WO1997047975A1

WO1997047975A1 - Analysis apparatus and analysis method

Info

Publication number: WO1997047975A1
Application number: PCT/DK1997/000248
Authority: WO
Inventors: Hans Svankjaer Jacobsen
Original assignee: Danfoss A/S
Priority date: 1996-06-07
Filing date: 1997-06-04
Publication date: 1997-12-18
Also published as: AU3165497A; DE19622847A1

Abstract

An analysis apparatus and an analysis method for the continuous determination of a species in a fluid is disclosed. For that purpose at least one reagent (R1, R2) reacting with the species is added to the fluid (T) in order to produce a reaction product dependent on the concentration of the species. The reaction product is evaluated in a detector device (9, 10). It is desirable to be able to carry out such continuous analyses quickly and reliably. To that end, at a first time after addition of the reagent (R1, R2) the reaction product is determined. A further determination is effected later at a different time. From the difference in the signals obtained with the findings, conclusions are drawn about the concentration of the species in the fluid.

Description

Analysis apparatus and analysis method.

The invention relates to an analysis apparatus for the continuous or semi-continuous determination of a species in a fluid, the analysis apparatus having a sample-removing device, a reaction channel, which is connected to the sample removal device and to at least one reagent supply device, a detector arrangement and a control device. The invention furthermore relates to an analysis method for determining the concentration of a species in a fluid, in which method at least one reagent reacting with the species is added to the fluid in order to produce a reaction product dependent on the concentration of the species, and the reaction product is evaluated in a detector device through which the fluid is channelled.

The invention can be applied, for example, when the concentration of a species in a fluid is to be monitored continuously or se i-continuously, for example, the phosphate content or nitrate content in the water of a sewage treatment plant. The fluid charged with the species, which fluid can be produced, for example, by means of a type of dialysis, is mixed with the reagent, which forms a reaction product with the species. If required, additional reagents can be added. The fluid can also be removed directly. The reaction product that can ultimately be evaluated is, for example, a colourant, which changes the light- transmitting capacity of the fluid. Accordingly, the light-transmitting capacity of the fluid charged with the reaction product can be ascertained in the detector arrangement. Other possibilities for the reaction product and the detection thereof are also conceivable, of course.

A widely used procedure consists in measuring intermittently an exact volume of a representative sample of the fluid and adding to that volume an equally exactly measured volume of at least one reagent. If A is the content of the species in the sample volume and B is the reagent content in the reagent volume that reacts with A,

A + B → P, is obtained, P being the reaction product. P is here formed in an amount that corresponds to the original content of A, because B is present in excess. The reaction does not take place immediately, however; it takes a certain time for the reaction to finish. Consequently, on the whole a waiting time is necessary before the measurement can be taken.

Alternatively, the reaction can be evaluated even before it has concluded. It is a precondition here that the reaction in the case of all samples that contain the same concentration of A takes place at the same speed, and that the detector arrangement has previously been calibrated using samples of which the concentration of the species A is known. In order to be able to evaluate a sample, several calibration measurements may possibly be required.

Nevertheless, measuring is beset by the problem that a sample often contains other substances C in addition to the species A. These other substances C can influence the detector arrangement as though they were the reaction product P. The detector arrangement thus gives a false result. In order to solve that problem, previously the content of C in the sample was measured first, that is, without the reaction product P having previously been produced by admixture of the reagent B. On subsequent measurements, the determined value P then had to be corrected by the contribution of C. With continuous measurements, however, the content of C can vary over time. Calibration therefore has to be carried out at shorter or longer intervals.

US 4 120 657 describes how the speed of a chemical reaction is measured. A special programme for controlling the admixture of reagents is used in this instance. Initially, no reagent is admixed. Admixing is then increased step by step to a maximum and then decreased step by step to zero again. In this manner it is possible to ascertain how much time is needed for the reaction product to reach certain values, that is, the amplitudes of a signal obtained from the reaction product are determined, that is, a signal corresponding to the amount of reaction product formed.

