Sequential charged particle reactions, that is, reactions induced by secondary charged particles generated due to neutron reactions, may cause non-negligible radioactivity in low activation material. A computer code ACT4, which had been developed to calculate radioactivity, decay heat and gamma ray source for fusion reactors was supplemented an ability to deal with the reactions. Experiments conducted with Fusion Neutronics Source in Japan Atomic Energy Research Institute for the radioactivity by sequential charged particle reactions were analyzed with the newly revised code of ACT4. As the result, the activations via sequential reaction in LiF showed reasonable agreement with experimental data, which means that the ACT4 program will work well. However, the other activations in vanadium and low activation ferritic steel did not. The disagreement must be caused by incorrectness in neutron cross sections, charged particle cross sections, charged particle emission spectra, etc., and they need to be fully reviewed for the detailed and reliable design analysis of fusion reactors. [Copyright &y& Elsevier]

Abstract: In order to investigate the activation performance of prospective materials of the IFMIF facility, we have carried out irradiation experiments with the AVF cyclotron in the TIARA facility of the Japan Atomic Energy Research Institute (JAERI). We measured the production cross sections of the nuclides Cu-61, Cu-64, Zn-62 and Zn-63 by deuterons irradiation of copper with natural isotope abundance (natCu) in the energy range of 22–40MeV. The obtained cross sections were compared with data from the ACSELAM library and with previously measured cross section data. The comparisons revealed that some of the ACSELAM cross section data are different from the experimental data and have to be improved. [Copyright &y& Elsevier]

Abstract: Double-differential charged-particle emission cross section (DDXc) is a fundamental value to estimate nuclear heating and material damages in fusion reactor development. In the present study, measurements of DDXc for several light elements have been carried out with a newly developed spectrometer. In the spectrometer, a good signal-to-noise ratio and very low minimum measurable energy were realized by using a pencil-beam DT neutron source of FNS in JAERI and a counter-telescope system with very thin silicon surface barrier detectors of appropriate thicknesses. The detailed DDXc data in a wide emission energy range were obtained for 27Al, 9Be, natC and 19F. For 27Al(n,xα) reactions, the measured total cross section agreed well with the evaluated data. For 9Be(n,xα) reactions, there were differences in the α-particle emission DDX between measurement and the evaluated data in the detail of energy structure. For natC(n,xα) reaction, a large discrepancy between the present result and the previous experimental data appeared. For 19F(n, charged-particle) reactions, some peaks corresponding to excited states of the residual nuclei were observed in the energy spectrum of emitted proton, deuteron, and triton particles. [Copyright &y& Elsevier]

Abstract: The neutronics study in support of the DEMO blanket concept has been performed using the DT neutrons at the FNS facility in JAERI on blanket mockups and a bulk beryllium assembly. From the results obtained in the blanket mockups, it has been clarified that the C/Es of the integrated tritium production are 1.01–1.04 and 1.11–1.13 in the experiments without and with the neutron reflector, respectively. The tritium production can be very accurately predicted in the experiment without the reflector. In the mockups with the 12.6 and 25.2mm thick tungsten armors, the integrated tritium productions are reduced by 3 and 6% relative to the case without the armor, respectively. Based on analysis of the integral experiment on the beryllium assembly, it is possible to conclude that the calculation uncertainty of the reaction rates in beryllium provides an acceptable agreement with the experimental data related to tritium breeding. Tritium techniques used for TPR measurements have been tested in the international benchmark experiment. There is a very good agreement within the estimated uncertainties between results obtained at JAERI and at TUD, though the results obtained at ENEA show an underestimation (about 15–20%) in comparison with FNS and TUD results. [Copyright &y& Elsevier]

Abstract: Two nuclear microprobe techniques; deuteron-induced nuclear reaction analysis (d-NRA) and neutron elastic recoil detection analysis (NERDA), are presented to obtain hydrogen isotope depth profiles of fusion reactor materials in a shallow and deep region, respectively. Tritium and deuterium depth profiles in a bumper limiter of TFTR used in the D–T experiments were measured as an application of d-NRA. The T(d, α)n and D(d, t)H reactions were used to determine the depth profiles. The tritium concentration had a peak, whereas deuterium showed a broad distribution in the profile. Also, the application of NERDA is demonstrated. A proof-of-principle experiment was performed using a standard sample of deuterated polyethylene. The depth profile and resolution for NERDA are discussed. [Copyright &y& Elsevier]

Abstract: This paper presents an evaluation of accessibility of the Li loop piping considering activated corrosion product. International Fusion Materials Irradiation Facility (IFMIF) is a deuteron-lithium (Li) stripping reaction neutron source for fusion materials testing. Target assembly and back wall are designed as fully remote maintenance component. Accessibility around the Li loop piping will depend on activation level of the deposition materials due to the back wall erosion/corrosion process under liquid Li flow. Activation level of the corrosion products coming from the AISI 316LN back wall is calculated by the ACT-4 of the THIDA-2 code system. The total activities after 1 day, 1 week, 1 month and 1 year cooling are 3.1×1014, 2.8×1014, 2.3×1014 and 7.5×1013 Bq/kg, respectively. Radiation dose rate around the Li loop pipe is calculated by QAD-CGGP2R code. Activated area of the back wall is 100cm2. Corrosion rate is assumed 1μm/year. When 10% of the corrosion material is supposed to be deposited on the inner surface of the pipe, the dose rate is calculated to be less than a permissible level of 10μSv/h for hands-on maintenance, therefore, the maintenance work is assessed to be possible. [Copyright &y& Elsevier]

