WO1996001999A1

WO1996001999A1 - Rapid prediction and real time determination of total organic carbon in water supplies

Info

Publication number: WO1996001999A1
Application number: PCT/US1995/009463
Authority: WO
Inventors: Adam D. Folley
Original assignee: Millipore Corporation
Priority date: 1994-07-12
Filing date: 1995-07-10
Publication date: 1996-01-25

Abstract

Methods are presented for obtaining the amount of total organic carbon in a water supply. In one method the amount of total organic carbon is calculated from a proportionality relationship between resistivity measurements and a previously determined assessment of resistivity for a water sample where all the organic carbon is completely oxidized. In another method the amount of total organic carbon is calculated from two resistivity measurements that are obtained within a fraction of a minute from each other. One of the measurements is taken of the background water resistivity and the other measurement is taken after a sample of the water supply has been subjected to a short period of carbon oxidation. The presented methods allow rapid assessments of total organic carbon in water supplies to be made and determinations of total organic carbon in water supplies when the resistivity of the water supply is changing.

Description

RAPID PREDICTION AND REAL TIME DETERMINATION OF TOTAL ORGANIC CARBON IN WATER SUPPLIES

Background of the Invention

It is often desired in industrial applications for water, used in these applications, to be free of organic carbon. For example, the manufacture of computer chips or pharmaceuticals requires the use of water, for cleaning or solution purposes, of the highest purity obtainable. The presence of organic carbon in the water will introduce unwanted carbon-containing substances into the products. The organic carbon compounds are quite often electrically neutral. Such compounds are not easily completely removed from water supplies and, in fact, may be introduced into the water from the organic based ion-exchange resins of water purification systems.

A convenient means to remove organic carbon from the water of water supplies is to oxidize the organic carboncontaining molecules to break the carbon-carbon covalent bonds and form small carbon-oxygen molecules which are more readily extracted from the water by, for example, ion-exchange beds. The oxidation of the organic carbon- containing molecules can be accomplished by a variety of processes including the addition of strong oxidants, such as peroxides, into the water or by irradiation with radiation of sufficient energy and intensity. For

example, far ultraviolet (UV) light of sufficiently short wavelength can effectively be used to initiate oxidation reactions with organic carbon compounds. The subsequently formed oxidized carbon molecules are then extracted from the water supply.

Since water which is free of organic contaminants is often desired, the water supply is periodically monitored to assess the water quality and to assure that the amount of organic carbon in the water supply is below a tolerable threshold. Periodic monitoring is required for a variety of reasons. Organic substances may be introduced into the water supply during the course of water being taken from the supply, e.g. from water purification ion-exchange resin beds. The processes and materials used to remove organic carbon contaminants from the water supply may not be functioning properly due, for example, to aging or inability to properly remove the organic carbon load in the water supply. It is often necessary to have an accurate record of the organic carbon content in the water supply as the water is being used.

Monitoring of water supplies, particularly flowing water supplies, has been, heretofore, a time consuming process. For this reason, valuable time is lost from productive use of the water in order to evaluate the organic carbon content. The water may be inefficiently diverted from a productive use to an evaluative use. The monitoring or evaluating process is time consuming because multiple measurements must be made in order to obtain a reasonably accurate estimate of the amount of organic carbon, or the evaluated water sample must be subjected to conditions which oxidize the organic carbon until this carbon is completely oxidized before the monitoring measurement is made. As these monitoring activities are being performed, the organic carbon content of the water supply may be undergoing a change, i.e. the organic carbon content of the water supply may not be at a steady state.

For these reasons, a need exists for a rapid method of ascertaining the content of organic carbon in water supplies. This need exists both for static and flowing water supplies as well as for water supplies where the organic carbon content is in an unsteady state. Summary of the Invention

The present invention pertains to methods for

determining the amount of organic carbon that is present in a water supply at any time. In one of the methods the following steps are performed. First, the resistivity of an initial sample of water from the water supply is estimated as if all the organic carbon in the sample were completely oxidized. The amount of completely oxidized organic carbon will reduce the resistivity of the water sample in proportion to the quantity of organic carbon present in the sample. Second, a sample of water from the same water supply is subjected to conditions which oxidize organic carbon for a specified period of time and the resistivity is measured. Third, another sample of water from the same water supply, at a different time, is subjected to the same conditions which oxidize organic carbon for the same specified period of time that occurred for the initial water sample. The resistivity for this other sample is measured. Fourth, the resistivity of this second sample of water is calculated as if all the organic carbon in the sample were completely oxidized. This calculation is made from the proportionality condition that the ratio of the second sample resistivity when all organic carbon is completely oxidized to the resistivity that occurs when this second sample is 'subjected to conditions which oxidize organic carbon is the same as the ratio of the initial sample resistivity when all organic carbon in it is completely oxidized to the resistivity that occurs when the initial sample is subjected to 4;he conditions which oxidize organic carbon. Since the only unknown value in this proportionality relationship is the resistivity of the second sample when all the organic carbon is completely oxidized, this value is readily calculated from the proportionality relationship and the other three known quantities. Finally, the amount of organic carbon in this second sample of water is

determined from known equivalences between the amount of organic carbon ascertained to be present in given water samples and the resistivities of these respective given water samples when all organic carbon in these samples has been completely oxidized. These known equivalences allow the amount of organic carbon in this second sample to be determined when resistivity values are known for this sample and all organic carbon is completely oxidized. The amount of organic carbon in this other sample is

indicative of the amount of organic carbon present in the water supply at the time the second sample was taken from the water supply.

Another method of the present invention pertains to the determination of the amount of organic carbon that is present in a flowing water supply at any time. This method is particularly suitable for water supply systems where the water flows from a water source to a water delivery point. In this method, the following steps are performed. First, the resistivity of an initial sample of water from the flowing water supply is estimated as if all the organic carbon in the sample were completely oxidized. Again, the amount of completely oxidized organic carbon in the sample will reduce the resistivity of the water sample in proportion to the quantity of organic carbon present in the sample. Second, as the water is flowing past a position where a source is located that constantly

oxidizes organic carbon in the water, a sample of this water is subjected to conditions which further oxidise organic carbon beyond that obtained during the normal water flow past the source that constantly oxidizes organic carbon. The conditions which further oxidize organic carbon include subjection of the water sample to the oxidation conditions for a longer than normal time period and the imposition of an additional or a more intense oxidation source. These conditions are applied for a specified period of time and the resistivity is measured. Third, again as water is flowing past the position where a source is located that constantly

oxidizes organic carbon in the water, a second sample of water from the same water supply, at a different time, is subjected to the same conditions which further oxidize organic carbon for the same specified period of time as occurred for the first water sample. The resistivity for this second sample is again measured after the procedure which further oxidizes the organic carbon in the water sample is performed. Fourth, the resistivity of this second sample of water is calculated as if all the organic carbon in this sample were completely oxidized. This calculation is made from the proportionality condition that the ratio of the second sample resistivity when all organic carbon is completely oxidized to the resistivity that occurs when this second sample is subjected to the conditions which further oxidize organic carbon is the same as the ratio of the initial sample resistivity when all organic carbon in it is completely oxidized to the resistivity that occurs when the initial sample is