Measuring the time between predetermined amplitudes presupposes, however, that these amplitudes can even be reached. Measurements of very low concentrations can therefore in some cases lead to very long measuring times. It may also be that only the first amplitude value, but not the second amplitude value, is reached. This method can, on the whole, only be used if at least one measurable concentration is known in advance, or if the time factor is unimportant. The measurements cannot be performed continuously, but only in batches, which can lead to measuring errors if the concentration of species in the fluid changes at relatively short intervals.

US 4 224 033 describes a programmable analysis apparatus in which a sample amount taken discretely - A - from a supply is mixed with a specific volume of at least one reagent. Depending on the desired programme, this mixture has to pass along a relatively short or relatively long reaction path before the mixture is directed to a detector arrangement.

The invention is based on the problem of allowing continuous analyses to be carried out quickly and reliably.

That problem is solved in the case of an apparatus of the kind mentioned in the introduction in that by means of the detector arrangement the control device determines a signal dependent on a reaction product in a mixture of fluid and reagent at a first time and at a later time, and by means of the difference in the signals determines the concentration of the species in the fluid.

The concentration of the species in the fluid is therefore derived from a difference between two signals. A zero displacement, which would arise because of a foreign substance C, for example, is absent. The zero displacement therefore comes to have no influence on the measurement result. Zero calibrations of the apparatus can be dispensed with. In practice is has been shown that there is a traceable correlation between the measured difference and the concentration of the species in the fluid. Since the measurement is carried out at different times on the same volume of fluid-reagent mixture, the difference can be used to determine the increase in the amount of reaction product. This increase is significant for the concentration in a certain respect. If a lot of species is present in the fluid, the increase in reaction product is effected more rapidly than when only a small amount of species is present in the fluid. It is no longer important that certain amplitudes are reached. Measuring works even with small concentrations.

The detector arrangement preferably comprises several detectors arranged one behind the other on a reaction channel. The time that elapses between individual measurements of the same sample volume can therefore be defined by the time that this sample volume requires to pass from one detector to another. When a measurement is carried out at the later detector, a measurement for a following sample value can be taken at the same time at the preceding detector. In this way a semi- continuous measurement of the species in the fluid can be achieved. It is merely necessary to store the measured value of a preceding detector until the following detector is able to deliver its measurement result. Monitoring the flow speed is also a useful step.

Alternatively or additionally thereto, flow control means can be provided; these are connected to the control device and stop the flow of the mixture in the reaction channel between the first and the later times. This can be exploited for two procedures. Firstly, a single detector can then be used to examine the same sample volume at different times. In that case, differences between individual detectors are unimportant. A zero displacement of one detector is filtered out during formation of the difference. In another procedure the reaction times can be extended. In particular with very small concentrations of the species in the fluid, convincing measurement results can therefore be obtained. The flow control means preferably comprise valves. The valves, especially magnetic valves, not only interrupt the flow. They also isolate individual sample volumes from other fluid portions, so that mutual influence is prevented.

Additionally or as an alternative thereto, the flow control means can comprise icropumps. These control not only an interruption of the flow through the device, but also enable the flow speeds to be controlled and so allow a certain influence to be exerted on the mixing of fluid and reagent.

Preferably, separate micropumps which are synchronizable with each other are provided for both fluid and reagent. Thus, on the one hand, the flow of the reagent fluid can be exactly matched to the flow of the fluid, so that the desired ratio of volumes can be adhered to accurately. Furthermore, the mixture ratio of fluid and reagent can be influenced. Finally, by stopping the micropumps synchronously, it is possible to achieve an interruption in the flow, which can in turn be used to measure the same sample, that is, the same fluid volume to be examined, at different times.

The micropumps preferably close their outlets during a suction stroke and their inlets during a pressure stroke. In this way the flow is stationary during suction of the pumps, and the flow of fluid and/or reagent takes place only during the pressure stroke of the pumps. A pulsed advance of the flow through the apparatus is therefore achieved semi-automatically.