Abstract: Experiments have been conducted with breeding blanket mock-ups consisting of layers of candidate materials for JAERI’s pebble bed blanket design: lithium titanate, beryllium and reduced activation ferritic steel F82H. The mock-ups were irradiated with 14 MeV neutrons from the DT neutron source at FNS. The lithium titanate was enriched to 40 and 95% in . The local tritium production rate in the 12 mm thick breeding layers was measured by means of pellet detectors of lithium titanate, Li2CO3 and liquid scintillation counting techniques. Additional verification of the neutron spectrum was obtained from material activation such as activation foils. The results were analyzed with the Monte Carlo neutron transport code MCNP-4C and several transport cross section libraries such as FENDL/MC-2.0. The calculation/experiment ratio was found to be within the combined error estimate of calculation and experiment but in some cases, the calculations showed an overestimation tendency. [Copyright &y& Elsevier]

Abstract: For the evaluation of the damages, nuclear heating, and also tritium inventory in the fusion materials, the precise information of the energetic charged particles produced in the materials by nuclear reactions is important. Fusion neutronics source in Japan Atomic Engineering research Association, JAERI/FNS, has started the measurements of the double differential cross-sections (DDXs) and the angular differential cross-sections (ADXs) for the candidate materials of fusion reactors using the collimated D–T neutron beam system and the precise results of the DDXs and ADXs for the 9Be(n,αx) and 9Be(n,t) reactions have been obtained. Also, from the comparison of the evaluated nuclear data for the beryllium, it was found that the calculated DDX of 9Be(n,α) above about 3MeV was underestimated and the total 9Be(n,t) cross-section obtained with the measurement of the DDX and ADX was in good agreement with the evaluated nuclear data. In the following sections, the details of the measurement methods for the charged particles and the obtained DDX and ADX for the beryllium are described. [Copyright &y& Elsevier]

Abstract: An integral breeding blanket experiment has been done at Fusion Neutronics Source of JAERI with an assembly of three layers of lithium titanate, beryllium, and low activation ferritic steel F82H. The experimental assembly and the neutron source were enclosed in a SS316 stainless steel can and irradiated with D–T neutrons. The tritium production rate was measured with small Li2CO3 pellet detectors. The tritium production from7Li was estimated by means of the reactions32S(n, p)32P and35Cl(n, )32P which have effective cross-sections similar to7Li(n, n)T but are more sensitive methods. The neutron flux in the assembly was checked with activation foils. The measurement results were analyzed with the three-dimensional Monte Carlo code MCNP-4C and three different beryllium cross-section evaluations taken from FENDL/MC-2.0, FENDL/E-2.0, and EFF-3.03. The library FENDL/MC-2.0 was also used for the other materials in the assembly. All three beryllium libraries model the fast flux in the assembly well. Differences were seen for indium foils. An overestimation was observed for the tritium production. [Copyright &y& Elsevier]

A shutdown dose experiment was conducted by using the FNS at JAERI under the ITER R&D Task T-426, in order to evaluate the accuracy of the shutdown dose evaluation method. The activation and the delayed gamma-ray distribution after passing through the straight duct were simulated in the experiment. The experiment was conducted at the 80° beam line of the FNS and the irradiation was carried out over 6 days. The experimental analyses were carried out by using the new direct one step Monte Carlo method that is used as the design tool to estimate shutdown dose of the ITER/FEAT. The nuclear data packages, FENDL-2/MC and FENDL-2/A, were compared through the analysis of the experiment. As a result, the following facts were found for the accuracy of the shutdown dose evaluation: (1) the C/E values of the shutdown dose at around 106 s after the irradiation were 0.92–1.07; (2) the shutdown dose rate could be evaluated by the new direct one-step Monte Carlo method within the experimental error, and the accuracy of the method is sufficient for shielding design of the International Thermonuclear Experimental Reactor (ITER). [Copyright &y& Elsevier]

Degradation of electrical insulation property of mineral-insulated (MI) cables under fusion neutron irradiation was examined by use of Fusion Neutronics Source at Japan Atomic Energy Research Institute. Observed increase of leakage current in an MI cable (diameter; 4.8 mm, length; 2.0 m) was ∼5.0 pA for the DT neutron flux of 2.9×107 n/cm2/s and bias voltage of 200 V between the central wire and sheath. The bias voltage and neutron flux dependence of the leakage current indicates that the degradation of insulation property of the cable is dominated by the production rate and drift of electric charge in the insulator. Furthermore, the appearance of large transient leakage currents immediately after the start and stop of the irradiation and the growth of leakage current even for the bias voltage of 0 V imply that electrical properties of MI cables under irradiation are considerably affected by the distribution of electric charge trapped in the insulator. [Copyright &y& Elsevier]