GB2359138A - Method and apparatus for monitoring adsorbed water - Google Patents

Abstract

A method and apparatus for monitoring adsorbed water and in particular for monitoring adsorbed water in hardening and formed concrete and on concrete aggregates ready for mixing. The method of monitoring adsorbed water in a material includes the steps of providing a sample of the material between dielectric probes (C1,C2), and determining the impedance (capacitance) of the sample in the presence of an alternating electric field, the material having a water content below its saturation level. The method separates the real and imaginary components of the capacitance measurement and uses the imaginary component to determine the enhanced dielectric effect of surface adsorbed water. Phase sensitive detector means (70) are adapted to detect the sample impedance at a consistent point of cycles of the alternating electric field.

Description

2359138 1 - METHOD AND APPARATUS FOR MONITORING ADSORBED WATER

FIELD OF THE INVENTION

This invention relates to a method and apparatus for monitoring adsorbed water, and in particular for monitoring adsorbed water in hardening and formed concrete and on concrete aggregates ready for mixing.

BACKGROUND TO THE INVENTION

Accurate and rapid determination of the moisture content of concrete has long been considered important. Industry observers believe that the properties of concrete vary with moisture content, and in particular that the proportions of water and cement present in a concrete mix greatly affect the physical characteristics of the hardening concrete.

Faults in many concrete structures once built are impossible to rectify, yet during construction the concrete often has to be prepared and poured in difficult conditions. As a well known example, the concrete for offshore oil-drilling platforms needs great strength, durability and water tightness, which means that a close check should be kept not only on the constituents prior to mixing, but also even during a pour so that the mixture can be regulated as necessary. Subsequent measurements to check for possible deterioration are also desirable.

Distress in hardened concrete is frequently the result of excessive water or inadequate cement content. Recent studies on the corrosion of reinforcing steel in concrete have revealed that in addition to providing a beneficial effect on the strength of hardened concrete, a low water to cement ratio also ensures a long service life for reinforced concrete structures exposed to de-icing salts. The importance of determining the setting time of concrete has 2 also long been apparent. This recognised need has itself resulted in a demand for suitable means of monitoring the available water, as a preliminary to controlling the water to cement ratio of concrete made both under site and under 5 factory conditions.

The possible variation in the amount of water contained in the aggregates is believed by many to be the main factor which has to be considered in concrete quality control.

Thus, the importance of monitoring the water to cement ratio has been suggested by a number of studies.

Laboratory and on-site investigations confirm that deleterious chemical reactions between the aggregate and the surrounding cement paste are an important reason for concrete deterioration. The most common reaction is believed to be that between the susceptible silica constituents of the aggregate and the alkalis in cement. The reactive forms of silica are opal (amorphous) chalcedony (kryptocrystalline fibrous), and tridymite (crystalline). The reaction starts with the attack on the siliceous minerals in the aggregate by the alkaline hydroxides derived from the alkalis (sodium oxide and potassium oxide) in the cement. As a result alkali-silicate gel is formed and alteration of the borders of the aggregate take place. The gel is of the "unlimited swelling" type i.e. it imbibes water with a consequent tendency to increase in volume.

Since the gel is confined by the surrounding cement paste, internal pressures result, and eventually lead to expansion, cracking, and disruption of the cement paste (pop-outs).

Factors influencing the progress of the alkali-aggregate reaction will include the availablility of non-evaporable water in the paste. As little as 0.5% of defective aggregate can cause damage. The aggregate is harmful if under test a sample prepared from the aggregate expands more than 0. 05% af ter 3 months or more than 0. 1% af ter 6 months; a considerable delay occurs before judgement on the 3 aggregate can therefore be pronounced. A rapid and conclusive test for aggregate reactivity is required, including the important measure of available water content.

