USING A FAST NEUTRON FLAW DETECTOR TO SOLVE CERTAIN PROBLEMS IN THE INSPECTION OF REINFORCED CONCRETE STRUCTURES

V. I. GORBUNOV, V. B. ELAGIN, G. Sh. PEKARSKII

The widespread introduction of reinforced concrete in the construction industry, and the manufacture of complex, critical structures from reinforced concrete, require the introduction of quality control for products and, above all, nondestructive methods.

In the inspection of reinforced concrete structures, two tasks are mainly considered:

1. Inspection of reinforcement and metal embedded elements (presence, fastening at joints, position, absence of breaks, etc.).

2. Inspection of concrete quality (presence of voids and cracks, detection of foreign inclusions, density distribution, etc.).

The first task is successfully solved using gamma radiation from isotope sources and bremsstrahlung radiation from X-ray installations and accelerators [1, 2, 3]. Depending on the thickness and density of the concrete, the energy of the X-ray and gamma radiation is selected.

Thus, for concrete thicknesses up to 15 cm it is most expedient to use a 100 kV X-ray installation with a current of up to 5 mA. With a 300 kV X-ray installation (current up to 10 mA), products made of concrete up to 30 cm thick are inspected. When inspecting concrete thicknesses up to 60 cm, Cs-137 and Co-60 gamma sources are used [1]. A further increase in the thickness of concrete structures makes the use of isotope sources impractical because of the sharp decrease in the productivity of the method and the increase in the gamma-source activity required for inspection.

Thus, the exposure time when radiographing a concrete slab 40 cm thick with a Co-60 source of activity 1 curie is more than 30 hours [1].

The use of bremsstrahlung radiation from a betatron for inspection made it possible to increase substantially the inspected thicknesses and to increase productivity considerably [4]. Radiography of a reinforced-concrete structure 50 cm thick using a PMB-6 betatron requires 25–30 min. A clear image was obtained of reinforcing bars 6–7 mm in diameter located at a depth of up to 20 cm.

High sensitivity to the detection of defects in the metal components of reinforced-concrete structures is achieved owing to the large difference between the attenuation coefficients of gamma radiation in iron and in concrete. The difference in the attenuation coefficients of γ-radiation by concrete and iron will determine the difference in radiation intensity,

that has passed through a reinforced-concrete structure when a section of the structure with reinforcement, \(I_1\), and without reinforcement, \(I_2\), is irradiated by a beam,

\[ \frac{I_1}{I_2}=e^{(\mu_{\text{bet}}-\mu_{\text{st}})\cdot x}, \]

where

\(\mu_{\text{st}}, \mu_{\text{bet}}\) are the attenuation coefficients of radiation in steel and concrete, respectively;

\(x\) is the thickness of steel in the reinforced-concrete structure.

The linear attenuation coefficients of gamma radiation in concrete and iron are presented in Table 1.

Table 1

Linear attenuation coefficients of gamma radiation from isotope sources Co\(^{60}\), Cs\(^{137}\) in iron and concrete [1, 5]

While providing high sensitivity in selective inspection of reinforcement in reinforced concrete, the \(\gamma\)-method at the same time does not make it possible to solve a number of problems in quality control of the concrete itself in reinforced-concrete structures, such as, for example, determining air voids during the manufacture of structures. Air voids arise in structures mainly at reinforcement-joining nodes and significantly reduce the strength of the structures. The attenuation of gamma radiation by the reinforcement compensates for the increase in the flux of transmitted radiation caused by the air void. For example, a layer of iron \(2.15\) cm thick will attenuate the radiation intensity of the isotope Co\(^{60}\) by a factor of 1.5. Such a layer completely compensates for the increase in radiation intensity caused by a void in concrete (void \(2300\) kg/cm\(^3\)) with linear dimensions along the beam of \(8.6\) cm [6]. For radiation of lower energy the difference becomes still greater. In addition, irradiation has to be carried out through a metal form. This leads to the need to select a gamma-radiation source of higher energy and thereby reduce the sensitivity of the method in detecting defects in concrete.

From an analysis of the microscopic cross sections for the interaction of neutrons with iron and the constituents of concrete [7, 8], it may be assumed that, in inspecting reinforced-concrete structures with the aid of neutrons, it is possible to use that region of neutron energies in which the attenuation coefficients of neutrons for steel and concrete will be equal or, at least, close to one another. In this case the adverse influence of the metal reinforcement will be weakened, and the sensitivity to the detection of air voids will be sufficiently high.

To confirm these propositions, the dependence of the effective attenuation coefficients for iron and concrete was obtained. By changing the discrimination threshold, we thereby select the effective energy of the recorded neutrons. The geometry of the experiment is shown in Fig. 1. Here 1 is the Po—Be fast-neutron source; 2 is the absorber; 3 is the paraffin collimator; 4 is the fast-neutron detector.

