Bundesanstalt für Materialforschung und -prüfung

International Symposium (NDT-CE 2003)

Non-Destructive Testing in Civil Engineering 2003
Start > Contributions >Lectures > Electrical Methods: Print

Spectral Induced Polarization - a tool for non-destructive testing of soils and building materials

Ernst Niederleithinger
Federal Institute for Materials Research (BAM), Berlin, Germany


The Spectral Induced Polarization (SIP) is a newly developed non-destructive geophysical method. It is an extension of conventional electrical resistivity measurements. By measuring additional parameters a more detailed investigation of the subsurface is possible. Up to now SIP is mainly used in mineral exploration, but the number of application examples from near surface geophysics and civil engineering is increasing. Results are presented from the investigation of slag heaps and measurements on bricks.

1. Introduction

Electrical methods for corrosion and moisture detection are well established in civil engineering [Ref 1]. But due to the fact that the electrical resistivity is depending on many factors (matrix material, porosity, salt content, moisture, temperature and others) data interpretation is often difficult. The same problem arises in the use of electrical methods in geophysics. During the last decades much effort was spent on the use of more sophisticated electrical methods. Most promising is the SIP method, which is used for example to discriminate minerals in mining geophysics.

2. Method

Most electrical methods use separate electrodes for current injection and voltage measurements to avoid the influence of contact resistances (figure1). The apparent resistivity ra is calculated by

Fig 1: Principle of electrical measurements.


Fig 2:
Principle of 2D multielectrode electric measurements and data processing.

Fig 3:
Field layout of multichannel SIP equipment (photo: www.radic-research.de).

K (configuration factor) is dependent on the electrode arrangement at the surface. The apparent resistivities ra are integral values over a certain volume in the subsurface dependent on the electrode arrangement and the true resistivity distribution. Large electrode distances lead to large penetration depths and large volumes. The resistivity r at a certain point in the subsurface has to be reconstructed from large data sets of apparent resistivities ra from different positions and penetrations depths via model calculations or inversion schemes (figure2).

The Spectral Induced Polarization (SIP) method is very similar to AC impedance spectroscopy used for investigation on curing concrete or chloride diffusivity [Ref 2,Ref 3]. A sine current is applied to the body under investigation, The amplitude and the phase lag of the resulting voltage is measured (figure 4, left). This is repeated with discrete frequencies between 1mHz and 10kHz. The resulting frequency spectra of resistivity (figure4, right) can be fitted by model formulas derived from electrochemistry or equivalent electrical circuits. Most common, but not always applicable is the Cole-Cole-model formula [Ref 4]:

Fig 4: Principle of SIP measurements: a sequence of time series is used to produce the resistivity spectrum.


r 0 : DC-resistivity [Wm]
w : circular frequency (= 2pf) [Hz]
m : chargeability
t : time constant [s]
c : frequency dependance

The Cole-Cole-parameters can be interpreted of in terms of material parameters. For example in many cases the time constant t correlates with the mean grain size [Ref 5].

3. Measurements on slag heaps

Figure5 shows a slag heap in Lower Saxony consisting mainly of iron oxides. It was investigated in a research project conducted by the German Federal Institute for Geosciences and Natural Resources (BGR). The focus was on crusts which are developing on the surface of such heaps [Ref 6,Ref 7]. They were supposed protecting the heap from the ingression of rain water and by this way from solution of contaminants.

Fig 5: South slope of slag heap: Measurements described below are from profile 1 (photo: Buero fuer Geophysik Lorenz).
Fig 6: Data at 12 Hz of SIP section on heap slope. Left: Amplitude of apparent resistivity at 12 Hz. Right: Phase, same frequency. Data courtesy of BGR.

To proove whether such crusts can be detected by geophysical measurements, a SIP-section was conducted on the south slope (27electrodes, 0.25m apart, frequency 0.1 - 10000Hz) [Ref 6]. The measured data at 12Hz are shown in figure6.

The data show high apparent resistivities and high phase values at the surface and a steady decrease to the heaps interior. But data interpretation by model calculations led to a model (figure7), which shows an intermediate layer with increasing resistivity and high chargeabilitym. This is due to an older crust from an earlier stage of material deposition, now covered with newer material. The site owner confirmed the existence of such layers. The parameters of the major model blocks are listed in table1. The older buried crust (block2) is identified by a chargeability value similar to the top crust.

Fig 7: Model for SIP section on slagheap slope. Parameters see table 1.

Block no. r[Wm] Chargeability m [s]
Table 1: Model parameters for SIP section (figure7).

Figure8 shows a test pit in the slagheap, 20m from the SIP section. The hard top crust is clearly visible; its thickness is about 0.7m. Beneath the crust is a layer of very loose, slightly wet material. Slope instability inhibited digging deeper.