subjected to the conditions which further oxidize organic carbon. Again, since the only unknown value in this proportionality relationship is the resistivity of the second sample when all the organic carbon is completely oxidized, this value is readily calculated from the proportionality relationship and the other three known quantities. Finally, the amount of organic carbon in this second sample of water is determined from known

equivalences between the amount of organic carbon

ascertained to be present in given water samples and the resistivities of these respective given water samples when all organic carbon in these samples has been completely oxidized. Again, these known equivalences allow the amount of organic carbon in this second sample to be determined when resistivity values are known for this sample and all organic carbon is completely oxidized. The amount of organic carbon in this second sample is

indicative of the amount of organic carbon present in the flowing water supply at the time the second sample was taken from the flowing water supply.

It should be apparent that the principal difference between the method for determining the amount of organic carbon that is present in a flowing water supply (the latter method) and the amount of organic carbon that is present in any water supply (the first described method) is the use of conditions which further oxidize organic carbon on samples from the flowing water supply. This difference is employed when the flowing water is routinely subjected to a source of conditions which oxidize organic carbon. Otherwise, the two methods are essentially the same.

The present invention also pertains to methods for determining the amount of organic carbon that is present in a water supply by obtaining two measurements of

resistivity of the water in the water supply and then by inserting the results of these measurements into a curve fitting algorithm which calculates a minimum resistivity for the water supply as if all the organic carbon in the water supply were completely oxidized. One of the

resistivity measurements is a baseline or background resistivity of the water supply which is taken just prior to the procedure that culminates in the second resistivity measurement. The second resistivity measurement is taken from a sample of the water supply that has been subjected to conditions which oxidize organic carbon present in the water. The water sample is taken just after the first resistivity measurement. The period of time for

performing the carbon oxidation can be 20 seconds or less. Thus, the two resistivity measurements can be accomplished in approximately a 20 second time frame. The amount of organic carbon in the water supply is determined from known equivalences between the amount of organic carbon ascertained to be present in given water samples and the resistivities of these respective water samples when all organic carbon in these samples has been completely oxidized. The calculated minimum resistivity obtained from the curve fitting algorithm is the resistivity value used in the equivalences procedure.

Brief Description of the Drawings

Fig. 1 is a graphical representation of the estimated minimum resistivity of water samples, when the organic carbon in the samples is completely oxidized, compared to the measured residence time resistivities for these water samples.

Fig. 2 is a graphical representation of the estimated minimum resistivity of water samples, when the organic carbon in the samples is completely oxidized, compared to the measured residence time resistivities for these water samples.

Fig. 3 is a graphical representation of the estimated minimum resistivity of water samples, when the organic carbon in the samples is completely oxidized, compared to the measured residence time resistivities for these water samples.

Fig. 4 is a graph which occurred during the

acquisition of resistivity data for both 'rapid' and

'rigorous⁷ estimations of total organic carbon content in a flowing water supply. The x-axis represents time in minutes and the y-axis represents the resistivity at a position immediately following a UV light source. Each division on the x-axis represents 2 minutes and each major division on the y-axis represents 1 MΩ-cm. Detailed Description of the Invention

This invention is directed to methods for determining the amount of organic carbon that is present in a water supply at any time. The water supply can be a closed system, an open system where water continually or

intermittently enters the supply from an external source, or, preferentially, a flowing supply where the water is fed through a conduit system to a water tap. In this latter system, the water is preferentially fed through a water purification system in order to extract

particulates, mineral ions and other contaminants before delivery of the water to the water tap. Often, the water purification system includes chambers containing organic based ion-exchange resins.

The term "organic carbon" is meant, in the usual sense, to imply organic carbon-containing compounds where the carbon is covalently bonded to other atoms. Often, these compounds have multiple carbon atoms which form myriad organic structures.

The organic carbon that is present in the water supplies is oxidized to small carbon-oxygen molecules that are subsequently removed within the water purification system. The oxidation of the organic carbon can be accomplished by a variety of techniques. A preferred means to irradiate the organic carbon-containing molecules in the water is with far-ultraviolet light of shorter than 210 nm wavelength through a fused quartz window capable of transmitting far-ultraviolet light in the water containing vessel. The water is transparent to this irradiation but the organic-carbon containing molecules absorb it and become oxidized by subsequently reacting with the oxygen contained as a gas in the water solution. This latter process is the organic carbon oxidizing means employed in the exemplifications of the present invention, but other techniques are available and known to persons skilled in the art.

The present invention provides a rapid, simple and convenient method for determining the amount of organic carbon in a water supply at any time desired by the user. The water supply can be stagnant or it can be flowing, for example, from a head, such as a distillation apparatus, to a tap. The organic carbon quantity determination can be made as the water is undergoing a purification procedure. Since the present method for determining the amount of organic carbon in a water supply includes the use of conditions which oxidize organic carbon, that converts organic carbon to simple molecules with increased

conductivity, further purification of the water of organic carbon is easily performed and readily adapted to the present method of determination. For example, an

additional ion-exchange resin bed or an additional ionexchange resin bed additionally containing activated carbon can be inserted into the water supply system following the organic carbon determination position. This additional ion-exchange resin bed or additional ionexchange resin bed additionally containing activated carbon would scavenge the oxidized carbon molecules thereby removing these molecules from the water supply. If the organic carbon oxidation procedure of the present method is allowed to operate until all the organic carbon is oxidized, the additional ion-exchange resin bed or additional ion-exchange resin bed additionally containing activated carbon, sometimes referred to as a "polisher", would remove all the previously present organic carbon from the water supply. Thus, organic carbon determination and subsequent removal can readily and advantageously be coupled when the present method of organic carbon

determination is being employed. The method of the present invention is accomplished by performing the following steps.

First, the resistivity of an initial sample of water from the water supply is estimated as if all the organic carbon in the sample had been completely oxidized. This resistivity can arbitrarily be given the symbol R_{Tot, 1}. Resistivity is conventionally measured in units of ohm-cm or megohm-cm. The actual measurement on the initial sample can be performed by a variety of known techniques. For example, the electrical resistance can be measured across the water sample contained in a cell after the organic carbon in the sample has been completely oxidized. The water sample in the cell is subjected to oxidation conditions sufficient to completely oxidize all the organic carbon in the sample and the resistance across the cell is measured and converted to resistivity values.

Alternatively, and preferably, a series of water samples can be subjected to oxidation conditions for different specified periods of time, resistivity measurements taken for these samples, and the resulting values used to predict the resistivity of a water sample if it were subjected to oxidation conditions for an infinite period of time, i.e. after complete oxidation of the organic carbon. Such a procedure is shown in U.S. Patent No.