DISCLOSURE OF THE PRIOR ART

Water content has been determined simply by a microwave-oven drying method in a laboratory, whereby a sample of fresh concrete mix is heated in a microwave oven for about 30 minutes to remove the water, and the weight loss checked. It has also been proposed to measure microwave transmission properties of fresh concrete mixes to seek a correlation between the energy demand (voltage drop) and the water content; however, this method is also only really suitable for laboratory use, and for thin samples.

The electrical conductance of concrete samples has also been compared. The application of an alternating electrical field to a sample of hardened cement paste results in a conductance not only dependent upon the water distribution within the matrix, but also upon the ionic conductivity between electrodes and so is affected by the concentration of ionic species (such as Ca2+ 11 OH- and Na+) in the capillary pore water solution, which concentration changes with time as the concrete sets and hydration products are formed.

It has also been suggested that the AC reactive element of a measurement will be related to polarisationmechanisms operative within the paste, dominated probably by Maxwell Wagner (interfacial) effects, with adsorbed water relaxation processes also making a contribution. Earlier workers considered that water distribution within the hardened cement paste would therefore have a significant influence not only on electrical conduction measurements of a concrete sample but also generally on its measured dielectric value 4 - remarking that the dielectric constant of dry cement paste is between 2 and 5 whilst that of bulk water is about 80.

Society For Testing A study of the moisture content of hardened concrete by the measurement of its dielectric properties was reported in Volume 63 Pages 997-1007 of The Journal Of The American Materials (ASTM Proceedings), but without a recognition of the importance of utilising the imaginary component of surface adsorbed water at below the saturation level.

SUMMARY OF THE INVENTION

We seek an accurate and rapid method and apparatus for moisture measurements in aggregates, and curing and cured concrete, and for curing measurements in hardening concrete.

We have recognised that most naturally occurring aggregates, such as siliceous and calcareous aggregates, exhibit chemically unsaturated surfaces with a pattern of surface electrical charges, and that these surfaces are capable of adsorbing large quantities of water. We have further recognised that as water is adsorbed onto the charged surfaces, hydrogen bonding between molecules appears to be greatly reduced, and the lack of inter-molecular hydrogen bonding greatly reduces the energy barrier necessary for the water molecules to rotate, and so allows them to reorientate even in weak electromagnetic fields in greater numbers, and to a greater extent than water molecules in bulk water. We have concluded that as the surface adsorbed water molecules re-orientate to align their dipole moments in the electromagnetic field, they exhibit a dielectric constant many orders of magnitude greater, often several thousand times greater, than that of bulk water.

We have further recognised, however, that at the measuring frequency, available and dissolved salts can also have a measurable contribution, leading to false results. Thus, in a preferred embodiment suitable for use not only with washed sand aggregate but also unwashed sand, and also for concrete with ionic species in the capillary pore water from hydration reaction processes, we separate the measured signal into its real and imaginary components, and rely only upon the measured imaginary component. In this manner, we discard the contribution from dissolved salts.

It will be understood that by utilising the imaginary component, the need to use washed sand is avoided, since our results show equivalence between the readings from washed and unwashed sand ready for mixing. The imaginary component will allow the separation of the contribution from surface adsorbed water and pore dissolved charged species from the cement undergoing hydration.

We have also noted that when water is added to a concrete mix, it spreads through the agglomerations of sand and aggregate particles, and into the clusters of cement grains comprising mineral phases such as tricalcium, silicate (alite), dicalcium silicate (belite), tricalcium aluminate (aluminate), tetracalciumaluminoferrite (ferrite) and gypsum. During the f irst ten minutes or so of mixing for a typical OPC (ordinary portland cement) made to British Standard 12, when the cement grains eventually react to form aluminate rich gel by the hydration of tricalcium aluminate and/or ferrite solid solution with calcium sulphate in solution, they adsorb large amounts of water onto surface reactive sites, with large dipole moments as a result of a low degree of inter-molecular hydrogen bonding.