Fig 8: Cut into uppermost layers of slagheap (photo: Buero fuer Geophysik Lorenz).

4. Measurements on bricks

A preliminary study on SIP measurements on bricks has been carried out recently [Ref 8]. Main questions have been:

  • Are reliable measurements possible?
  • How much does moisture influence the measurements?
  • Have different bricks different SIP response?

Fig 9:
SIP-Measurements on bricks. A: measurement device, B: sample with AgCl-Electrodes, C: control computer, D,E: other samples.

A simple measurement setup was realized with a SIP-Mini device (Radic research, courtesy of the Federal Institute for Geosciences and Natural Resources) and Ag/AgCl electrodes from a medical supply store (figure9). It was found that measurements with less than 1% standard deviation in amplitude are possible unless the bricks are completely dried out.

Three different bricks (clay, lime-sand and aerated concrete) have been put in rainwater for seven days. After wetting resistivity spectra were measured once a day. The resistivity amplitude for a certain frequency is shown in figure10. All curves show similar shape, but a different level. Resistivity increases monotonic with time more than one decade. The resistivity range overlaps between bricks. Thus data interpretation of amplitude at a single frequency (or DC) would be ambiguous in real cases.

Fig 10: Resistivity increase during drying of bricks.

Figure11 shows the resistivity spectra of three different bricks transformed into Argand diagrams. In an Argand diagram the real and imaginary parts of resistivity are plotted against each other. Different types of spectral behavior can be easily seen here. All curve show different forms in the low frequency and the high frequency part. The steep decrease for high frequencies is possibly to capacitive coupling between cables and bricks. More interesting is the low frequency part. Between 0.09 and 23Hz less than 0.5% resistivity change for the aerated concrete, about 1% for the lime brick and about 3.5% for the clay brick. In addition the curvature is higher for aerated concrete than for the bricks. These results show that there might be a chance for the development of an electric non-destructive technique for the investigation of masonry, its moisture content and the detection of certain brick types in plastered walls. Similar polarization effects have been detected in wood [Ref 9].

Fig 11: Argand diagram of SIP measurements for different brick types. Each brick has specific spectral characteristics.

5. Perspective

The SIP method proves to be a promising tool for the investigation of soil and building materials. It delivers information on constituents, structure and moisture beneath the surface. But still there is a high demand for systematic investigations on the influence of different physical (for example pore size and form) and chemical (for example salt content) factors on the results.


  1. Schickert, G., Henschen, J., Krause, M., Maierhofer, C., Weise, F., Wiggenhauser, H. & Borchardt, K.: ZfPBau-Kompendium, 1999. http://www.bam.de/service/publikationen/zfp_kompendium/welcome.html
  2. Shi, M., Chen, Z. & Sun, J.: Determination of chloride diffusivity in concrete by AC impedance spectroscopy. Cement and Concrete Research 29 (1999) 1111-1115
  3. Knoedel, K., Krummel, H. & Lange G.: Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten. Band 3: Geophysik. Springer Verlag, Berlin, 1997.
  4. Luo, Y. & Zhang, G.: Theory and Application of Spectral Induced Polarization. Geophysical Monographs, SEG, Tulsa, 1998.
  5. Pelton, W.H., Ward, S.H., Hallof, P.G., Sill, W.R. & Nelson, P.H.: Mineral discrimination and removal of inductive coupling with multifrequency IP,1978. Geophysics 56, p. 941-950.
  6. Buero fuer Geophysik Lorenz: Wiederholungsmessungen mit der Methode der Induzierten Polarisation zur Erkundung von Bergbau- und Schlackehalden. Report to the Federal Institute for Geosciences and Natural Resources (BGR), Hannover (unpublished), 1999.
  7. Niederleithinger, E., Grissemann, Ch. & Rammlmair, D.: SIP Geophysical Measurements on Slag Heaps: a New Way to Get Information About Subsurface Structures and Petrophysical Parameters. Proceedings of ICAM Congress, Göttingen, 2000.
  8. Niederleithinger, E. & Weller, A.: SIP: Messung des komplexen elektrischen Widerstandes von Beton und Mauerwerk - Ansätze zu einer neuen Prüfmethode. Proceeding of DGZfP conference, Weimar, 2002. http://www.ndt.net/article/dgzfp02/papers/v47/v47.htm
  9. Thierry, B., Weller, A., Schleifer, N., Westphal, T.: Polarisation effects of wood. Proceedings of 7th Meeting of EEGS-ES, Birmingham, 2001.
STARTPublisher: DGfZPPrograming: NDT.net