5,272,091 which is herein incorporated by reference. An implicit assumption in the series of samples alternative of determining the resistivity after complete oxidation of the organic carbon is that the baseline resistivity does not vary between the samples. Otherwise, the

predictiveness of the series of resistivity measurements becomes much more difficult to achieve. Thus, the

practice of the series of samples alternative is

practically limited by this assumption.

The resistivity of water at 25° C is about 18.2 megohm-cm when the water is free of ionic or other contaminants that affect its resistivity. As organic carbon is oxidized, the oxidized carbon molecules are conductive so the resistivity of the water sample

containing the oxidized carbon molecules is lowered from the 18.2 megohm-cm value. The more organic carbon that is present and oxidized in the water sample, the lower the resistivity of the sample.

Any dissolved ionic species, including inorganic cations or anions, will contribute to the lowering of the resistivity of a water sample containing the ions.

Usually, the water supply whose organic carbon content is to be determined has been preliminarily purified of particulate and ionic species contaminants, e.g. by passing the water through ion-exchange resins, activated carbon, etc. However, such preliminary purification is not required for the method of the present invention to successfully operate. The presence of inorganic ionic species is tolerated in the present invention provided that the concentration of such ionic species remains constant as the resistivity measurements of the second and third steps are performed. Of course, such ionic species as well as organic carbon should be eventually removed from the water when pure water is desired.

The second step of the present method is to measure the resistivity, in a sample of water from the same source as used in the first step, that occurs when the sample is subjected to a set of conditions which oxidize organic carbon for a known period of time. This resistivity difference can arbitrarily be given the symbol R₁. The set of conditions which oxidize organic carbon is

sufficient to at least partially oxidize the organic carbon that resides in the water sample. It is not required that all organic carbon be completely oxidized, but at least a portion of it must be oxidized to form molecules that are conductive. By doing so, a change in resistivity is realized. The result of this oxidation procedure is that the resistivity after the sample is subjected to the oxidation conditions is lower than the resistivity before the oxidation conditions are applied. The size of the resistivity difference is dependent on the amount of organic carbon present in the water sample, the strength or intensity of the oxidation conditions and the length of time that these oxidation conditions are

applied.

The third step of the present method is to measure the resistivity in a water sample taken at the time when the organic carbon content of the water supply is sought to be determined. This resistivity can arbitrarily be given the symbol R₂. This resistivity is that which occurs when the water sample is subjected to the same set of conditions which oxidize organic carbon and for the same period of time as for the second step. If the amount of organic carbon that is present in the water supply has changed between the resistivity measurements of the second step and the third step, then these resistivities will not be identical. Again, as for the second step, it is not required that all organic carbon be completely oxidized in this step.

The fourth step of the present method is to determine the resistivity of the water sample taken in the third step that would occur if all the organic carbon in that sample were completely oxidized. This resistivity can arbitrarily be given the symbol R_{Tot 2}, This is the resistivity for the water supply when the organic carbon content is sought to be determined. This resistivity determination is made by realizing that the ratio of resistivity of the water sample of the third step to the resistivity of the water sample measured in the third step is equivalent to the ratio of the resistivity of the water sample determined in the first step to the resistivity of the water sample measured in the second step. Simply stated, R_Tot,2 = R₂ x R_{Tot, 1}/R₁. Another way of viewing this determination is by noting that the resistivity of the water sample whose organic carbon content is sought to be determined is related to the resistivity of the initial water sample through the ratio of the measured resistivity for the water sample whose organic carbon content is sought to be determined to the measured resistivity for the initial water sample.

In the practice of the present invention, the

sequence of the first three steps is unimportant. The measurement of the resistivity for the initial water sample can be taken after the measurement of the

resistivity for the water sample whose organic carbon content is sought to be determined. The resistivity of the initial sample of water can be determined after either of the two resistivity measurements have been taken.

Usually, however, the sequence of the first three steps of the method of the present invention will be in the order as described in the preceding paragraphs.

Another aspect of the present invention is that either or both of the resistivity measurements (for the initial water sample or for the water sample whose organic carbon content is sought to be determined) can be taken when the baseline resistivity of the water supply from which the water samples are drawn is changing. For example, the number or type of ionic species in the water supply may be changing as the water samples are withdrawn for resistivity measurements. If such baseline

resistivity changes are occurring, it will not affect the accuracy of the determination of the amount of organic carbon that is present in the water supply at the time such content is desired to be known, i.e. when the sample is taken for the third step described above. This ability to use water samples when the baseline resistivity is varying is an advantage provided by the method of the present invention. Such an ability is not present when the methods of the prior art, e.g. U.S. Patent No.

5,272,091, are used. The method of the present invention can be used when the baseline resistivity is constant or varying.

The final step of the present method is to use the resistivity for the water supply when the organic carbon content is sought to be determined (i.e., the value obtained in the fourth step) to determine the amount of organic carbon in that water sample and thereby in the water supply. This determination is made by equating resistivity values for water samples when the organic carbon is completely oxidized to organic carbon amounts. Such equivalences can be in tabular form or in the form of an analytical relationship (a graph or an equation of organic carbon amount as a function of resistivity when the organic carbon is completely oxidized). These

equivalences can be obtained from standard sources or from calibration measurements made earlier or subsequently at the water supply site of interest. The determination of the amount of organic carbon from the resistivity value when the organic carbon is completely oxidized completes the method of the present invention.

It is often desired to remove organic carbon from the water supply before or as the water is being delivered for its desired use. This removal can be accomplished by a variety of procedures known to the skilled worker. For example, the organic carbon can be oxidized to ionic or conductive carbon-containing molecules which can

subsequently be removed by passing the water through ion- exchange resin beds. The oxidation of the organic carbon compounds can be achieved by a variety of procedures known to the skilled artisan. The irradiation is constantly provided to ensure that the organic carbon entering the water supply is oxidized. In other words, the oxidation conditions are continuously applied. A readily available procedure is to irradiate the water supply with

ultraviolet (UV) light of sufficient energy and intensity, e.g. a high intensity mercury or xenon lamp whose

luminance is embedded in the water supply or directed through a fused quartz window to the water supply. This procedure can easily be adapted to a flowing water supply, i.e. where the water flows from a source through a pipe or conduit system, preferably including water purification sites (e.g. tanks of appropriate resins or adsorbents), to a water tap for delivery of pure water. The UV oxidation source can easily be mounted on one of the pipes of the water purification system.

When oxidation conditions of organic carbon are continuously applied to a water supply that is flowing past the oxidation source, e.g. a UV lamp, there is often a difference in resistivity of the water that occurs as the water flows past the oxidation source. That is, if resistivity measurements of a given water sample were taken immediately preceding and immediately following the oxidation source, a resistivity difference for this sample would be apparent. Even under such circumstances, the methods of the present invention can be utilized. In these situations, the measured resistivities in the present invention are designated as resistivity

differences and arbitrarily given the symbols ΔR₁ and ΔR₂ for the measured resistivity of the first sample of water and the measured resistivity of the second sample of water, respectively.

