Re: sensor to measure the conductivity of mud

[Martin]

NOTE:  To get off this list, send email to majordomo@aqua.ccwr.ac.za
with the body of the message containing the line:
unsubscribe sowacs

Hello Bruce

Thanks for the comments - feel free to use the material as you wish. It's
mainly drawn fron "The safe use of marginal water in agriculture: Guide for
the water resource planner." (1997) C L Abbott & N J Hasnip HR
Wallingford/DFID Report OD 140 

I didn't quote verbatim, so there shouldn't be any copyright problems.

[Those less technically interested had better opt out now! 
However, there are some interesting issues raised in regards different
"editions" of Thetaprobes and their modes of operation. This raises the
importance of sensor manufacturers fully declaring the exact details of each
"vintage" of sensors, so that we may compare apples with apples when
different workers use these commercial sensors over different years.  Bruce :)]
 
To answer the other query, on why it's important, will probably take you
into the discussions on effects of magnetic treatment on water.
More is being understood about this phenomena. Application of an
appropriate electro-magnetic field of specific frequency has an influence
on the arrangement of dissolved ions, within water flow through this field. 

In the area of soil water measurement by "tdr", there appear to be some
related anomalies. Dr Max Hilhorst and his colleagues have been working on
this topic for a long time. Dr Hilhorst kindly provided a copy of his
thesis for me to read. At the same time, I read another thesis of a student
at Newcastle upon Tyne. Their conceptual pictures seem to differ, in
relation to  behaviour of ions in solution at microscopic dimensions. I
passed on some queries to Dr Hilhorst, based on my reading of both sets of
work. So far Dr Hilhorst has not had opportunity to respond. Some of those
queries are listed below.

................I have tried to get a consistent picture, from reading your
thesis, that of Moijd, the paper of Wyseure et al and from some of my own
information. This has raised a series of questions, which I hope that you
might answer.

1 Counterion diffusion and the Maxwell Wagner effect.

You distinguish these phenomena, on the basis of their relaxation
frequences, viz counterion diffusion (1-100 kHz) and Maxwell Wagner (0.1 -
500 MHz). Moijd (and probably others) develops a relationship to describe
bulk electrical conductivity sb based on the electric double layer concept.
He measures bulk electrical conductivity from the attenuation of a tdr
signal and relates this to electrical conductivity of pore water s, when
the latter has been measured separately. For saturated soils, he obtains a
number of similar linear regressions of sb (0-7 dSm-1) on s (0-28 dSm-1).
The wave frequency of his instrument was judged to be 145 MHz. So agreement
of a theoretical model for sb vs s could only be coincidental, since
counterion diffusion effects on dielectric determinations (and hence bulk
conductivity estimates) would not be significant at these frequencies?
   

In p83-92 of your thesis, you specifically relate sb and s, by
consideration of the Maxwell-Wagner effect. I find figure 4.9 a bit
puzzling. This shows ionic conductivity of soil solution to be independent
of water content. Earlier on p24 you said "Note that for a soil drying as a
result of evaporation, the concentration of ions in the pore water will
increase, and so do s and fMWr." 

2 "Apparent dielectric constant" vs bulk or solution electrical
conductivities.

Wyseure et al highlight a finding of Dalton (1992) (SSSA special public.)
that the dielectric constant (permittivity) of electrolyte solutions
decreases with increasing conductivity (Robinson & Stokes, 1959), while in
saline soils the dielectric constant  (permittivity) increases with
increasing electrical conductivity. Measurements by Moijd using tdr clearly
show his measured "apparent dielectric constant" increasing with increasing
electrical conductivity determinations for solutions (0-6.5 dSm-1). So that
water content in saline soils measured by tdr is over-estimated.

Gaskin & Miller (1996) describe the development of their fd sensor (Theta
probe) and it's calibration. Instrument output voltage is linked to square
root of "apparent dielectric constant". Their measurements also showed a
reduction of output voltage (and hence estimate of apparent dielectric
constant) with increasing levels of ionic conductivity, in beakers of
increasing concentrations (0 to 20 dSm-1) of potassium chloride.  The
instrument was reported to have a 100 MHz oscillator.

Parkes et el (1997 unpubl) describe a correlation between an fd (Theta
probe) voltage output and electrical conductivity of soil saturated paste
extract (0-20 dSm-1). This shows increasing voltage (and hence "apparent
dielectric coefficient") with increasing soil electrical conductivity (and
hence bulk electrical conductivity) for sb > 4 dSm-1. The commercial sensor
used in 1996 might have a somewhat different operating frequency to that
tested by Gaskin & Miller. How else might the contradictory results with
the fd sensor be explained?  

It is the Maxwell Wagner effect, shown in Fig 2.8 of p25, which explains
the increase of either "apparent dielectric constant" or real component of
permittivity e' with bulk soil conductivity sb .  This is the case for
fixed water content -and occurs despite the fact that e' does not vary with
ionic conductivity s - (viz fig 3.12 of thesis), when using fd or tdr
sensors. These assumptions are true?

3 Instrument calibration.

In the notes you gave me you said that the current instrumental operating
frequency of the IMAG sensor is 30 MHz. You also suggested that the
auto-calibration feature is not appropriate. Can instruments be calibrated
along the lines of equation 2.60 of your thesis, if a soil water
characteristic is available for the soil type? Presumably this would be
combined with measurements on the soil, for the 3 measuring frequencies.
Alternatively, if K in eqn 2.65 is a function of soil texture and ionic
conduction, then can the results in Figure 4.9 be generalised to get a
rough relationship?

As regards temperature, do you recommend simply making measurements at
times when soil temperature is comparable to calibration temperature?      



.................. A puzzling topic which has challenged a lot of able
people for quite a long time!

regards

Martin  

WasteTrim at Technology Transfer Centre, Alrick Building,
Mayfield Road, Edinburgh EH9 3JL, SCOTLAND 
tel (44) 131 472 4708
fax (44) 131 662 4678