The duration of a suction stroke need not necessarily be the same as that of a pressure stroke. Preferably, the duration of the suction stroke is shorter than the duration of the pressure stroke. In this way short interruptions in the flow are effected, so that the individual measurements can proceed relatively quickly one after the other.

The detector arrangement preferably has at least one photometer, in particular a spectrophotometer. The measurements of the species, that is, of the reaction product, can therefore be performed without physical contact with the mixture of fluid and reagent. A spectrophotometer moreover has the advantage that not only can the turbidity that is caused by the reaction product be measured, but an effect of the reaction product that is different in different regions of the spectrum can be determined. Additional information can therefore be obtained about the species contained in the fluid. If required, even several different species can be evaluated.

Alternatively or additionally thereto, the detector arrangement can comprise at least one conductivity sensor and/or at least one ion-selective electrode and/or at least one pH sensor. The choice of individual detectors depends on the reaction product produced. If the reaction product causes no change in the turbidity, it can cause, for example, a change in the electric conductivity or a change in the ion concentration or a change in pH.

The problem is also solved by a method of the kind mentioned in the introduction, namely, in that at a first time after addition of the reagent a signal dependent on the reaction product is determined and at a further, later, time a signal dependent on the reaction product is again determined and the difference in the signals is formed in order to determine the concentration of the species in the fluid.

After addition of the reagent to the fluid, a reaction starts between the species and the reagent, which ultimately leads to the reaction product. If a predetermined time later a first measurement is carried out, at that time a signal can be obtained which corresponds to the progress of the reaction at that time. Advantageously, a time will be chosen at which the reaction has not yet progressed very far, that is, not very much reaction product has yet formed. At a later time, at which the reaction advantageously ought still not to have finished, a further signal can be obtained. Disturbing influences common to both signals, for example a zero displacement or the influence of foreign substances, that in the detector lead to the same result as the reaction product, are consequently eliminated. This is especially advantageous when the content of these foreign substances changes relatively quickly, which may be the case, for example, in the waste water of sewage treatment plants. Here, for example, it is desirable to measure the nitrate content, but exclude the content of organic nitrate compounds from having any influence on the measurement.

In the period between the first time and the further time, further signals dependent on the reaction product at the particular measuring times are preferably determined. The course of the reaction can be monitored more efficiently in this manner. More information is therefore available from which conclusions about the species can be drawn. Such a procedure may admittedly possibly yield redundant information, but this can be evaluated in order to carry out a reliability or plausibility check.

It is especially preferred in this connection for a curve characteristic to be determined from the signals. Such a curve characteristic is in the simplest case the attempt to describe the determined signals by means of a function. Other, numeric, methods for forming such a curve characteristic can be put forward of course. The curve characteristic can also be formed, for example, in that the detected signals are subjected to signal- processing, for example, are filtered. The wealth of information obtained through the many signals is thus reduced again to one or a few succinct details.

In this connection it is especially preferred for the curve characteristic to be compared with a curve characteristic stored in a memory. Apart from the absolute amplitude values and the differences between the amplitudes of the signals, the signal course or the curve characteristic also enables significant information to be obtained about the content or the concentration of the species in the fluid. In many cases, however, the concentration can be determined only with considerable computing or processing effort from this curve characteristic. If this curve characteristic is now compared with known curve characteristics, this effort is not needed. On the contrary, a check is only made to see where the curve characteristic determined "fits".

In an especially preferred practical form, inferences about faults in the analysis apparatus can be drawn from specific differences between the determined and the stored curve characteristics. It has been ascertained that in the undisturbed state the curve characteristics have certain similarities. They differ in certain parameters which thus enable information to be obtained about the concentration or the content of the species in the fluid. If there are other differences, they may be attributable to faults in the analysis apparatus. For example, the efficiency of a pump may decline, the channels may become blocked, the supply of a reagent fluid can run out or a detector can become clogged. In many of these cases such a fault will show up in the curve characteristic so that it can be taken into consideration in the evaluation.