Since conditions where organic carbon is oxidized are, in many instances, constantly applied to the water supply in procedures to remove the organic carbon from the water supply, the method of the present invention can be modified, in these circumstances, such that both the oxidation of organic carbon and a determination of organic carbon in the water supply are simultaneously achievable. This simultaneous achievability is accomplished by

modifying the second and third steps of the above

described method. In these modified steps, the oxidation conditions are increased, i.e. performed at higher

intensity or on a static sample in a normally flowing water supply, for a known period of time. This period of time is the same in both the second and third steps. The increase in intensity can be performed in a variety of ways. For example, a second oxidation source can be directed at the water sample in addition to the

continuously applied oxidation source. Alternatively, the continuously applied oxidation source can be increased in intensity for the known period of time. In any event, the added oxidation conditions will cause more organic carbon to be oxidized to ionic or conductive species. This will, in turn, cause a change in resistivity in the water sample from the normal operating condition that is measurable for each water sample. The resistivities that result from these changes are assigned the symbols R₁ and R₂ as described above for the situation when continuous

oxidation of organic carbon is not occurring. Other than the increased oxidation conditions, the method of the present invention is the same whether continuously applied oxidation conditions are used or are absent.

The known period of time to which the sample of water is subjected to a set of oxidized conditions, or increased oxidation conditions, can be of any duration. Preferably, the period of time is 20 seconds or less. The oxidation period should be of sufficient length to generate a measurable amount of conductive molecules from the

oxidized organic carbon but short enough to be of

practical use as an assessment that yields the organic carbon content of the sample of water from the water supply whose content of organic carbon is sought.

In practical situations, particularly where the water samples are obtained from a flowing water supply, it is often advantageous, in the economy of time, to obtain the first sample of water soon after water flow is initiated and to determine the resistivity (R_Tot,1) of tne initial sample of water that would occur if all the organic carbon in this sample had been completely oxidized. In the present invention, the obtainment of this first sample of water can be performed within 10 minutes after water flow has been initiated. In most instances, the determination of the R_Tot,1 resistivity of this sample of water must await stabilization of the baseline resistivity before the procedures which result in this determination can be performed due to the underlying basis of these procedures. However, the resistivity measurement (R₁) for this first sample can be obtained at any time after or as the sample is obtained, even if the baseline resistivity of the water supply is changing. Thus, R_Tot,1 determination sampling of the first sample of water should normally be done when the baseline resistivity of the water is unchanging but measurement of the R₁ resistivity does not depend on such an invariance. For either set of measurements, the sample of water can usually be obtained within 10 minutes of water flow initiation. By that time, the baseline

resistivity of the flowing water has stabilized to a constant value.

Another aspect of the present invention is the ability to calculate a minimum resistivity of a sample of water from a water supply by obtaining two resistivity measurements and inserting these resistivity measurement values into a curve fitting algorithm. The first

resistivity measurement is a baseline or background or "residence time" resistivity of the water supply. Such a measurement can be taken when the baseline resistivity is in a steady state or is undergoing a resistivity change, i.e., an unsteady state. This first resistivity

measurement is usually obtained just prior to an oxidation procedure which oxidizes organic carbon in a sample of water from the water supply for a predetermined period of time. This period of time can be any reasonable time but oxidation times of 20 seconds or less are preferred.

After the organic carbon in the sample of water has been oxidized for the predetermined period of time, the second resistivity measurement is obtained. These two resistivity measurements can thus be obtained within approximately 20 seconds or less of each other. These resistivity

measurement values can be inserted into a curve fitting algorithm, e.g., in a standard computer program, for the calculation of a minimum resistivity of the water sample as if all the organic carbon in the water sample had been completely oxidized. Such algorithms can be generated by the user or obtained as a curve fitting package from appropriate vendors. The minimum resistivity calculated from the curve fitting algorithm can be related to the total organic carbon in the water supply by applying the previously discussed known equivalences between

resistivity values and organic carbon amounts.

The advantages of the just described methods are the quick (20 seconds or less) calculation of total organic carbon in a water supply (including a flowing water supply) and the ability to calculate such a total organic carbon amount when the resistivity of the water supply is in an unsteady state.

When a flowing water supply is used, it is often advantageous, although not necessary, to insert a 3-way valve upstream of a cell in the water line where the UV irradiation and the resistivity measurements are

performed. Water can enter the cell and the water supply then diverted by use of the 3-way valve. The water in the cell is then subjected to the conditions which oxidize organic carbon for the known period of time and

resistivity measurements are taken before and after the oxidation procedure. The 3-way valve can again be

actuated thereby flushing the water from the cell and resuming the normal water flow.

Such a 3-way valve arrangement is also useful when the cell is the position of the continuously applied oxidation conditions. Under normal flow operation, the water supply is constantly subjected to oxidation of organic carbon as the water flows through the cell. When increased oxidation conditions are desired, the 3-way valve is actuated to divert the water flow, thereby allowing the water in the cell to remain for the known period of time. The water in the cell is subjected to the same or increased oxidation conditions for this period of time. This subjection causes an increase in oxidation of organic carbon in the water since the water cannot escape from the cell during the irradiation process as it does under normal flow operation. Resistivity measurements are taken before the water diversion and after the known period of time. The 3-way valve is actuated again to resume normal water flow and continuous application of oxidation conditions.

This 3-way valve arrangement is particularly useful when connected via a pump and conduit system to another 3- way valve downstream of the cell in the water line where the UV irradiation and the resistivity measurements are performed. The water flow system established with this coupled set of 3-way valves allows water to be recycled through the oxidation (UV irradiation) path as the organic carbon in the water is oxidized. When the recycling waterway encompasses the water flowthrough compartment, the UV irradiation path and, optionally, the water purification system, it becomes possible to monitor the resistivity of the water and thereby determine the amount of organic carbon present as well as the amount of

oxidation required to completely oxidize the organic carbon. The recycling operation allows the user to perform a series of resistivity difference measurements as the water is successively subjected to oxidization (UV irradiation) conditions. This series of successive resistivity difference measurements can be continued until the change between two immediately succeeding resistivity difference measurements is about zero, i.e. when the organic carbon in the recycling water sample is

essentially completely oxidized. Such a determination provides an indication of the amount of oxidation needed to purify the water supply of organic carbon.

Example 1

Resistivity Ratio Technology and Total Organic Carbon

Predictions

The following procedure is an example of the method used to predict the total organic carbon discharged from a water purification system according to the method of invention. The water source was reverse osmosis water made from Bedford, Massachusetts tap water. A Millipore Milli Q UV Plus water purification, commercially available from Millipore Corporation, Inc., was modified to include the proper apparatus to measure total organic carbon content.