Thus, if a sample containing hydrating cement is subjected to an electromagnetic field generated by an alternating electrical signal, a relatively quantitative determination of the amount of water in this state can be determined by measuring the impedance of the sample, which comprises both real and imaginary components corresponding to the resistance and reactance of the sample. Changes in the amount of water in this state affect the dielectric constant and so are reflected in changes in the reactance and therefore also the impedance.

Thus, according to one feature of our invention we propose a method of monitoring adsorbed water in a material which includes the steps of providing a sample of the material between dielectric probes, and determining the impedance of the sample in the presence of an alternating electric field, the material having a water content below its saturation level.

According to another f eature of our invention we provide 15 electrical circuit apparatus which includes means to determine the imaginary component of the impedance of a sample in the presence of an alternating electric field, and means to connect the apparatus to capacitor probes.

capacitor circuit means.

We recognise that our method and apparatus are most suitable for use up to the water saturation level of the aggregate. Above the water saturation level, the effective dielectric constant becomes that of bulk water.

Bulk water should not, however, be present within properly mixed and setting concrete i.e. after the initial set, and a low dielectric measurement warns of this. However, if a small quantity of bulk water is absorbed after setting e.g.

into the capillary pores, this can be ignored in our dielectric measurement since its value will be about 80 only.

Furthermore, we recognise that dissolved organic materials can provide spurious capacitance results, but such water is already known to be unsuitable for concrete mixing and would have been rejected before the capacitance testing was performed.

7 - Thus, the method and apparatus of the invention are useful for a moisture content up to a known and measurable percentage (typically 5-8%) by dry weight.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be further described, by way of example, with reference to the accompanying drawings, in which:- Fig. 1 is a diagramatic interpretation of a capacitance plot for the hydration of OPC concrete against time; Fig.2 is a diagramatic plot similar to that of Fig.1, but for selected concrete constituents; Fig.3 is a model of the electrical characteristics of a sample; Fig. 4 Fig. 5 Fig. 6 Fig. 7 is a phasor diagram, with indicated real and imaginary components of the impedance of a sample; is a graph of the imaginary component of the measured capacitance against moisture content, for unsaturated sand, washed and as delivered (unwashed); is a graph similar to that of Fig.5, extending to beyond saturation; and is a circuit suitable for a phase sensitive detector.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Fig.1 is a plot of measured capacitance against time for a sample of hydrating concrete made using ordinary portland cement. Fig.2 is similar, being for the principal hydration products, though a calorimetry plot after about four hours setting time would be broadly similar. It is desirable for repeat measurements to be made on hardened concrete as in checking the status of road bridges and the like.

peaks ma Fig.1 and Fig.2 can be extended to show a longer term capacitance plot against time as the concrete is setting to form the principal hydration products expected. However, if there is excess water present in the original mix, then the concrete may not be as durable as expected; there can for instance be alkali silica reactions, alkali carbonate reactions, and delayed ettringite reactions, and these conditions will be manifested by an above- normal reading of capacitance, perhaps for a 30 day or 300 day capacitance reading, either generally or as localised "peaks". These y be observed before deterioration becomes significant, since they are sensitive to the initial stages of deterioration before major damage occurs, to concrete and thus to a structure built therewith. Thus, if probes are set in situ in the structure, measurement apparatus can be connected to the probes whereby the circuitry gives an indicated reading, preferably digital, and if the reading is above the expected value, then close attention can be given to the particular structure to check further whether there has been decay or deterioration of the concrete, before an accident occurs.

There can be minor variations between the results from different samples. Thus, the inventor believes that the quantity of surface adsorbed water is a function both of the total gel pore surface and of the capillary pore surface in the cementitious material, and which changes as the hydration proceeds. He has noted that any water present in the set concrete will typically be water inf lowing along capillaries from the external environment as bulk water e.g. rain water, and as such will not generate such "unexpected" dielectric value dependence, since this water will have a dielectric constant of approximately 80; the small amount of capillary water adsorbed onto the surface of the concrete constituents in which the inter-molecular bonding has broken down will show an increased dielectric constant which may be measurable, but the surface area of the capillaries will be several orders of magnitude smaller than the surface area of the constituents within any reacting gel, so that the effects of the reacting gel will be predominant when this is present in significant quantities.