The steel absorber consisted of plates measuring \(500 \times 500\) mm. The concrete specimens were made in the form of blocks measuring \(300 \times 300\) mm. To reduce the influence of boundary effects, the absorber was supplemented with shielding concrete blocks of the same composition. The effective attenuation coefficients were calculated on the basis of the exponential character of the attenuation of the fast-neutron flux in the absorber. The results are illustrated in Fig. 2 [9]. The total thickness of the absorber was brought up to 350 mm. In this case, the dependence of the effective attenuation coefficient on the absorber thickness was insignificant. The results of determining the attenuation coefficients agree well

Fig. 1. Geometry of the experiment

Fig. 2. Dependence of the effective neutron attenuation coefficients in iron and concrete

with the data of [10], where neutron attenuation in concretes of various grades was studied using an “all-wave” counter. The error in measurements with the “all-wave” counter is estimated at 2.5%. It is seen from Fig. 2 that the ratio of the effective attenuation coefficient for steel \(\mu_{\mathrm{eff.st}}\) to the effective attenuation coefficient for concrete \(\mu_{\mathrm{eff.con}}\) varies over wide limits depending on the effective energy of the recorded radiation. When the ratio

\[ \frac{\mu_{\mathrm{eff.st}}}{\mu_{\mathrm{eff.con}}} \approx 1 \]

the presence of reinforcement in the structure does not affect the detectability of air voids. As a result of the experiments, defects were recorded in reinforced-concrete slabs with dimensions of \(15 \times 15 \times 15\) mm at a total structural thickness of 300 mm. The signal-to-noise ratio in this case was 3. The defects were simulated by means of the concrete slab shown in Fig. 3.

The defect simulator was moved perpendicular to the source–detector axis at a speed of 0.5 m/min. The width of the inspected section was 30 mm. In the experiment, a Po—Be neutron source with a yield of \(10^7\) n/sec was used. In selective inspection of structures at the most dangerous points, the activity of the source can be reduced by at least an order of magnitude by increasing the measurement time. By setting a discrimination threshold that provides a large ratio

\[ \frac{\mu_{\mathrm{eff.st}}}{\mu_{\mathrm{eff.con}}}, \]

the flaw detector can be used for inspection control of embedded parts and reinforcement in reinforced-concrete structures in order to search for hidden

metallic utilities, etc. However, the advantage of a neutron flaw detector over a gamma one will probably appear only at large thicknesses, and the advisability of using it for these purposes must be considered additionally.

Another problem where the use of a neutron flaw detector with direct recording of the transmitted flux of fast neutrons is preferable is the determination of concrete moisture content and the monitoring of aggregate moisture content during the preparation of concrete. Continuous monitoring of aggregate moisture makes it possible to maintain the required ratio and to dose correctly the water being added. The method of determining moisture with the aid of neutrons has proved to be the most suitable and accurate for these purposes, making it possible to measure moisture in a short time with an accuracy of up to 1%. Thus, under laboratory conditions, with a Po–Be source of activity 100 microcuries, measurement of moisture content in the range from 3 to 22% with an accuracy of ±0.5% is carried out in 2 min [11].

Fig. 3. Flaw-detecting slab

Most concretes used for biological protection against ionizing radiation contain boron and other elements with resonance neutron captures, and the use of fast neutrons excludes their influence when determining moisture content.

REFERENCES

1. E. Pol’. Nondestructive Methods of Testing Concrete. Stroiizdat, Moscow, 1967.

2. V. A. Vorob’ev. Candidate’s dissertation. TPI, Tomsk, 1966.

3. Radiation Technology. Collection of papers of the All-Union Scientific-Research Institute of Radiation Technology. Issue 2, Atomizdat, Moscow, 1968.

4. L. M. Anan’ev et al. “Industrial Construction,” No. 11, pp. 44–45, 1957.

5. O. I. Leipunskii et al. Propagation of Gamma Quanta in Matter. Fizmatgiz, Moscow, 1960.

6. Instructions for Working with Radioactive Substances and Other Sources of Radiation. Publishing House of the USSR Academy of Sciences, Moscow, 1969.

7. M. I. Shal’nov. Tissue Dose of Neutrons. Atomizdat, Moscow, 1960.

8. I. V. Gordeev et al. Nuclear-Physical Constants. Handbook. Atomizdat, Moscow, 1963.

9. V. I. Gorbunov et al. Investigation of the attenuation of fast neutrons in reinforced concrete. Collection of papers of the Second Conference on Radiation Methods of Control, Tomsk, (in press).

10. I. I. Avaev et al. In: Problems in the Physics of Reactor Shielding. Atomizdat, Moscow, 193, 1963.