A steady state ("residence time") resistivity value for the water supply, which was measured at a position immediately following a UV light source ("post UV

resistivity"), was observed for a period of about ten minutes. This value was 11.57 megohm-cm at 25 degrees centigrade. During this time the water purification system was operating in a recirculation mode. This is defined as a state of operation where water continuously recirculates through the system; that is, the same water stays in the system and recirculates through it in a closed loop.

In an initial determination of the total organic carbon content of this water sample, the following

resistivities were observed for water samples from the recirculating water supply that were held in the UV light path for the designated time periods.

A nonlinear regression analysis of these data

according to the equation:

where R(0) is the resistivity at time t, = 0, i.e. the baseline resistivity (T is a time constant), yielded the resulting equation:

Resistivity(t) = 8.18 + 3.79-exp (-0.048 ·time).

The calculated resistivity value at infinite time of oxidation was 8.18 megohm-cm at 25 °C. The calculated resistivity value at infinite oxidation time is R_min. The ratio of infinite time resistivity to residence time resistivity was calculated to be 8.18/11.57 or 0.71. This ratio can also be expressed as R_Tot,1/R₁ = 0.71 since, here, R_min = R_Tot,1.

The total organic carbon (TOC) of the water sample can be calculated from the resistivity value at infinite time (R_Tot,1) by the formula:

TOC = -2.3896 + 45.049/R_Tot,1 + 133.78/R² _Tot,1 +

35.202/R³ _Tot,1

Substitution of 8.18 MΩ-cm for R_Tot,1 yields a value of 5.18 ppb total organic carbon for this particular water sample.

This water purification system was changed from a recirculation mode to a production mode. This latter mode allows the water purification system to take in reverse osmosis water, purify it, and then send it to a container or an analytical instrument.

The water purification system was operated for about 10 minutes during which time the steady state ("residence time") resistivity value for the water supply measured at the position following the UV light source was 3.74 MΩ-cm.

In a determination of total organic carbon content for this water system, the following resistivities were measured:

A non regression analysis of these data yielded the resulting equation: Resistivity(t) = 2.51 + 1.15-exp (-0.036 ·time)

The calculated resistivity value at infinite time of oxidation was 2.51 megohm-cm at 25 °C. The ratio of

infinite time resistivity to residence time resistivity was calculated to be 2.51/3.74 or 0.69. This ratio can also be expressed as R_Tot,2/R₂ = 0.69. Note the change of subscripts from '1' to '2' for this second water sample since the water supply was changed.

The total organic carbon of the second water sample can be calculated from the resistivity value at infinite time (R_Tot,2) by the empirically derived formula:

Substitution of 2.51 MΩ-cm for R_Tot,2 yields a value of 38.94 ppb TOC for this particular water sample.

A summary of the two different water samples and their respective resistivities is shown below:

These two different ratios are 2.9% different from one another, i.e. they are essentially identical.

The total organic carbon content can be calculated at a particular resistivity value with the aid of the

following procedure and formula (s):

A. Measure the resistivity, R, at the position following the UV light source for a specified period of time (usually the residence time value) for a particular water sample.

B. Calculate the total resistivity for this water sample: R_Total = R · ratio of estimated resistivity at infinite oxidation time for initial sample (R_min) to resistivity measurement of initial sample for same specified period of time (R_θ). This algebraic relationship can be formally derived in the following manner:

The equation R (t) = R_min + [R₀ -R_min ] e^-t/τ can be arranged as R(t) = R_min[1 - e^-t/τ]+R_θ e^-t/τ o_r _

The limit

which is the ratio of the estimated

and measured resistivities for the initial ssaammpplle. A second limit with new conditions would yield

or

This can be rearranged as:

C. Calculate total organic carbon content by

_otal

The following is a compilation of total organic carbon content values for the two different water samples for different baseline (residence) resistivities which were determined by using the above method. The values were calculated by using the two different ratios stated above. It is of interest to compare the total organic carbon values at the same resistivity.

The unique advantage to this procedure is the ability to quickly estimate these total organic carbon (TOC) values in real time. The determination of the resistivity ratio can be done with four data points in about 3.5 minutes or can be done with two data points in 20 seconds (see next example). Once the resistivity ratio is known, then TOC values can be estimated once the residence time resistivity at the position following the UV light source is known. This is constantly being measured in real time also since the water system has water flowing constantly across the UV light. A commercially available TOC measuring device, not employing the concepts of this disclosure, would take about 8-15 minutes to measure each TOC sample. Thus, it would take approximately 96-195 minutes to measure all of the above TOC values that could be measured in real time from either an initial 3.5 minute or 20 second set of

resistivity measurements.

Example 2

Comparisons of Minimum Resistivity Values [R_min] to

Residence Time (Baseline) Resistivity Values [R(θ) 1

A series of procedures were performed using the protocol and data analysis described in Example 1. Steady state, i.e. baseline or "residence time", resistivities of the water samples were measured. These measured

resistivity values were graphically compared with minimum resistivity values that were calculated in nonlinear regression analyses from the equation: R(t) =R_min+[R(θ)-[R-_{mi n}].e^-t/r

These calculated minimum resistivity values represent the estimated resistivity of the water samples when the organic carbon in these respective samples is completely oxidized.

A. The graph of Fig. 1 shows the relationship of the

calculated minimum resistivity value to the measured residence time resistivity for two different

estimations of total organic carbon in a flowing water supply. The resistivity values used in the nonlinear regression analysis were taken after 20, 30, 40 and 50 second oxidation times. There was one recirculation mode estimation and one recirculation mode estimation.

The recirculation mode is when the water is internally recirculating its own water throughout its components. The external feedwater does not enter nor exit the water system.

The production mode is when the water

purification system is taking in new feedwater and purifying it. This purified water is then discharged from the system to a user (glass washing, solution making). The graph of Fig. 2 shows the relationship of the calculated minimum resistivity value to the measured residence time resistivity for several estimations of total organic carbon in a flowing water supply. The resistivity values used in the nonlinear regression analyses were taken at 20, 30, 40 and 50 second oxidation times. There were six production mode estimations and one recirculation mode estimation.

The recirculation mode is when the water system is internally recirculating its own water throughout its components. The external feedwater does not enter nor exit the water system.

The production mode is when 'the water

purification is taking in new feedwater and purifying it. This purified water is then discharged from the system to a user (glass washing, solution making) .

The feedwater in this particular experiment began as service deionization water made from Bedford, MA tap water. This water was then further treated by an ultrafiltration membrane device before entering the water system. A tank was used to store this water before it entered the water purification system. The resistivity ratio from the recirculation mode estimation was 5.00/6.95 or 0.72.

