WO2005045422A1

WO2005045422A1 - A biosensor having exchangable biopart

Info

Publication number: WO2005045422A1
Application number: PCT/SE2004/001616
Authority: WO
Inventors: Bo Mattiasson; Jing Liu
Original assignee: Bo Mattiasson; Jing Liu
Priority date: 2003-11-06
Filing date: 2004-11-08
Publication date: 2005-05-19
Also published as: SE0302935D0

Abstract

The present invention relates to an integrated flow cell for biosensor analysis in flow-through or flow-injection assay and designed for easy renewal of bio-receptor, which is essential for inducing a catalytic reaction (or reaction chain) or non-catalytic molecular interaction (or interaction chain). One part of the flow cell (1) can be plugged in or pulled out of the other part (2) along a track. A layer of bio-receptor (3) can be formed inside an internal hollow channel of a plastic sheet (4) when two parts of the flow cell are held together and filled with the bio-receptor component. The bio-receptor layer is retained in direct spatial contact with a transducer (5) and separated from the carrier and sample by a selective membrane (6) to prevent leakage of the receptor and direct interferences from the carrier flow. The bio-receptor inside the internal hollow space can be replaced without dissembling the sensor by removing the used receptor and injecting a new one through two small conduits (7) that connect the layer of bio-receptor to the outside. An automated flow injection assay apparatus that is used in conjunction with the flow cell is also described.

Description

TITLE

A biosensor having exchan^able biopart DESCRIPTION 5 Technical field The present invention relates toa system for analysis, in particular an integrated flow- cell involving a biosensor with a special arrangement for easy renewal of bio-receptor in insoluble form, an automated flow injection assay (FIA) apparatus that is used in conjunction with the flow cell, and their utilization for detecting analytes in a complex 10 medium.

BACKGROU ND OF THE INVENTION In recent years, biosensors have generated increasing interest among scientists because of their ability to quickly determine the concentration of a given substrate, and for their 15 simplicity relative to other analytical techniques. A biosensor is an integrated device that is capable of providing specific quantitative or semi-quantitative analytical information using a biochemical receptor, which is retained in direct spatial contact with a physical transducer. Different kinds of biological material have been used as the bio- receptor for sensor construction. Various kinds of biosensor can be classified according 20 to whether the specific binding process results in a catalytic process or involves non- catalytic molecular interaction.

In general, biosensors have a limited operational period due to lysis or inactivation of the immobilized bio-receptor. Consequently, the old or rather used bio-receptor has to 25 be replaced with new one and re-construction of the sensor is often required. Bio- receptors are immobilized most commonly by adsorption (i.e. the biological material is simply in direct contact with the transducer surface), entrapment (i.e. physical entrapment behind a semi-permeable membrane or physical entrapment into a polymeric network), covalent binding (i.e. a stable linkage based chemical functionalities 30 both on the support and on the biomolecule), crosslinking (i.e. intermolecular covalent linkages being formed with the aid of a bifunctional crosslinking agent), and hybrid techniques where several immobilization methods are simultaneously applied. These methods involve complicated procedures that require both labour and technical skill. Therefore, they are not suitable for industrial applications in the field, where simple and 35 robust methods are preferred. To overcome the problem, easy renewal of immobilized bio-receptors seem to be an interesting and promising solution. Enfors and Nilsson, in Enzyme Microb. Technol. (1979), vol. 1, pages 260-264 describe an autoclavable enzyme electrode based on pH probe for fermentation. The electrode is immersed in the fermentation liquor during the measurement and is operated by injecting an enzyme solution into a chamber for enzymatic reaction after the outer surface of the electrode being sterilized. The electrode, however, has a disadvantage that the distance between the transducer tip and the membrane is difficult to keep constant. This leads to a severe problem with reproducibility. Furthermore, the thickness of the reaction chamber cannot be precisely adjusted, and a second transducer must be introduced as the reference in the system.

The injection of the recognition element (SIRE) technology designed by Kriz et al., as described in Biosens. Bioelectron. (1996), vol. 11, pages 1259-1265 and in US Patent No. 5,417,923 is based on the flow injection of soluble recognition elements into a buffer solution passing through an internal chamber inside the sensor, and as such it needs replacement of recognition elements for each assay.