In this connection it is especially preferred for fault-elimination routines to be initiated when predetermined faults occur. For example, if the efficiency of a pump is declining, the drive capacity is increased. If channels are becoming constricted, descaling or cleaning can be carried out. The system thus not only determines its own faults, but also eliminates them, and at a stage that is early enough for measurements not to be adversely influenced.

The invention is described in the following with reference to preferred exemplary embodiments in conjunction with the drawings, in which: Fig. 1 shows a first embodiment of the invention, Fig. 2 shows a second embodiment, and Fig. 3 shows a third embodiment with synchronously activatable pumps.

The invention is explained in the following with the help of analysis apparatuses by means of which the concentration of a species in a liquid is to be determined. A comparable analysis could, of course, also be carried out for gaseous fluids. Fig. 1 shows a simple embodiment of an analysis device 1 having a tank 2, which holds a carrier liquid T. This carrier liquid T can be constituted, for example, by distilled water. The device 1 also comprises two tanks 3, 4 for reagents Rl, R2 and a further tank 5 holding the liquid that is to be analyzed.

The carrier liquid T flows out of the tank 2 through a serpentine element 6, which passes through the tank 5, to a first mixing point 7. As it flows through the serpentine element 6, the carrier liquid T absorbs the species from the liquid in the container 5, for example, by dialysis or a different concentration equalization method. Other potential methods of removing samples can also be used, of course, for example, a coil could be used in place of the serpentine element, or the liquid could be removed directly from the tank.

At the mixing point 7 the reagent Rl is added at a time TR,. At a later, following mixing point 8 at a further time TR₂ the second reagent is added. The first reagent Rl produces, for example with the species, a first reaction product, which causes a colour change with the reagent R2. That colour is then the reaction product by means of which the concentration of the species is determined.

The carrier liquid T enriched with the species and which is mixed with the two reagents Rl, R2 , then flows onward through a reaction channel K through a detector arrangement which is formed by a first detector 9 and a second detector 10. After flowing through the second detector 10 the liquid flows onward to a collecting tank 11. During the time that the combined liquid flow comprising charged carrier liquid T and the two reagents Rl, R2 requires to pass from the first detector 9 to the second detector 10, the reaction that leads to the reaction product to be detected continues to develop. Whereas in the first detector 9 only the reaction product P that has formed in the section of the reaction channel K between the second mixing point 8 and the first detector 9 can be detected, in the second detector 10 generally a larger amount of reaction product P can be detected. Both detectors 9, 10 produce an output signal which is proportional to the amount of the reaction product contained in a specific volume.

Furthermore, a control device 12 is provided, which is connected to the two detectors 9, 10. The control device ensures that the individual signals that are emitted by the detectors 9, 10 are associated with each other correctly as regards time. This can be called "controlling" for short. The correct association as regards time means that a signal that is associated with a specific section or liquid volume that is passing through the first detector 9 is stored until the same liquid volume has reached the second detector 10. The control device 12 can then form the difference between the two signals.

The time delay arises from the length of the reaction channel K between the first and the second detector 9, 10 and the flow speed. If desired a speed-measuring device can be provided, in a manner not illustrated, for the flow.

Determination of the reaction product at two different times is in most cases sufficiently accurate, especially when the change in a possible content of the tank 5 proceeds slowly in relation to the reaction of the analysis apparatus 1. The reaction or response time is only as great as the flow time from the tank 5 to the second detector 10. When the control device 12 controls the two detectors 9, 10 so that at the same time as the second detector 10 is making a measurement the first detector 9 is measuring a subsequent volume portion of the mixed liquid, a semi-continuous determination, or even, in fact, a continuous determination of the species in the tank 5 can be performed.

The embodiment of Fig. 1 is very simple, because neither pumps nor valves are present. The flow through the reaction channel K is here exclusively effected by gravity, that is, it depends on the difference in level between the tanks 2, 3, 4 and the collecting tank 11. The cross-section of the reaction channel K can here be made very small so that the amount flowing through can be restricted to the order of magnitude of a millilitre per hour or less. Even with small tank capacities, operation can consequently be continued for a long time without the need for the carrier liquid T or reagents Rl, R2 to be replenished. The collecting tank too is able to hold the required amount of liquid.