The inventor has set out to provide electrical circuit apparatus to measure the enhanced dielectric effects of surface moisture in aggregates and in a sample of hydrating concrete. This apparatus when connected to capacitor probes of suitable shape for the application (including a hopper wall as one probe when measuring the surface water percentage in discharging sand) will measure the enhanced dielectric effect of surface adsorbed water e.g. 10,000 as compared to about 80 for bulk water.

When water is added to cement to f orm a paste, a series of complex chemical reactions begin, and which result in the eventual setting and hardening of the cement paste or gel. As indicated by the results shown in graphical form in Fig.1, which are typical for OPC pastes, mortars and concrete, and other hydration products, during the initial setting stage 2, lasting about ten minutes, there is believed to be a rapid formation of aluminate rich gel, with pores having a much smaller size than capillary pores, but of slightly greater magnitude than the water molecules.

Thus, the water in the gel pores is not in a "free water" condition but is adsorbed. Furthermore, the specific surface area of the gel is very great, estimated to be between 200,000 m2/kg and 600,000 m2/kg. The adsorbed water exhibits the enhanced dielectric properties identified by the inventor as useful for the capacitance measurements utilised in this invention, and confirmed by the rapid rise in the graphical representation of the capacitance measurement plot over this initial period of hydration.

Following this initial stage the concrete undergoes an initial set i.e. the formation of gel and a loss of plasticity.

After about 70 minutes, the plot of the capacitance peaks, and thereafter starts to decline progressively. This is believed to correspond with the onset of the concrete final setting, and the formation of capillary pores, which rapidly displace the spaces of the initial gel, and thus of the gel pore surface area.

This belief is consistent with the suggestion that the amount of dielectrically enhanced adsorbed surface water in the gel pores is reduced, in consequence of it being progressively locked into hydration products where its dielectric contribution is reduced. As the gel/pore structure is replaced by hydration products and capillary pores, there is a net loss of internal surface area and surface adsorbed water. This is reflected in the capacitance plot, which shows a steep decline 4 after the peak 6.

At about the 18 hour stage, the capacitance plot exhibits a shoulder broadening 8. It is believed that a secondary gel formation process occurs, whilst there is also an indication at about the three day stage that there is a further peak 10, possibly attributed to further internal hydration reactions. Thus, the capacitance measurement component is very sensitive to the combined surface area of the curing concrete gel and the capillary pores.

11 Cement grains are known to include several mineral phases, such as tricalcium silicate, dicalcium. silicate, tricalcium aluminate and gypsum. In the graph of Fig.2, the capacitance component for adsorbed water (below saturation level) is indicated for separated component reactions in a concrete mix, such as that of Fig.l. In this embodiment, the separated reactions are fa) for the first 10 hours the hydration of tricalcium silicate to form a cement gel shell; {b) after about 18 hours, tricalcium aluminate and/or ferrite solid solution produce long rods of ettringite; and (c) after about 3 days, the reaction of tricalcium. aluminate with ettringite to produce hexagonal plates of monosulphate.

Thus, and in relation to the graphs of Fig.1 and Fig.2, for about the first two hours after mixing the fresh cement paste is a plastic network of particles of cement in water; after this time there is a gradual densification of the set paste by the growth of hydrates of the various mineral phases of the cement clinker. After such densification (setting), the paste develops a randomly orientated, interconnected capillary pore network. As the paste further hardens, this network is in a constant state of change as filamentary pores, which were continuous through the paste sample, become constricted or even blocked by hydration products e.g. primarily calcium silicate hydrates and calcium aluminate hydrates. As access to pores becomes blocked, the continuous path through the hardened cement paste becomes increasingly circuitous.