Magnetic force has also been used for entrapping the biocatalysts on electrode surfaces for biosensors construction. Miyabayashi et al., in Anal. Chim. Acta (1988), vol. 213, pages 121-130 and in Biotechnology and Applied Biochemistry (1989), vol. 11, 379-386 presented a magnetic electrode in which a homogeneous magnetic field is applied to the sensing part of the electrode, thereby making it possible to immobilize enzyme-carrying magnetic particles in a reproducible way. Similar methods have also been applied on living cells by Sakai et al., in J. Ferment. Bioeng. (1995), vol. 80, pages 300-303 and antibodies (antigens) by Santandreu et al., in Biosensors and Bioelectronics (1998), vol. 13, pages 7-17 and Li et al., in Anal. Chim. Acta (2003), vol. 481, pages 191-198. The electrode with magnetic particles is based on the use of the same flow system for introduction of the biocatalyst as is used in the subsequent analytical procedure. That may be a suitable method when dealing with pure solutions and immobilized pure enzymes. This is however not useful for handling microorganisms that need to be kept in good metabolic condition and at the same time with a reasonable balance between different constituents. Moreover, it is difficult to control the characteristics of the immobilized biocatalysts, such as the thickness of the bio-layer on the electrode surface.

In recent years a number of reports have been published on development of new analytical methods for cost-effective on-line monitoring of certain water quality parameters, such as BOD (Biological Oxygen Demand) as summarized by Liu and Mattiasson, in Water Research (2002), vol. 36, pages 3786-3802. These new potential on-line methods will facilitate real time monitoring of environmental treatment processes, thereby contributing to a better understanding of the dynamic characteristics of biological treatment processes.

The biosensor of the present invention, as described below in the present application, can be used in the assessment of BOD in accordance with the prior publications. The new sensor construction will facilitate long-term utilization, thus making the BOD biosensor suitable for applications in real life under severe conditions.

The determination of BOD is an empirical test in which standardized laboratory procedures are used to determine the relative oxygen requirements of wastewater, effluent and polluted waters. The BOD values indicate the amount of biodegradable organic material (carbonaceous demand) and oxygen used to oxidize inorganic material such as sulfides and ferrous iron. The test has its widest application in measuring waste loadings to treatment plants and in evaluating the BOD removal efficiency of such treatment systems. BOD is conventionally determined by taking a sample of water, aerating it well, placing it in a sealed bottle, incubating for a standard period of 5 or 7 days at 20 + 1 °C in the dark, and determining the oxygen consumption in the water at the end of incubation. The conventional BOD test has the limitations of being time consuming, and consequently it is not a suitable method for on-line process monitoring.

Fast determination of BOD could be achieved by the biosensor-based methods based on the respirometric principle. Most of the previously reported BOD sensors have been biofilm-type whole-cell-based microbial sensors, which rely on measurement of the bacterial respiration rate in close proximity to a suitable transducer. A common feature of these sensors is that they consist of a microbial layer sandwiched between a porous cellulose membrane and a gas-permeable membrane as the bio-receptor. The bio- receptor is an immobilized microbial population that can bio-oxidize the organic substrate to be quantified. The response is usually a change in concentration of dissolved oxygen or other phenomena such as light emission. A physical transducer is used to monitor this process. The result is a change in an electrical or optical signal that is amplified and correlated to the content of biodegradable material measured.

There are, however, a whole range of BOD-sensors described over the years. Many are based on use of pure cultures or at best of mixtures of two pure cultures as reviewed by Liu and Mattiasson, in Water Research (2002), vol. 36, pages 3786-3802. In general, the whole-cell based BOD biosensors have a limited operational period due to lysis of the immobilized microbes. Consequently, the bio-receptor has to be changed, and the sensor needs to be re-fabricated. The microbial cells are immobilized most commonly by adsorption, e.g . the cells are placed directly on a porous cellulose membrane by suction, or hydrogel entrapment, i.e. the cells are entrapped on the surface of a porous matrix membrane by using an aqueous solution of polyvinyl alcohol (PVA) or poly(carbamoyl)sulfonate (PCS). These methods involve complicated procedures that require both labour and technical skill. Therefore, they are not suitable for field service, where simple and robust methods are preferred.

