DEVELOPMENT OF 3.95 MeV X-BAND LINAC-DRIVEN X-RAY COM-

BINED NEUTRON SOURCE*

J. M. Bereder†, Y. Mitsuya, Y. Takahashi, K. Dobashi, M. Uesaka,

The University of Tokyo, Tokyo, Japan

J. Kusano, Accuthera Inc., Kanagawa, Japan

Y. Tanaka, Institute of Industrial Science, Tokyo, Japan

Y. Oshima, M. Ishida, Public Works Research Institute, Ibaraki, Japan

Abstract The existing non-destructive inspection method em-

ployed for concrete structures uses high energy X-rays to detect internal flaws in concrete structures and iron rein-forcing rods. In addition to this conventional method, the authors are developing an innovative inspection system that uses a mobile compact linac-driven neutron source that utilizes neutron backscattering, to measure the mois-ture distribution in concrete structures and estimate the corrosion probability distribution of iron reinforcing rods.

INTRODUCTION

Research Purpose

During the period of rapid economic growth from the

1950s to the 1970s, Japanese governmental investment in

infrastructural projects such as bridges and buildings

expanded rapidly. However, most of the industrial and

social infrastructure projects had an estimated lifetime of only approximately 50 years, so the declining strength of

concrete structures has become an issue of national im-

portance.

From a cost-performance point of view, it is better to carry out on-site non-destructive inspections regularly and repairs as and when needed, rather than demolishing or rebuilding. Non-destructive inspection methods for social infrastructures, aimed at detecting internal flaws in con-crete structures and iron rods, have therefore been devel-oped[1]. In addition to the conventional high energy X-

ray method, the development of a neutron backscatter moisture detection system using a linear accelerator driv-en neutron source to measure moisture distribution in concrete structures is now under development. By com-bining the knowledge of the moisture distribution in con-crete structures, the corrosion probability distribution of iron reinforcing rods can be estimated. (Figure 1)

Moisture Detection for PC Bridges

About 60% of the bridges, the most important infra-structures, are concrete bridges, and about 40% of them are prestressed concrete bridges (PC bridges).

A PC bridge has cylindrical structure of iron inside the concrete, and steel wires (PC wire) with tensile stress (prestress) are passed through, thereby giving compres-sive stress to the entire concrete, and strengthening the whole structure. The PC wires are filled with grout filler and placed under basic conditions, so that a passive film of Ca(OH)2 is formed on the surface of the PC wires that

prevents internal corrosion. This passive film is destroyed by neutralization due to CO2 invasion into the concrete. Corrosion of the wire is caused by moisture penetration in the grout unloaded part due to initial construction failure, leading to deterioration of the concrete (Figure 2). There-fore, it is important to acquire internal moisture distribu-tion information for the soundness evaluation of concrete structures, but it is not yet realized in non-destructive inspection technology usable on site.

Corrosion probability distributionNeutron probe

NDT for bridges using neutron probe

Moisture distribution information

NDT for bridges using high energy x-ray

50 years～Year after construction

High energy X-ray NDT

Detection of inner flaw

Structure analysis (FEM, Beam theory)

～20 yearsYear after construction

Figure 1: Applications of X-ray source and neutron

source in NDT of social infrastructure.

Figure 2: Description of prestressed concrete structure.

PC wire corrosion due to Neutralization(CO2) & moisture penetration

PC wire Grout filler

Prestress Prestress

Moisture Moisture

Compressive stress

Compressive stress

Sheath

CO2

THPVA098 Proceedings of IPAC2017, Copenhagen, Denmark

ISBN 978-3-95450-182-34692Co

pyrig

ht©

2017

CC-B

Y-3.

0an

dby

ther

espe

ctiv

eaut

hors

08 Applications of Accelerators, Technology Transfer and Industrial RelationsU05 Applications, Other

DESCRIPTION OF NEUTRON

SOURCE

Target Design

Beryllium 9Be, having the lowest threshold energy for photo-nuclear reaction 9Be(γ, n)8Be*, is used in a mobile linac-driven neutron source, as designed by an Italian group in the previous study [2]. A beryllium photo-

neutron target has been combined with a lead beam colli-mator, a boric acid resin layer for neutron shielding, and a lead layer for γ-ray shielding (Figure 3). The 9Be target and the 3.95 MV mobile X-ray source together comprise the mobile neutron source system (Figure 4). Since main-ly fast neutrons are used in the neutron source, a beam line using a high-Z material that does not moderate the neutrons is used. Optimization of the beryllium target size and neutron/γ-ray shielding simulation is performed using the Monte-Carlo code.

Figure 3: Description of neutron source.