During hydration, water is held in varying degrees of firmness. At one extreme there is free water held in the capillaries, containing calcium and hydroxyl ions, amongst others; at the other extreme exists chemically combined water forming a definite part of the hydrated products.

Between these two extremes there is surface adsorbed water and interlayer water. Adsorbed water is held onto the gel and other hydration product surfaces (exposed for example in 12 - the capillary cavities) by surface forces, and is believed to be approximately four molecules thick, whereas interlayer water is held between the sheetlike structure of the calcium silicate hydrate, and is believed to be approximately one molecule thick, and as a consequence is tightly bound.

As the surface adsorbs water molecules, it is suggested that there is competition with the intermolecular hydrogen bonding. The water molecules thus become more freely available to contribute to the dielectric measurement, increasing the capacitance value of the sample. However, dissolved ionic species from the hydration reactions alter the resistance of the sample. The capacitance value can be separated from the resistance value by means of a phase sensitive detector.

The model of Fig.3 represents the combined effect of the surface adsorbed water molecules and the ionic species wherein the effect of water molecules is represented by the capacitor 34, and the effect of the ionic species is represented by the resistor 36. If an alternating potential is applied across the terminals 30,32 a phase sensitive detector can separate the capacitance measurement (the imaginary component) from the resistance measurement (the real component), thus separating the effect of the surface adsorbed water molecules from the effect of the ionic species.

In the embodiment for Fig.3, the capacitance is in the range 0 - 500 nanofarads whilst the resistance is in the range 100 to 1000 ohms for a typical sample, using an applied EMF of 10 Volts.

The phasor diagram of Fig. 4 indicates the total impedance Z, and the real r and imaginary j components of the impedance for a sample of aggregate such as sand. In this phasor diagram, "a" is a function of the circuit resistance - 13 and "b" Classical I is G is is R is V is C is Then and is a reciprocal function of circuit capacitance. theory of alternating currents provides that if: the current the ionic conductance or loss component E is the Electromotive force F is the frequency of the applied EMF the circuit resistance the output voltage the capacitance V = I.R I = E R + 1/(G+j.wX) where 1/(G+j.wX) for this measurement circuit is the impedance of the measured concrete, and with j.w.C being the impedance of the capacitance effect of 20 the measured concrete.

N B j = 4r---P; w = 2nF It follows that V = E x E x E x (a + jb) R R + 1/(G+j.w.C) GR(I+R.G) + (R.WX)2 (1 +R.G)2 + (R.WX)2 j R.w.C (1 +R.G)2 + (R.W.C)2 jb is the imaginary component of the impedance Z and is a function of the capacitance of the sample and is dependent on the polarisability of the medium.

14 - Use of a phase sensitive detector gives the quadratic component of V to be:- is V = E x R.w.C (1 +R.G)2 + (R.w.C)2 The quadratic component is proportional to the capacitance C if:- (a} w = 2nF is fixed, as also is E (electromotive force), R (circuit resistance) and F (frequency - though in practice the method can be utilised provided that F is sufficiently constant; a variation of 10 ppm or less is not necessarily essential but is ideal); {b) R.G is less than 1; and fcj (R.w.C) is less than 1.

The inventor has noted that the requirement that R.G be less than unity becomes more difficult to satisfy as G increases, corresponding to higher concentrations of conductive ionic species in the setting concrete. The product R. w. C can be kept less than unity by a suitable choice of the circuit resistance, in accordance with the values of w and C.

From Fig.5, it will be observed that using the imaginary component avoids the need for separate measuring instruments or calibration, as between the results for washed (50) and unwashed (52) sand samples. Specifically, the combined data for sand "as delivereW (shown as circled results 50) and sand "as washeW (shown as crossed results 52) are compared, and indicate that the measurement using the imaginary component gives comparative results, without need for recalibration of the test instrument or apparatus.