The novel design of BOD biosensor according to the present invention has a new feature of easy and fast replacement of bio-receptor without dismantling the biosensor. The present construction therefore goes far beyond what is earlier published in that it makes it possible with frequent shifts of the biocomponents - from almost continuous replacement of the catalytic components to intermittent changes when it is needed.

The present invention separates itself from described published constructions in several ways. It has solved all the drawbacks of the cited publications and achieved a construction that gives reliable and reproducible results. Furthermore, by electronic control of the analysis sequence and data evaluation and processing, it is a sensor system that is capable to meet far more advanced analytical challenges than the previously published constructions.

SUMMARY OF THE INVENTION

A first object of the present invention is to provide an integrated flow-cell for easy and rapid immobilization and replacement of bio-receptor without dismantling the sensor system. The flow-cell is designed for use together with the flow through and flow injection techniques, but it can easily be used as a separate electrode-sensor.

A second object of the present invention is to introduce a general approach for solving the problem of losing sensitivity and poor stability in biosensors. That is to allow rapid and convenient renewal of the bio-receptor on a regular basis to compensate the loss of catalytic activity or non-catalytic molecular interaction.

A third object of the present invention is to provide a fully automated FIA apparatus that is used in conjunction with the flow cell described and service as a universal platform for various sensor applications.

One embodiment of the present invention is to develop a biosensor system for biochemical oxygen demand (BOD) measurement according to the invention and to use the automated FIA apparatus for detecting easy biodegradable organic matter in wastewater.

DETAILED DESCRIPTION OF THE PRESENT INVENTION The present invention solving the above problems is characterized in having exchangeable bio-part, wherein the bio-part in insoluble form is arranged to be introduced and replaced without having to dismantle the biosensor, thereby maintaining sensitivity and stability of the biosensor.

A preferred embodiment is characterized by a construction allowing a fixed distance between the sensitive part of a transducer, an electrode, and the contact surface with the surrounding medium that is to be monitored.

A further preferred embodiment is characterized by an arrangement for introduction of the bio-part in a restricted space defined by the surface of an electrode on one side and a membrane that is in contact with the surrounding medium on the other side, the membrane being arranged in such a way that the distance between the transducer surface and the surrounding medium is adjustable, and whereby the biosensor is kept mounted while replacing the bio-part.

A preferred embodiment is characterized by that the biosensor is used in conjunction with an integrated flow-cell, the latter being constructed such that the bio-part is arranged to be introduced and removed via a set of at least two thin channels while the flow system is present in an arrangement for applying the biosensor in analysis of samples that are taken from a medium to be analyzed.

A further preferred embodiment is characterized by that the sensitivity and response time is controlled by using spacers of varying thickness and membranes of different permeabilities at the tip of the electrode, thereby controlling the thickness of the layer of the bio-part and at the same time controlling the flux of metabolites that w.ill be analyzed by the bio-part.

A preferred embodiment is characterized in that the biosensor is designed to be used by immersion in the reaction mixture to be analyzed.

A further preferred embodiment is characterized by being based on a mixed culture of organisms operating as a metabolic consortium. A preferred embodiment is characterized by being based on a mixture of two or more well-defined microbial species/strains that cooperate metabolically.

A further preferred embodiment is characterized by being based on a bio-part that is active by binding the target molecule and thereby causing a signal that is picked up by the transducer.

A preferred embodiment is characterized by being based on a microbial organism, either alone or in combination with another bio-part that will change its metabolism as a result of a chemical stimulus from external sources.

A further aspect of the invention relates to a biosensor system with easily replaceable bio-part being an integrated part of a flow injection system which is capable of performing sequence operations automatically including preparations of sample in a desired concentration range, sample loading, sample injection, signal recording, data processing and self-cleaning of the system.