Neutron Yield and Energy Distribution

The calculated neutron yield in the neutron source is approximately ~ 7n/s. The target weight is about 100 kg, which is manageable by manpower. Radiation safety is satisfactory as well. The distribution of neutron energy produced is shown in Figure 5.

Figure 4: Neutron source setup.

Neutron energy

Neu

tro

n f

lux

a.u

.

MeVmeV keVeV

1100101100101100101

Figure 5: Produced neutron energy distribution.

MOISTURE DETECTION USING

BACKSCATTER NEUTRON

Detection Principle Moisture detection using neutrons is performed by irra-

diating concrete with fast neutrons and detecting backscattered moderated neutrons due to multiple elastic scattering with light elements especially hydrogen nuclei.

The substance used for neutron detection including 3He has a high reaction cross section with neutrons in the thermal region, and therefore the count of detectors in-creases with the existence of moisture (Figure 6).

1.00E+00

1.00E+01

1.00E+02

1.00E+03

1.00E+04

1.00E+05

1.00E+06

1.00E-11

1.00E-10

1.00E-09

1.00E-08

1.00E-07

1.00E-06

Neutron energy

Neu

tro

n f

lux

(a.u

.)

Cro

ss sectio

nb

arn

Thermal neutron

MeVmeV keVeV

1100101100101100101

101

100

102

103

104

105

106

With moisture

No moisture

Incident neutron

3He (n,p)cross section

Figure 6: Backscatter neutron energy (with and without

water), incident neutron energy and 3He(n,p) cross section.

Moisture Detection Experiment of PC Concrete

Sample

Moisture detection experiments simulating immersion of PC wire duct in bridges were also conducted (Figure 7). 100 g of water was placed in a concrete sample having a cylindrical sheath structure at a depth of 10 cm in the concrete, and the measurement was performed. As a re-sult, the measurement was successful with accuracy of 3σ (Figure 8).

Boric acid resin

Lead

Neutron collimator

23cm

2cm

5cm

2cm

Target weight 100 kg Total neutron yield ～107 n/s Neutron flux(@Target surface) ～104 n/s/cm2

Beryllium target

5cm×5cm×5cm

Neutron target

3He detector

X-ray source

Proceedings of IPAC2017, Copenhagen, Denmark THPVA098

08 Applications of Accelerators, Technology Transfer and Industrial RelationsU05 Applications, Other

ISBN 978-3-95450-182-34693 Co

pyrig

ht©

2017

CC-B

Y-3.

0an

dby

ther

espe

ctiv

eaut

hors

Figure 7: Experiment simulating immersion of PC wire duct.

Figure 8: Result of PC sheath moisture detection experi-ment.

Concrete Moisture Content Measurement

Concrete moisture content measurement experiment was conducted as a fundamental study for measurement of moisture distribution in concrete structures(Figure 9). Each concrete sample with a water cement ratio (W/C ratio) of 36% and 50% was adjusted to 0% moisture con-tent and 100% moisture content and 600 second-

measurement were carried out. As a result, even for con-crete samples with different W/C ratios, the difference in neutron detector counts between the moisture content 0% case and 100% case is approximately the same, which means the moisture information is successfully acquired regardless of the formulation of concrete sample it-self.(Figure 10)

3He detector

Concrete sample

Neutron

Figure 9: Concrete moisture content measurement setup.

Figure 10: Result of concrete water content measurement experiment.

CONCLUSION

An experiment simulating flood conditions in actual bridges and a measurement experiment of concrete mois-ture content - using a mobile linac driven compact neu-tron source - were carried out.

The development of a measurement system using neu-tron energy information by TOF to improve inspection accuracy is now under development. The first outdoor actual bridge test using an accelerator driven neutron source will also be conducted in this fall.

ACKNOWLEDGEMENT

This work was supported by Council for Science, Tech-nology and Innovation (CSTI), Cross-Ministrial Strategic Innovation Promotion Program (SIP) (Funding agency: JST).

REFERENCES

[1] M. Uesaka et al., J. Phys. B: At. Mol. Opt. Phys. 47, 234008, 9pp, 2014.

Neutrons

Concrete sample

X-ray

3He detector

Sample

[2] L. Auditore et al., Nucl. Instrum. Meth. B, vol. 229, pp. 127143, 2005.

THPVA098 Proceedings of IPAC2017, Copenhagen, Denmark

ISBN 978-3-95450-182-34694Co

pyrig

ht©

2017

CC-B

Y-3.

0an

dby

ther

espe

ctiv

eaut

hors

08 Applications of Accelerators, Technology Transfer and Industrial RelationsU05 Applications, Other