From the example resulting in the measurement graph of Fig.6 it is seen that the capacitance of sand (washed or unwashed) measured using the apparatus and method of the invention is linearly dependent upon surface adsorbed moisture up to the saturation point 60, following which it is suggested that hydrogen bonding in the bulk water breaks down the surface dependent dielectric effects. At or above saturation, extensive hydrogen bonding within the bulk water extends to the surface adsorbed species, inhibiting the freedom to reorientate. Thus, the method and apparatus of the invention are useful for a moisture content up to a known and measurable percentage (typically 5-8%) by dry weight.

It is desirable for repeat measurements to be made on hardened concrete as in checking the status of road bridges and the like. Fig. 1 and Fig. 2 can be extended to show a longer term capacitance plot against time as the concrete is setting to form the principal hydration products expected. However, if there is excess water present in the original mix, then the concrete may not be as durable as expected; there can for instance be alkali silica reactions, alkali carbonate reactions, and delayed ettringite reactions, and these conditions will be manifested by an above-normal reading of capacitance, perhaps for a 30 day or 300 day capacitance reading, either generally or as localised "peaks". These peaks may be observed before deterioration becomes significant, since they are sensitive to the initial stages of deterioration before major damage occurs, to concrete and thus to a structure built therewith.

Thus, if probes are set in situ in the structure, measurement apparatus can be connected to the probes whereby the circuitry gives an indicated reading, preferably digital, and if the reading is above the expected value, then close attention can be given to the particular structure to check further whether there has been decay or deterioration of the concrete, before an accident occurs.

16 - There can be minor variations between the results from different samples. Thus, the inventor believes that the quantity of surface adsorbed water is a function both of the total gel pore surface and of the capillary pore surface in the cementitious material, and which changes as the hydration proceeds. He has noted that any water present in the set concrete will typically be water inflowing along capillaries from the external environment as bulk water e.g rain water, and as such will not generate such "unexpected" dielectric value dependence, since this water will have a dielectric constant of approximately 80; the small amount of capillary water adsorbed onto the surface of the concrete constituents in which the inter- molecular bonding has broken down will show an increased dielectric constant which may be measurable, but the surface area of the capillaries will be several orders of magnitude smaller than the surface area of the constituents within any reacting gel, so that the effects of the reacting gel will be predominant when this is present in significant quantities.

Thus, there is provided apparatus and method to measure the enhanced dielectric effects of surface moisture in aggregates and in a sample of hydrating concrete. This apparatus measures the enhanced dielectric effect of surface adsorbed water recognised by the inventor, as a function of its interaction with the imaginary component of the measured capacitance in the presence of an alternating electrical current fed to excite the sample capacitor, i.e. the dielectric or capacitor probes.

A frequency of 1,000 Hz was found to produce useful results, though in alternative embodiments frequencies in the range 100 Hz to 100,000 Hz were used.

The use of the imaginary component discriminates against the contribution from the ionic species produced during the hydration process, since these affect primarily the real component of the applied alternating current, without therefore significant effect upon the imaginary component at these frequencies. The discrimination permits the high dielectric value of surface adsorbed water up to saturation level to be measured by the apparatus, with enhanced contrast against the readings from cement, aggregates, bulk water and cement hydration products, both as a preliminary measurement of aggregates prior to mixing and to follow the progress of the hydration reaction during setting and subsequent hardening of the concrete.

The processing flow plan of Fig.7 indicates a schematic flow measurement diagram. The sample is located between the capacitor probes Cl and C2, in this embodiment capacitor plates.

Preferably, and in order to avoid possible effects at the electrodes (or capacitor probes/plates), such as polarisation or diffusion processes, theinventor proposes the use of electrically insulated electrodes. In the embodiment of Fig.7 the capacitor plates are optionally coated, to inhibit corrosion and/or enhance the insulative value. Such insulated electrodes effectively add two large capacitors in series with the sample capacitance, but as their contribution to the overall capacitance is a function of the inverse of their specific capacitance, it is small, and constant over time, and as between comparative readings.