The resistivity ratios from the production mode estimations were 2.73/3.20, 2.88/3.62, 2.92/3.90, 3.03/4.07, 3.11/4.17 and 3.02/4.28. These ratios, when divided out, became 0.85, 0.80, 0.75, 0.74, 0.75 and 0.71. The graph of Fig. 3 shows the relationship of the calculated minimum resistivity value to the measured residence time resistivity for two different

estimations of total organic carbon in a flowing water supply. The resistivity values used in the nonlinear regression analysis were taken after 20, 30, 40 and 50 second oxidation times. There was one production mode estimation and one recirculation mode estimation.

The production mode is when the water

purification is taking in new feedwater and purifying it. This purified water is then discharged from the system to a user (glass washing, solution making).

The feedwater in this particular experiment began as reverse osmosis (RO) water made from Bedford, MA tap water. This RO water was then further treated by deionization resin and carbon, commonly known as service deionization or SDI, before entering the water system. A tank was used to store this water before it entered the water purification system.

The resistivity ratio from the production mode estimation was 1.67/2.34 or 0.71.

The resistivity ratio from the recirculation mode estimation was 8.09/11.85 or 0.68. The results of these comparisons of estimated minimum resistivity values to measured residence time or baseline resistivity values reveals that a proportionality

relationship exists between these values. This is

demonstrated by the linear relationship that occurs when the estimated minimum resistivity values are graphically plotted against their counterpart measured residence time resistivity values. Such graphical representations are shown in Figures 1-3. This proportionality relationship signifies that, for a given water supply system, the ratio of estimated minimum resistivity to measured residence time resistivity is constant, i.e.

where K is a specific dimensionless quantity for a given water system (the slope of the linear relationship). Example 3

Twenty Second Determination of Total Organic Carbon Content

The water purification system and apparatus of

Examples 1 and 2 were used including the UV light source for carbon oxidation and the resistivity measurement position following the UV light source. The graph of Fig. 4 shows the acquisition of resistivity data during both "rigorous" and "rapid" estimations of total organic carbon content of the flowing water supply.

A residence time resistivity drop occurs from 18 MΩ-cm to 11.5 MΩ-cm at minute 2. During minutes 2-4 the

establishment of steady state resistivity is seen (the per trace is "flat"). A "rigorous" estimation is done during minutes 4 through 8 (HKIN 11). The lowest four valleys or dips come from successive 20, 30, 40 and 50 second oxidation times. The resistivity valleys become lower as the oxidation times are increased. During minutes 8 through -9.5 the resistivity is steady state. After minute 9.5 the water purification system was switched from a recirculation mode to a production mode. This means that the deionized water was no longer being recirculated through the UV light irradiation and ion-exchange resin but was instead now being dumped to a drain or sink. The new feedwater entering the water system has more total organic carbon than the recirculated water and thus causes a larger resistivity drop. The resistivity drop is defined by the vertical distance between the 18 MΩ-cm line and the trace (solid black) line.

Minutes 9.5 through 10.5 shows the "post UV"

resistivity is in an unsteady state. This is indicated by the steep change in the slope. A "rigorous" estimation could not be performed here because of the changing

baseline resistivity. However a 20 second estimation is easily done. This is indicated by the RAPID EST marker at minute 11. A second RAPID EST was done at minute 12.5 and is labeled as such.

A second and third set of "rigorous" estimations were done at minutes 17 - 21.5 and minutes 24 - 27.5. These are labeled as HKIN12 and HKIN13 respectively. These two estimations were done at somewhat steady state conditions of water supply resistivity. HKIN13 is definitely done at more steady state conditions than HKIN12.

The calculation of the total organic carbon content of the water supply when data were acquired for a "rigorous" estimation was performed by the nonlinear regression analysis procedure used in Examples 1 and 2.

For the "rigorous" estimation labeled HKIN11 in Fig. 4, the resistivity values for the water supply which were measured at a position immediately following a UV light source (post UV resistivities) were:

A nonlinear curve fit of these data to the equation R(t) = R_min + [R(θ) - R_min] · e^-t/τ yielded R_min = 6.95 MΩ-cm and r = 35.92 sec. Evaluating the total organic carbon content by the equation: TOC = -2.3896 + 45.049/R^ + 133.78/R² _min + 35.202/R³ _mjn, a TOC of 6.98 ppb and a ratio of R_min/R(θ) which is 6.95/11.70 = 0.59 were obtained.

From these same resistivity measurements, an

estimation of the minimum resistivity was performed using only the resistivity measurements at 0 sec. and 20 sec. ( a "rapid" estimation). The resistivity values of 11.70 MΩ-cm and 9.07 MΩ-cm were curve fitted to the same equation as used for the "rigorous" estimation.

The equation R(t) = R_min + [R(θ) - R_min ] · e^-t/τ, at first inspection, appears to not be properly curve fitted with the two data points of resistivity at 0 and 20 seconds. However, there is an implicit relationship between T and R_min,R(θ). It has been observed that r is short when the difference between R(θ) and R_min is large. It has also been observed that τ is long when the difference between R(θ) and R_min is small. It has been found that it is not necessary to know the exact relationship between T, R(θ) and R_min in order to perform the curve fitting procedure that yields R_min . The curve fit that was performed with the data in this example was done with an iterative computer algorithm.

This algorithm performs a least squares fit of the data to the model equation: R(t) = R_min + [R(θ)-Rmin] · e^-t/τ. The least squares fit is accomplished by minimizing the sum of the squares of the differences between the observed data t, R(θ) and R(t) and the values calculated by the model equation for progressive estimations of τ and R_{min .} This algorithm was performed on an EXCEL version 4.0 worksheet.

A nonlinear curve fit for the 11.70 and 9.07 MΩ-cm data points to the equation

R(t) = R_min + [R(θ) - R_min] · e-^t/τ

yielded R_min = 6.68 MΩ-cm and T = 26.94 sec. Evaluating

TOC = -2.3896 + 45.049/R_min + 133.78/R² _min + 35.202/R³ _min, yielded TOC = 7.47 ppb and a ratio of R_min/R(0) which is 6.68/11.70 = 0.57.

As previously shown, the recirculation mode estimation (HKIN11 on the graph) indicated a TOC value of 6.98 ppb entering the UV light. The "rapid" estimations performed in the above manner at 11 and 13 minutes on the graph yielded values of 13.48 and 25.32 ppb. The "rigorous" estimations done in production mode (HKIN12 and HKIN13 on the graph) gave TOC values of 37.84 and 45.36 ppb

respectively. It should be noted that the lower

resistivity values, which are graphically illustrated in

Fig. 4, are indicative of the higher TOC values obtained by the "rigorous" and "rapid" estimations.

Tables 5-10 are summaries and comparisons of six separate experiments where the amount of total organic carbon (TOC) in water supplies was estimated by the

"rigorous" and the "rapid" estimation methods just

described. The actual resistivity measurements as well as the determined R_min τ, R(θ) and TOC values are presented. The resistivity units are MΩ-cm, the τ unit is seconds and the TOC unit is ppb. For comparison purposes, a separate determination was made of the total organic carbon in the water supplies using an Anatel TOC instrument. (See, e.g., U.S. Patent No. 5,275,957 issued January 4, 1994). For the "rigorous" estimation procedures (RIGEST), the resistivity values at 20, 30, 40 and 50 second oxidation times were used. For the "rapid" estimation procedures (RAPEST), the resistivity values at 0 and 20 second oxidation times were used.