The embodiment of the present invention will now be described in more detail with reference to the accompanying drawing Fig. 1. One part of the flow cell 1 is designed to be plugged in or pulled out of a second part 2 forming a housing for the flow cell 1 along a track. A Clark-type probe 5 for dissolved oxygen could be screwed into the flow cell part 1. The flow-cell also comprises a plastic sheet 4 with an elliptical channel to form a chamber 10 for a layer of bio-receptor 3 when two parts of the flow-cell are held together and filled with a bacterial cell paste. The bio-receptor layer is retained in direct spatial contact with the oxygen probe 5 and separated from a carrier and sample by a piece of dialysis membrane 6 to prevent leakage of cells and direct interferences from the carrier flow. The oxygen probe 5 is the transducer 5 here being a BOD sensor, also comprising an electrode. Instead of an electrode an optical sensor or another electrochemical sensor can be used to read changes in metabolism occurring in a bacterial cell paste 3. Two small conduits 7 are introduced to penetrate the flow cell part 1. They are placed on opposite sides and used to transport the bacterial cell paste to the surface of oxygen probe 5 for immobilization on the dialysis membrane 6. The bio-receptor, bacterial cell paste, is easily injected into the internal hollow space or chamber through one of two small conduits 7 from the outside. Likewise, used cells can be removed in the same manner. Normally, used cells are removed from the chamber simultaneously with introducing a new batch of bacterial cell paste. Therefore, the bio- receptor layer can be replaced without dismantling the sensor by removing the old, used receptor and injecting a new one. The thickness of bio-receptor layer is controlled by selecting the thickness of the plastic sheet 4. Additional fine adjustment can be carried out by tuning the transducer 5 along the screw thread 8 on the surface of the transducer and flow cell 1. For the flow cell chamber in wall-jet type, a small tube 9 is used as inlet channel and the inlet port can be adjusted to stretch out and draw back in order to adjust the distance between the inlet port and the sensing part of the oxygen probe 5. This allows adjusting the flow dynamics of carrier and sample solution inside the flow cell chamber. The thickness of the sheet 4 will work as a spacer, thereby controlling the thickness of the layer of the bio-part and at the same time controlling the flux of metabolites that will be analyzed by the bio-part. The movement of the transducer 5 using the threaded part 8 will also work as a spacer.

This biosensor system is an integrated part of a FIA apparatus which is capable of performing sequence operations automatically including preparations of a sample in a desired concentration range, sample loading, sample injection, signal recording, data processing and self cleaning of the system. The major hardware features of this FIA apparatus consists of a sample preparation unit A, a FIA unit B, an electronic control unit C, a signal converter and amplifier unit D, and data acquisition (DAQ) unit E. Fig. 2 shows a block diagram representation of the major components of the integrated system. However, the medium to be analyzed can be transformed to the flow cell using an either manual or automated sampling .device.

The sample preparation unit A is responsible for diluting the sample by the carrier liquid into a desired concentration. A variable volume pump (0-1250 μl) is used to precisely dose the sample to a 10 ml round bottomed flask with help of 1 three-way and 2 two- way diaphragm valves. A peristaltic pump with remote switching is used to control the carried liquid flow into or out of the flask together with 2 three-way diaphragm valves. The sample is well mixed with the carrier liquid in the flask by a small stainless stirrer that is driven by a DC motor, and then diluted into the desired concentration in the flask with help of an optical coupler circuit for liquid level control. A stepper-motor interface is responsible to control the dosing volume of variable volume pump according to the control signal from the control unit.