In a particular utilisation of this principle, a sample of aggregate or concrete can be placed between dielectric plates Cl and C2 in a container, for example a polythene bag in order that the sample should not contaminate the plates and not cause polarisation processes to occur at the plates.

In another utilisation, a (sheathed) probe can be located 35 centrally in a bulk sand dispensing hopper, with the probe and hopper forming the two capacitor plates Cl and C2.

As a subsidiary feature, the inventor has noted that if the real component is also to be measured, then embedded plates or capacitor probes can be used, insulated as above described; if an appropriate frequency is used, in-situ long term monitoring applications such as in disclosing chlorine ingress into the concrete e.g. a concrete bridge deck, are permitted. Thus, electrode polarisation effects are eliminated and the data is more reliable. In one example, a concrete batch was prepared, and introduced into a polythene sleeve, and then into a capacitance measurement cylinder.

The central rod was covered in a polythene sleeve, and lowered into the concrete. The sample capacitor was placed in a room at 200C, and electrical contact made to the two probes before measurements were taken the water adsorbing surface area During hydration, of the solid phase increases greatly, and thus large amounts of otherwise free water become adsorbed on the solid surfaces. It is this change from "free" water with a relative dielectric constant of 80.37 at 200C to surface adsorbed water with a dielectric constant many orders of magnitude greater at this temperature, that is monitored by the inventor's apparatus.

The representative circuit diagram of Fig.7 includes also a sample and hold circuit unit, numbered 70. Voltage meter 71 is connected to unit 70, and also to line 72 of power supply lines 72,73. The crystal controlled (digital) sine wave generator or oscillator 76 is coupled to the voltage regulator 74 (selected for high accuracy), and sends pulses to sample and hold unit 70 in accordance with instructions from crystal (75) controlled oscillator 76, so that the sample and hold circuit unit 70 is synchronised with the alternating applied voltage.

Resistors R1 in this embodiment are of 100 Ohms, whilst 35 resistors R2 are of 150 Ohms. Resistor Rp is of 1000 Ohms, and is formed as a zeroing potentiometer so that its utilised resistance can be changed.

Rm is a resistor of f ixed value, of very low temperature coefficient, and as shown is fitted in series with the sample. By its use, and with the selection of a suitable sine wave generator 76, in use the circuit of Fig.7 has a fixed resistance Rs. Provided the product of Rs and the measured ionic conductance of the sample (G) is less than unity, and provided the product of Rs, capacitance (C), sine wave frequency (F) and 2 x W (6.28) is less than unity, the use of a phase sensitive detector will give a reliable reading for the quadratic component of Vo, and in particular a reading of Vo which is dependent upon the imaginary value of the capacitance of the sample.

Operational amplifier 77 measures the voltage drop across 15 resistor Rm and provides an output to the sample and hold circuit 70 proportional to the voltage drop across Rm. The sample and hold circuit unit 70 is synchronised with the oscillator 76 to take a reading of the voltage drop across Rm at the same point in each cycle. Because the resistor Rm is in series with the sample the voltage drop across Rm is inversely proportional to the voltage drop across the sample, which voltage drop is proportional to the imaginary component of the impedance of the sample. Hence, provided that the aforementioned conditions {a}, {b} and {c} are satisfied, changes in the measured voltage drop across Rm at a consistent point in the cycle are inversely proportional to changes in the imaginary component of the sample impedance.

Amplifier 77 has a gain such that the output voltage swing can be measured by voltage meter 71. In another embodiment, voltage meter 71 is replaced by a data logger, and in other embodiments different output devices, or combinations of output devices, are used.

Example 1.

During the installation of an oil exploration platform in sea water, about 30,000 cubic metres of concrete had to be placed in 2,000 linear metres of wall, with a very low slipform sliding speed of about 1 metre per day; the method and the apparatus of the invention can be used in this situation (and similar situations) to learn how long the mix would remain workable, and to ensure that placement time and slipform sliding speed was related to time of mixing so that no premature setting occured.

