In view of the steadily growing world population and the increased residential density in cities, the requirements for efficient underground and overground infrastructures are becoming more and more demanding. Tunnel boring machines (TBMs) are used for the realization of tunnel systems for metro, road and rail traffic for the purpose of passenger transport as well as for the supply and disposal of waste water. Due to different geological conditions and the associated hardness of the soil depending on its quartz content, wear and adhesion degrade the cutting tools of a TBM, causing cost-intensive downtimes. To avoid this, the soil to be removed must be conditioned in a targeted manner by applying foam (consisting of water and surfactants) to soften the soil.

 
The aim of the feasibility study in preparation for a subsequent cooperative project is to demonstrate the successful classification of the soil type (e.g. sandy or clayey) of the overburden material of a TBM, as well as its water content by using prompt gamma neutron activation analysis (PGNAA). To this end several PGNAA measurement campaigns with realistic reference samples are carried out and analyzed regarding their chemical composition. Based on these results, a compact measuring system based on PGNAA will then be developed for a conveyor belt, which will implement these two tasks and monitor them online. In a subsequent joint project, the measuring method will be further developed for specific applications in order to solve the tasks mentioned above.

In AiNT's technical center, experimental measurement data are obtained by investigating realistic samples from the overburden material of a TBM using an existing PGNAA measurement system. As illustrated on the left panel of the figure, atomic nuclei inside the sample are excited to a higher energy level. Deexcitation to a stable nucleus occurs under emission of prompt gamma radiation which is detectable in suitable particle detectors. The fact that the energy of the emitted gamma radiation is characteristic for each chemical element allows for a non-destructive identification of the elements present inside the sample. Because the time scale of this nuclear reaction is in the range of femto seconds it is possible to obtain results much faster and on-site compared to a chemical analysis in a laboratory. AiNT develops measuring systems based on PGNAA, which allow for a non-destructive elemental analysis of a wide range of materials without prior sample preparation. As part of the feasibility study for the TBM analyzer, the proportion of the respective soil type (e.g. quartz, calcite, gypsum, clay) is reconstructed from the measurement data by applying special AI methods, speeding up the analysis even more. The water content can be determined by quantifying the hydrogen contained inside the sample. Depending on these results, parameters such as the speed of the cutting tools, the feed rate of the TBM and the amount of foam added during conditioning can be determined in order to reduce tool abrasion on the TBM to a large extent.