Fig. 2 shows an alternative construction of the invention, in which identical parts have been provided with identical reference numbers.

There have been two changes here. Firstly, only one detector 13 is provided, which is controlled by a control device 14; that is, its measured values are delivered to the control device 14, which detects these measured values at quite specific points in time. Moreover, a series of valves 15-19 has been added, which are activated by the control device 14. A valve 15 is arranged between the tank 2 and the serpentine element 6. A respective valve 16, 17 is arranged between the tanks 3, 4 and the mixing points 7, 8. Two further valves 18, 19 are located on the two sides of the detector 13 in the flow channel K. The valves 15- 19 can be generally well-known magnetic valves. Alternatively, semi-conductor valves, for example, silicon semiconductor valves, can be used.

Operation of this analysis apparatus can be outlined as follows: the mixture of charged carrier liquid T and the two reagents Rl, R2 is introduced into the detector 13. The valves 18, 19 are then closed. At pre¬ determined times the control device 14 samples the measurement results of the detector 13.

As specified above, two different times with two different measurement results are sufficient to be able to calculate back to the concentration of the species in the tank 5. Once the valves 18, 19 are opened, another liquid volume can be fed into the detector.

By means of the valves 15, 16, 17, the inflow of charged carrier liquid T and the reagents Rl, R2 into the reaction channel K can be controlled. For example, an axial lamination of the individual components in the reaction channel K can be achieved, in order to improve mixing. Transport of the liquid is effected here, as in the embodiment of Fig. 1, by gravity.

A third construction is illustrated in Fig. 3. Here the valves of Fig. 2 have been replaced by pumps 20-24. The pumps are controlled by a control device 26, which in turn is connected to a detector 25. The pumps are in the form of micropumps, that is, they have a very small capacity. The pumps 20-24 can be operated synchronously or can be harmonized exactly with one another by way of the control device 26. When the pumps are operated intermittently, the flow through the apparatus can be reduced, without the response time being appreciably adversely affected. In that case a measurement result is merely less often available.

In a manner not illustrated, yet further tanks can be provided, which contain, for example, a cleaning liquid or a descaling liquid. With the pumps, also not illustrated, provided for that purpose, which are also controlled by the control device 26, the apparatus can be cleaned or descaled.

The embodiment illustrated in Fig. 3 operates as follows: first of all the carrier liquid T is pumped by means of the pump 20 through the serpentine element 6. At the same time or later, the pumps 21, 22 can be operated in order to feed the reagents Rl, R2 into the reaction channel K. There may, of course, be pauses between the individual pump operations. That depends on which reactions are to take place. When the pumps 23, 24 are operated, the liquid, which comprises a mixture of the charged carrier liquid T and the reagents Rl, R2, is pumped into the detector 25. A first measurement can be taken here. After a time that is controlled by the control device 26, a second measurement can be taken. The difference between the two measured values is an expression of the concentration of the species in the liquid in the tank 5. On conclusion of the measurements, the two pumps 23, 24 can be set in operation again in order to convey the liquid into the collecting tank 11. Alternatively, of course, a combination of pumps and valves can be used in order to achieve the desired flow control.

When the pumps close their outlets during a suction stroke and their inlets during a pressure stroke they also function as valves.

When the mixture is in the detector 25, it is also possible to carry out more than two measurements. Basically, measuring can be continuous, that is, many measurements can be carried out at very small intervals. In this way a curve characteristic can be recorded which reproduces the progress of the reaction. Basically, the difference between the first and the last measurement is sufficient to allow conclusions to be drawn about or calculation of the concentration of the species. Moreover, a great deal of further partial information is available from which inferences can likewise be made about the concentration. The individual items of information, some of which are redundant, can be compared with one another and used for a validity check.