Example 2.

As the pace of building erection has increased, failures of floor and wall surfacing materials have become more common, believed to be due to the presence of moisture which the speed of construction has prevented from drying out before the surface finish is applied to the floor slab or walls, i.e. it is not possible to tell by superficial inspection whether such structures are dry, because the movement of moisture from the interior to the surface is usually slower than the potential rate of evaporation from the surface. The apparatus and method of the invention can be used by the overseers in these (and similar) situations to check if bulk water is present, for instance because the operators had incorrectly made an over-wet mix.

Example 3.

Sand is purchased by calculating the net weight of a loaded 30 bulk transporter i.e. after the weight of the transporter has been subtracted. Excess water means both a shortfall in the quantity of sand, and a deterioration in the concrete made therefrom if the water to cement ratio is out of range.

Example 4.

Sand is often stored pending use. Samples from opposite sides of a storage pile can have different water - 21 proportions, perhaps because one side is dried by the prevailing wind or is rain-lashed. Thus, the overseer needs a rapid and accurate measuring facility even for material removed from a single stack.

The invention can also be used to monitor the above processes in cement paste, cement and other slurries, clays, cementitious grouts, mortars, and other non-organic materials. In addition, the invention can be used with certain organic materials, to measure surface-bound moisture, subject to the surface wetability characteristics of the material, and suitable calibration of the circuit.

Claims (12)

1. A method of monitoring adsorbed water in a material which includes the steps of providing a sample of the material between dielectric probes, and determining the impedance of the sample in the presence of an alternating electric field, the material having a water content below its saturation level.
2. 2. A method as claimed in claim 1 in which the material comprises one of an aggregate, curing concrete, cured concrete, mortar or other cementitious material.
3. 3. A method as claimed in claim 1 or claim 2 in which a calculation of the imaginary component of the impedance is made, the imaginary component being dependent upon the combined surface area of the curing concrete gel and the capillary pores.
4. 4.

   A method as claimed in any of claims 1-3 characterised by taking independent impedance measurements over time.
5. Apparatus for performing the method of any of claims 1-4 in which an alternating electric field supply is connected to terminals in a measurement circuit having a phase sensitive detector adapted to detect the impedance of the sample at a consistent point of cycles of the alternating electric field.
6. 6. Apparatus according to claim 5 in which the phase sensitive detector includes a sample and hold circuit connected to a synchronising output of the oscillator providing the alternating electric field.
7. 7.

   Apparatus according to claim 5 or claim 6 in which the product of the circuit resistance and the ionic conductance of the sample is less than one, and in which the product of the circuit resistance the frequency of the alternating electrical f ield the capacitance of the sample and two times pi is less than one.

   powering meanE
8. 8. Apparatus according to any one of claims 5-7 in which is provided to apply an alternating voltage of 10 volts, and circuit means is provided including a phase sensitive detector responsive to a capacitance in the range 0500 nanofarads and a resistance in the range 100- 1000 ohms.
9. 9. Apparatus as claimed in any one of claims 5-8 in which the quadrature component of the impedance is fed from a measurement unit to output devices, in which the output devices comprise a data log and microprocessor analyser connected to the output from a frequency sweep indicator which in turn receives a signal from an a.c. sine wave generator whereby to indicate to the output devices the measurement frequency.
10. 10. Electrical circuit apparatus for determining the amount of water in a concrete mix by a method according to any of claims 1-4 in which electrical capacitor circuit means is connectable to capacitor probes and measurement means responsive to the electrical circuit to provide an indication of the impedance of the sample.
11. A concrete structure characterised by probes set in situ for use by the apparatus of any one of claims 5-9.
12. 12. An apparatus for performing the method of monitoring water in a nonorganic material, constructed and arranged substantially as described in relation to Fig.7 of the accompanying drawings.