Table 11 presents comparisons of the total organic carbon (TOC) results for the six experiments. These comparisons show that the RIGEST estimation, the RAPEST estimation and the Anatel determination are in close agreement. It should be noted that the "rigorous" estimation takes approximately 3.5 minutes, the "rapid" estimation takes approximately 20 seconds (since the baseline resistivity can be constantly monitored) and the Anatel determination takes at least between 8 and 15 minutes.

The primary advantage of the 20 second estimation procedure is that it is very quick compared to the 3.5 minute estimation procedure. The presented six sets of experimental results show that the 20 second estimation procedure gives results that are as accurate as the 3.5 minute estimation procedure.

A second and equally important advantage is that the 20 second estimation procedure can be done at unsteady state conditions as illustrated in Fig. 4. Steady state conditions exist when the "post UV resistivity" is

relatively constant over time (usually 2-5 minutes).

Unsteady state conditions occur when the resistivity is changing over time. Since unsteady state condition exist when flowing water supplies are initially started, it is necessary to first discard 2-5 minutes worth of water to the drain before beginning a 3.5 minute estimation in order to reach the steady state conditions required for the 3.5 minute estimation. There are other instances where

achievement of steady state is not possible. An example is when the feedwater is coming from ion-exchange tanks, more commonly known as Service Deionization (SDI). If the "post UV residence time resistivity" is changing with time, a cumulative error will result if "rigorous" estimations of total organic carbon are performed. The advantage of the 20 second estimation is that it uses the residence time "post UV resistivity" value and only one measured

resistivity point. The curve fitting of these two points gives the sought relationship of residence time [R(θ)] and minimum time resistivity [R_min], i.e. the ratio R_min/R(θ). These values can, in turn, be used during normal operating conditions to determine the total organic carbon in the system. These normal operating conditions are at a time when a TOC estimation procedure is not normally being done.

While this invention has been particularly shown and described with references to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the

invention as defined by the appended claims.

Claims

CLAIMS What is claimed is:

1. A method for determining the amount of organic carbon at any time in a water supply which comprises:

a) determining the resistivity (R_Tot,₁) of an

initial sample of water, from said water supply, that would occur if all the organic carbon in said initial sample had been completely oxidized; b) measuring the resistivity (R₁) that occurs in

said initial sample of water when said initial sample of water is subjected to conditions where organic carbon is oxidized for a specified period of time;

c) measuring the resistivity (R₂) that occurs in a second sample of water, from said water supply, when said second sample of water is subjected to said conditions where organic carbon is oxidized for the same specified period of time as in step b);

d) determining the resistivity (R_Tot,2) of said

second sample of water that would occur when the organic carbon in said second sample is completely oxidized, said determining of R_Tot,2 being accomplished from the proportionality condition that the ratio of R_Tot,2 to R₂ is the same as the ratio of R_Tot,1 to R₁; and

e) determining the amount of organic carbon in said second sample of water from known equivalences between the amount of organic carbon known to be present in a given water sample and the resistivity in said given water sample when all the organic carbon in said given water sample is completely oxidized.

2. The method of Claim 1 wherein the water supply is flowing through a water purification system and said measuring steps are performed between the water purification procedure location and the water delivery for use location.

3. The method of Claim 2 wherein an ion scavenger is

present between the location of the procedure where organic carbon is oxidized and the water delivery for use location, said ion scavenger to absorb the ionic carbon produced by said procedure where organic carbon is oxidized.

4. The method of Claim 3 wherein said ion scavenger is an ion-exchange resin bed.

5. The method of Claim 3 wherein said ion scavenger is an ion-exchange resin bed containing activated carbon.

6. The method of Claim 3 wherein the delivered water for use has a resistivity of about 18.2 megohm-cm at 25°C.

7. The method of Claim 3 wherein said conditions where organic carbon is oxidized comprise the illumination of a water sample compartment with ultraviolet light.

8. The method of Claim 2 wherein said measuring of the resistivities comprises measurement, before and after said conditions where organic carbon is oxidized are applied, by a single resistivity measuring device located on the water flow efflux side of the procedure where organic carbon is oxidized.

9. The method of Claim 7 wherein said specified period of time is not more than 20 seconds.

10. The method of Claim 2 wherein step a) and step b) are performed within 10 minutes after water flow

initiation and steps c) to e) are performed anytime thereafter.

11. The method of Claim 2 wherein R_{Tot, 1} is determined by: i) measuring resistivities in separate samples of said flowing water where each sample is subjected to conditions where organic carbon is oxidized for different specified periods of time, wherein each measured resistivity is matched with its respective time period of carbon oxidation;

ii) inserting said resistivities and their matched said respective time periods into a curve fitting algorithm that thereby calculates the resistivity in a hypothetical water sample that

hypothetically was subjected to said conditions where organic carbon is oxidized for an infinite period of time; and

iii) relating said resistivity for said hypothetical water sample to R_{Tot, 1} wherein said relating is accomplished through known equivalences between known amounts of organic carbon in water samples, resistivities for such water samples when all the organic carbon in such water samples is completely oxidized, and R_{Tot, 1}.

12. The method of Claim 11 wherein said measured

resistivities for determining R_{Tot, 1} are measured when the reference resistivity for all said separate samples is the same value prior to subjection of said separate samples to said conditions where organic carbon is oxidized and wherein one of said measured resistivities is taken without subjecting the respective sample to said conditions where organic carbon is oxidized.

13. A method for determining the amount of organic carbon at any time in a flowing water supply which comprises: a) determining the resistivity (R_Tot,1) of an

initial sample of water, from said flowing water supply, that would occur if all the organic carbon in said initial sample had been completely oxidized;

b) while the water is flowing past a constant source of organic carbon oxidation:

i) measuring the resistivity difference (ΔR₁) that occurs in said initial sample of water when said initial sample of water is

subjected to an increased amount of organic carbon oxidation for a specified period of time; and

ii) measuring the resistivity difference (ΔR₂) that occurs in a second sample of water, from said water supply, when said second sample of water is subjected to said

increased amount of organic carbon oxidation for said specified period of time;

c) determining the resistivity (R_Tot,2) of said

second sample of water that would occur when the organic carbon in said second sample is completely oxidized, said determining of R_Tot,2 being accomplished from the proportionality condition that the ratio of R_Tot,2 to ΔR₂ is the same as the ratio of R_Tot,1 to ΔR₁; and d) determining the amount of organic carbon in said second sample of water from known equivalences between the amount of organic carbon known to be present in a given water sample and the resistivity in said given water sample when all the organic carbon in said given water sample is completely oxidized.

