Measurements and data - Induced Polarization

Introduction

There are four techniques for observing chargeability: two using a charging current that is turned abruptly off so that discharging can be observed in the time domain, and two in which the dispersive nature (or effect as a function of frequency) of the phenomenon is observed in the frequency domain.

Two types of time domain data

Consider the experiment illustrated in the following figure. Current is injected into the ground at the I source electrodes and voltage is measured at the V potential electrodes. The source is DC (direct current) in the sense that when it is on, there is no variation. However, in this case it is turned on and off with a duty cycle as shown in the figure. Two methods of measuring chargeability in the time domain are described below.

1. The following simple definition of chargeability is less commonly employed, since it is impractical to measure. The figure to the right shows voltage measured when the transmitter is first turned on and then turned off some time later. Using parameters from this figure, one definition of chargeability is M = V_s / V_m , where V_s and V _m are the maximum and "secondary" potentials, respectively.
• The leading edge potential, V_σ , is what would be measured in the absence of chargeability. This potential would yield the ground's resistivity.
• The maximum potential is the combined effect of current flowing in the ground and charges built up under the influence of the imposed electric field.
• The so-called secondary potential is entirely due to the charge imbalance resulting from the build-up of charge.
• Using this form, chargeability M will be 0 ≤ M < 1. If M = 0 the measured potential will follow the input current waveform exactly with no charging or discharging involved, as shown in the first column of the figure above.
  
2. The most commonly measured form of time domain IP is the normalized area under the decay curve. It can be represented by the following equation, using parameters specified in the adjacent figure. The decaying potential that follows V_s is written as V_s(t).

Chargeability M is essentially the red area under the decay curve, normalized by the source voltage.

Two types of frequency domain data

An oscillating source current can be employed to observe chargeability. The measurements are often still referred to as "DC resistivity" because the frequencies are relatively low. The resulting data will include (i) a "DC resistivity" based upon the voltages measured with the lowest source frequency, and (ii) a chargeability based upon the measurements explained next. Two methods of measuring chargeability in the frequency domain are described below.

1. If potential-signal amplitude is measured at two frequencies, a measure of chargeability is acquired, and be expressed as units of "percent frequency effect" or PFE. Since the ground has less time to respond at higher frequencies, the signal is expected to be smaller. Expressions for PFE are shown in the equations below. The data used in this calculation are illustrated in the figure below. Recall that _a = K |V| / I, where K is the geometric factor based upon electrode geometry (see the Geophysical surveys chapter, "DC resistivity" section), V is the measured potential, and I is the source current.



Alternatively:

If the voltage version is used, the Frequency Effect (FE) can easily be converted to a percent frequency effect by multiplying by 100.

2. Data with units of phase are gathered by transmitting a sinusoidal source current. Then the phase difference between this source and measured potentials is recorded as a measure of chargeability. Units are usually milliradians. The following figure illustrates:

Relating the four types of data

The different IP responses all result from the build up of polarizing charges, but they do not produce the same numbers. In fact, the units of the various measurements are different. Nevertheless, the following approximate rule of thumb allows conversion between the different data sets:

A chargeability of M = 0.1 is approximately equal to 10PFE, or 70mrad, or 70msec.

Data acquisition

Time domain IP

As noted above, when time domain IP is recorded, chargeability M  is measured as the area under the decay curve normalized by "primary" voltage V_P,  using .  The t₁ and t₂ times may be any limits within the off-time, and there are not really any standards, so comparison of different surveys can be difficult.

Source (input) current is a square wave with 50% duty cycle (equal on and off times) as per resistivity (repeated cycles of +on, off, -on, off). The use of positive and negative cycles in transmitter current is very important for time-domain IP work. The correct area under the decay curve will be measured only if the potential decays exactly to zero. This will not occur when there is a superimposed spontaneous potential (SP), which is usually the case. If only one polarity was used, the inevitable SP could not be detected and removed. Recording both positive and negative cycles allows the "off-time" potential (i.e. voltages recorded when the transmitter is off) to be estimated, and any non-zero component removed.

Many instruments record measured voltage, V_p, just before the transmitter is turned off, and then again 10 times while voltages decay during the off times. The results can then provide a calculated chargeability and an estimated spontaneous potential. The adjacent figure illustrates each measured parameter. Note that if the transmitter is not on for a long enough time, V_p will be measured before the charging time is finished, resulting in a voltage that is smaller than the actual V_p.

Other instruments use alternative time windows, and some newer instruments digitize the whole waveform, but the fundamental concepts are the same for all time domain systems.

Frequency domain IP

The percent frequency effect was defined above as either

or

The low frequency can be measured using DC or very low frequency, and the second type of resistivities can be measured at frequencies on the order of a few tens to hundreds of Hertz. In practice, measurements are often made at two or more frequencies, and a PFE is calculated from the results.

Phase IP

When the phase of voltage with respect to input current is measured directly, the impedance of the ground can be determined based on the material. However, this requires careful synchronization between the receiver and the transmitter, and may be difficult on large surveys in rough country.

Metal factor

Metal factor is a parameter given by PFE or chargeability, M, divided by the corresponding apparent (i.e. measured) resistivity. Plots of this parameter emphasize where both low resistivity and high chargeability exist, or where there are significant occurrences of metallic mineralization (or graphite). However, metallic minerals (such as sulfides) are often disseminated, in which case the ore will more likely have high resistivity correlated with high M.  For this and other reasons, the use of metal factor has declined and it is now rarely used.

Choice of time, frequency or phase measurements

Here are a few factors affecting whether to choose time domain or freqeuncy domain survey types:

• Time domain methods may provide more information, and will be less susceptible to induction effects (see the appendix on noise). However, decay curves are often shown in small units, such as millivolts or microvolts, so signal-to-noise ratio can be a problem. Stacking many repeat measurements can help, but this practice is expensive because it requires a longer investment in field time.
• Frequency domain methods require significantly smaller source currents and are less sensitive to most sources of noise. However the effects of EM coupling can be severe, and unless these are removed, freqeuncy domain and phase measurements may contain no usable chargeability information at all. ("EM coupling" is an un-wanted signal which arises from inductive interactions (like a transformer) between conductive near-surface ground and the wires carrying transmitter current. It can completely hide IP effects when it is severe).
• See Smith, 1980, for a comparison of time domain and frequency domain results recorded using three different instruments over the same ore body. Such studies are rare because of the cost, so this is an interesting examination of the pros and cons of various ways of measuring IP.

References

1. Smith, M.J., 1980, Comparison of induced polarization measurements over the Elura orebody, The Geophysics of the Elura Orebody, Cobar NSW, ASEG, 1980, 77-80.