Measurement System m:explore.eis

Impedance spectroscopy (also dielectric spectroscopy or electrochemical impedance spectroscopy) is a non-destructive, fast and low-cost measuring method that can record electrical material parameters, such as impedance, conductivity and permittivity. These parameters are frequency-dependent and vary based on the frequency of the alternating electric fields they are measured with.

Ultra-wideband impedance spectroscopy allows for insight into the examined materials that would otherwise not be obvious at first glance. This can be interesting for users that would like to have information that goes beyond conventional sensors. Various process parameters are reflected in electrically-measured quantities and can be captured with UWB technology.

In some cases, the direct measurement of one parameter is either not possible or cannot provide the required information. In these scenarios, the information can be derived through post-processing or a combination of the measured data.

An electromagnetic (EM) wave is emitted from the applicator into a substance.
The amount of the transmitted, absorbed or reflected signal gives insight into the material properties of the substance.

The interaction between the substance and the signal is based on the atomic or molecular building blocks of the substance and the frequency of the electric field of the propagating signal.

The frequency-dependent material parameter permittivity ε_r describes the transmissibility of an EM wave through a substance and is measured over a broad frequency range with a single measurement. This allows for a differentiation between very similar substances that is not possible with a single narrowband measurement.

A typical behaviour is found for a given substance component. Just as a fingerprint can be assigned to a specific human being, a certain substance component can also be assigned to an impedance spectrum. The wider the frequency interval over which the measurements are carried out, the more clearly the assignment succeeds.

The impedance spectrum of an unknown mixture of substances is measured through a superposition of the impedance spectra of the (known) individual components and leads to an indirect conclusion for the material composition (usually through machine learning methods).

At Ilmsens, the impedance spectroscopy consists of the following three parts:

1. The ultra-wideband measurement electronics, which are the same for both impedance spectroscopy and near-field radar (e.g. m:explore)
2. Applicator (sensing probe) that interacts with the material that is analysed
3. Data evaluation algorithms
   

Since every application is unique, every process requires different methods of interaction with the measured substance. We discuss the possibilities with our partners and the options include but are not limited to the following specific requirements:

• Size, available space
• Position of implementation
• Form factor
• Robustness

There are fundamentally different setups:

• In-line or bypass setup
• Flow-through or immersion probe

In the impedance spectroscopy, the accuracy of results depends on the following:

• Substances composition
• Complexity of substance compositions
• 
• Frequency range of impedance measurement and
• Quality of analysis software (scope of "learning")

If substance components have distinctive permittivity values (see graphic, values circled in red), they are easy to distinguish and even slight changes in the respective concentrations become apparent. Measurements can demonstrate detectable concentration changes within a few ppm.

If components of a liquid show similar permittivity values, only larger concentration changes become detectable.

In order to illustrate which materials are similar in their electric properties, the following table displays several exemplary materials and their permittivity values based on the indicated sources.

In contrast to some sources, it should be noted that permittivity values are not constant, but rather temperature and frequency dependent. Therefore, the permittivity values in the table can only provide a first indication.
Example: Water (ε_r = 80) and oil (ε_r = ca. 3) can be differentiated very well. Even small concentrations can be easily detected.

Name	εr
Acetone	21.5
Acetic acid	6.2
Ammonia	15.0 … 25.0
Ascorbic acid (vitamin C)	2.1
Gas	2.0
Cotton-seed oil	3.1
Brewer's yeast	25.0
Chloroacetic acid	33.4
Diesel fuel	2.1
Ethanol (ethyl alcohol)	16.2
Fatty acid	1.7
Fish oil	2.6
Hydrochloric acid	5.0
Isopropanol	18.0
Methanol (methyl alkohol)	33.0
Mineral oil	2.0 … 2.5
Palm kernel oil	1.8
Soy-bean oil	2.9 … 3.5
Transformer oil	2.2 … 2.4
Vinegar	24.0 / 37.6
Water (20°C / 120°C)	80.0 / 20.4

Impedance spectroscopy helps to detect the composition of mixtures of substances, such as:

• The composition of liquids (e.g. milk)
• The condition of industrial processes that include liquids (foods, chemistry, machinery (oils), industrial cleaning, etc.)
• Harmful substance inputs (e.g. water in engine oils, cooling lubricant in subsequent process steps)

Impedance spectroscopy can also be applied to identify properties of substance mixtures, such as:

• Material moisture
• Curing degree of adhesives or building materials
• State of cell emulsions
• Tumour tissue

If you have another application in mind, please feel free to contact us any time.

The main advantage is the ability to differentiate between components of a complex substance: Through ultra wideband signal measurements, the typical behaviour is found for a given substance component. Just as a fingerprint can be assigned to a specific human being, a certain substance component can also be assigned to an impedance spectrum. The wider the frequency interval (impedance spectrum) over which the measurements are carried out, the more clearly the assignment succeeds.

Two substances might be similar in a limited frequency range, but the wider the measured frequency range, the more likely it is to find differences in a specific frequency range.