Such a curve characteristic can then be evaluated in great detail. For that purpose, however, considerable calculation effort is sometimes necessary. It is easier if these characteristics are compared with characteristics that have already been stored. Neural networks, for example, can be used for that purpose. If the recorded curve characteristic is "similar" to a stored curve characteristic, then the concentration also corresponds to the concentration at the time the stored curve characteristic was recorded. The appearance of the curve characteristic or the signal form is an expression of how the reaction is progressing. Normally, the characteristics have certain coincident features. When the recorded characteristic differs in one of these features from a "normal" characteristic, which can be stored, for example, that is an indication that there may be a fault in the apparatus.

The control device, which can contain the necessary memory for storing the stored curve characteristics, can contain an additional memory or memory region in which a "catalogue" of faulty forms of the characteristics is stored. Through a comparison with that catalogue the control unit is able to find the probable cause of the fault. The control unit can then carry out further analyses to analyze the fault exactly. When the fault has been found, the apparatus itself can attempt to eliminate the fault. For example, a cleaning or descaling operation can be effected or the displacement of a pump can be adjusted. When the established fault requires intervention for maintenance, a corresponding signal can be delivered.

Claims

Patent Claims

1. Analysis apparatus for the continuous or semi- continuous determination of a species in a fluid, with a sample-removing device, a reaction channel, which is connected to the sample removal device and to at least one reagent supply device, a detector arrangement and a control device, characterized in that by means of the detector arrangement (9, 10; 13; 25) the control device (12, 14, 26) determines a signal dependent on a reaction product in a mixture of fluid and reagent at a first time and at a later time, and by means of the difference in the signals determines the concentration of the species in the fluid.

2. Analysis apparatus according to claim 1, characterized in that the detector arrangement (9, 10) comprises several detectors (9, 10) arranged one behind the other on a reaction channel (K) .

3. Analysis apparatus according to claim 1 or 2, characterized in that there are provided flow control means (15-19; 20-24) which are connected to the control device (14, 26) and stop the flow of the mixture in the reaction channel (K) between the first and the later times.

4. Analysis apparatus according to claim 3, characterized in that the flow control means (15-19; 20-24) comprise valves (15-19) .

5. Analysis apparatus according to claim 3 or 4 , characterized in that the flow control means (15-19; 2- 24) comprise micropumps (20-24).

6. Analysis apparatus according to any one of claims 3 to 5, characterized in that separate micropumps (20; 21, 22) which are synchronizable with each other are provided for both fluid (T) and reagent (Rl, R2) .

7. Analysis apparatus according to claim 5 or 6, characterized in that the micropumps (20-24) close their outlets during a suction stroke and their inlets during a pressure stroke.

8. Analysis apparatus according to claim 7, characterized in that the duration of the suction stroke is shorter than the duration of the pressure stroke.

9. Analysis apparatus according to any one of claims 1 to 8, characterized in that the detector arrangement 99, 10; 13; 26) has at least one photometer, in particular a spectrophotometer.

10. Analysis apparatus according to any one of claims 1 to 8, characterized in that the detector arrangement (9, 10; 13; 25) comprises at least one conductivity sensor and/or at least one ion-selective electrode and/or at least one pH sensor.

11. Analysis method for determining the concentration of a species in a fluid, in which method at least one reagent reacting with the species is added to the fluid in order to produce a reaction product dependent on the concentration of the species, and the reaction product is evaluated in a detector device through which the fluid is channelled, characterized in that at a first time after addition of the reagent a signal dependent on the reaction product is determined and at a further, later, time a signal dependent on the reaction product is again determined and the difference in the signals is formed in order to determine the concentration of the species in the fluid.

12. Method according to claim 11, characterized in that in the period between the first time and the further time, further signals dependent on the reaction product at the particular measuring times are determined.

13. Method according to claim 12, characterized in that a curve characteristic is determined from the signals.

14. Method according to claim 13, characterized in that the curve characteristic is compared with a curve characteristic stored in a memory.

15. Method according to claim 14, characterized in that inferences about the faults in the analysis apparatus are drawn from specific differences between the determined and the stored curve characteristics.

16. Method according to claim 15, characterized in that fault-elimination routines are initiated when predetermined faults occur.