14. The method of Claim 13 wherein the flowing water

supply is flowing through a water purification system and said measuring steps are performed between the water purification procedure location and the water delivery for use location.

15. The method of Claim 14 wherein an ion scavenger is

present between the location of the source and

procedure where organic carbon is oxidized and the location of the water delivery for use, said ion scavenger to absorb the ionic carbon produced by said procedure where organic carbon is oxidized.

16. The method of Claim 15 wherein said ion scavenger is an ion-exchange resin bed.

17. The method of Claim 15 wherein said ion scavenger is an ion-exchange resin bed containing activated carbon.

18. The method of Claim 15 wherein the delivered water for use has a resistivity of about 18.2 megohm-cm at 25°C.

19. The method of Claim 15 wherein said constant source of organic carbon oxidation comprises illumination of a water flowthrough compartment with ultraviolet light.

20. The method of Claim 19 wherein said increased amount of organic carbon oxidation comprises said

illumination of said water flowthrough compartment for said specified period of time while water flow therein is halted.

21. The method of Claim 19 wherein the water is recycled through said water flowthrough compartment, and wherein said step b) i) of measuring the resistivity difference (ΔR₁) is performed as a series of

operations until the change in resistivity difference between two successive instances of said operations is about zero.

22. The method of Claim 14 wherein said measuring of the resistivity differences comprises measurement, before and after said subjection to an increased amount of organic carbon oxidation, by a single resistivity measuring device located on the water flow efflux side of said water flowthrough compartment.

23. The method of Claim 20 wherein said specified period of time is not more than 20 seconds.

24. The method of Claim 14 wherein step a) and step b)i) are performed within 10 minutes after water flow initiation and steps b)ii), c) and d) are performed anytime thereafter.

25. The method of Claim 14 wherein R_{Tot, 1} is determined

by:

a₁) measuring resistivity differences in separate

samples of water, from said flowing water supply, where each sample is subjected to said increased amount of organic carbon oxidation for different specified periods of time, wherein each measured resistivity difference is matched with its respective time period of increased amount of organic carbon oxidation; a₂) inserting said resistivity differences and their matched said respective time periods into a curve fitting algorithm that thereby calculates the resistivity difference in a hypothetical sample of water that hypothetically was subjected to said increased amount of organic carbon oxidation for an infinite period of time; and a₃) relating said resistivity difference for said

hypothetical sample of water to R_{Tot, 1} wherein said relating is accomplished through known equivalences between known amounts of organic carbon in samples of water, resistivity differences for such samples of water when all the organic carbon in such water samples is completely oxidized, and R_{Tot, 1·}

26. The method of Claim 25 wherein said increased amount of organic carbon oxidation for different specified periods of time comprises illumination of a water flowthrough compartment with ultraviolet light while water flow therein is halted for said different specified periods of time.

27. The method of Claim 25 wherein said measured

resistivity differences for determining R_{Tot, 1} are measured when the reference resistivity for all said separate samples is the same value prior to subjection of said separate samples to said increased amount of organic carbon oxidation and wherein one of said measured resistivity differences is taken without subjecting the respective sample to said increased amount of organic carbon oxidation.

28. A method for determining the amount of organic carbon at any time in a water supply which comprises:

a) measuring the resistivity of the water supply at a specified time;

b) subjecting a sample of said water supply to

conditions where organic carbon is oxidized for a predetermined period of time beginning at said specified time;

c) measuring the resistivity of said sample of said water supply after said predetermined period of time;

d) determining the minimum resistivity (R_min ) of

said sample of said water supply that would occur when the organic carbon in said sample is

completely oxidized, said determining of R_min being accomplished by inserting the measured resistivity values of steps a) and c) into a curve fitting algorithm; and

e) determining the amount of organic carbon in said sample of said water supply from known equivalences between the amount of organic carbon known to be present in a given water sample and the resistivity in said given water sample when all the organic carbon in said given water sample is completely oxidized.

29. The method of Claim 28 wherein the water supply is

flowing through a water purification system and said measuring steps are performed between the water purification procedure location and the water delivery for use location.

30. The method of Claim 29 wherein an ion scavenger is present between the location of the procedure where organic carbon is oxidized and the water delivery for use location, said ion scavenger to absorb the ionic carbon produced by said procedure where organic carbon is oxidized.

31. The method of Claim 30 wherein said ion scavenger is an ion-exchange resin bed containing activated carbon.

32. The method of Claim 30 wherein the delivered water for use has a resistivity of about 18.2 megohm-cm at 25°C.

33. The method of Claim 30 wherein said conditions where organic carbon is oxidized comprise the illumination of a water sample compartment with ultraviolet light.

34. The method of Claim 29 wherein said measuring of the resistivities comprises measurement, before and after said conditions where organic carbon is oxidized are applied, by a single resistivity measuring device located on the water flow efflux side of the procedure where organic carbon is oxidized.

35. The method of Claim 33 wherein said predetermined

period of time is not more than 20 seconds.

36. A method for determining the amount of organic carbon at any time in a flowing water supply which comprises: a) measuring the resistivity of the flowing water supply at a specified time;

b) while the water is flowing past a constant source of organic carbon oxidation, measuring the resistivity difference (ΔR₁) that occurs in a sample of said flowing water supply when said sample is subjected to an increased amount of organic carbon oxidation for a predetermined period of time beginning at said specified time; c) determining the minimum resistivity (R_min ) of

said sample that would occur when the organic carbon in said sample is completely oxidized, said determining of R_min being accomplished by inserting the measured resistivity values of steps a) and b) into a curve fitting algorithm; and

d) determining the amount of organic carbon in said sample of water from known equivalences between the amount of organic carbon known to be present in a given water sample and the resistivity in said given water sample when all the organic carbon in said given water sample is completely oxidized.

37. The method of Claim 36 wherein the flowing water

38. The method of Claim 37 wherein an ion scavenger is

present between the location of the source and

39. The method of Claim 38 wherein said ion scavenger is an ion-exchange resin bed containing activated carbon.

40. The method of Claim 38 wherein the delivered water for use has a resistivity of about 18.2 megohm-cm at 25°C.

41. The method of Claim 38 wherein said constant source of organic carbon oxidation comprises illumination of a water flowthrough compartment with ultraviolet light.

42. The method of Claim 41 wherein said increased amount of organic carbon oxidation comprises said

43. The method of Claim 41 wherein the water is recycled through said water flowthrough compartment, and wherein said step b) of measuring the resistivity difference (ΔR₁) is performed as a series of

44. The method of Claim 37 wherein said measuring of the resistivity difference comprises measurement, before and after said subjection to an increased amount of organic carbon oxidation, by a single resistivity measuring device located on the water flow efflux side of said water flowthrough compartment.

45. The method of Claim 42 wherein said predetermined

period of time is not more than 20 seconds.