The FIA unit B keeps a flowing stream of carrier liquid in constant flow rate, as well as injects and delivers the diluted sample to the BOD biosensor. It consists of a peristaltic pump with analogue speed control, a cheminert valve with microelectric actuator, a sample loop with fixed volume and a storage tank for the air-saturated carrier liquid. The switching of the cheminert valve is triggered by a control signal from the control unit. The control unit C consists of an electronic circuit board that receives the operational parameters setting from a keyboard and controls the operational sequence including sending control signals to the sample preparation unit, triggering sample loading and injection, triggering data acquisition and starting self-cleaning procedure, etc. The key component of the circuit board is a 28-pin programmable CMOS Flash-based 8-bit microcontroller (PIC16F873, Microchip Technology Inc, USA) that serves as a central processing unit for various control tasks. In order to form an 8-bit data bus for signal transfer, the circuit board also consists of a high-speed l-of-8 decoder/demultiplexer (SN74LS138N, ON Semiconductor, USA), an octal non-inverting buffer (SN74HC244N, Texas Inst Semiconductors, USA), and an octal high-speed register (SN74LS273N, ON Semiconductor) . The keyboard includes seven shared functional buttons for starting (or stopping) the analysis procedure and setting the operational parameters, such as initial concentration of the diluted sample, concentration increment, time interval between analyses, number of analyses, as well as increment and decrement of digital settings. A liquid crystal display (LCD) is used to display the parameter settings and operational information in real-time.

The signal converter and amplifier unit D is responsible for converting original sensor signal from current changes to potential one with adequate amplitude.

The data acquisition unit E is used to record analogue signal from unit D, calculate slope of linear portion of the signal or the peak height, perform the calibration, as well as save all the data in files for off-line reference. The major components of this unit include a data acquisition (DAQ) board (such as PCI-MIO-16E-4 multifunction I/O board for PCI bus computers, National Instruments, USA) and a standard PC with Windows 9x operational system, or other equivalent operational system.

The sample preparation unit A and electronic control unit C are kept in a plastic case and are powered by a standard PC power supply that is capable of delivering 5 V at 0.3 A to the circuit board and 12 V at 0.3 A to the valves. The FIA unit B, signal converter and amplifier unit D and data acquisition unit E are powered directly by 220 V AC 50 Hz.

Two software programs have been used for the instrument, one for controlling the analysis sequence and the other for DAQ and graphic user interface (GUI). Figure 3 illustrates the major operation steps and functions according to the software programs. The control software was written in assemble language and was programmed into the 128 bytes EEPROM (Electrically Erasable Programmable Read Only Memory) data memory of the microcontroller. The instrument therefore can be used independently without occupying the resource of CPU in a PC. This allows the users to develop a simple distributed control system together with other equipment for a multi-parameter application when a few process parameters need to be monitored and controlled in realtime. The microcontroller serves as a local processor that takes care of the individual operational task, whereas the PC acts as a central computer for the data acquisition, data analysis, data communication, data presentation and process control. The software program controls the analysis procedure starting at rinsing the sample preparation flask with carrier liquid and followed by sending a series of digital bit-stream to the stepper- motor interface in order to pre-condition the variable volume pump and precisely control its dosing volume. The program ensures the sample is diluted into a desired concentration by checking the output signal given by the optical coupler circuit. Once the sample preparation has been done, the cheminert valve is switched into a position that allows the diluted sample to fill up the injection loop. Then the cheminert valve is shifted to an alternative position that allows the sample to be delivered to the sensor by the carrier flow. The DAQ is simultaneously triggered with the sample injection. In order to eliminate the risk of microbial contamination on internal surface of the flow system during a long-term operation, deionized water is used to wash out the sample residues inside the variable volume pump and tubing. The sample flask and tubing are also rinsed by carrier liquid before and after each analysis.

The software of DAQ was created in LabVIEW 6i (National Instruments). This software has a graphic user interface that contains a virtualized operational panel necessary for display of the monitoring sensor signal in real-time. Analogue signals are sampled at a pre-defined rate by the DAQ board, digitized into 12-bit words, and then transferred into the PC. The sensor signal is displayed in real-time during the DAQ and saved in a memory buffer. A digital low pass filter has been used to eliminate the high frequency noise. In the next step, the filtered data is analyzed and the slope of linear fraction of the data stream or the peak height is calculated. Based on the status of parameter settings, the software can either perform the sensor calibration procedure or calculate the BOD value of the analyzed sample according to the pre-calibrated coefficients. Finally, all data as well as status parameters of analysis are saved in corresponding files for off-line reference. EXPERIMENTS AND DISCUSSION

To exemplify the invention, a preliminary study has been made on the BOD biosensor constructed according to the present invention.

The sensor measurements were carried out in the initial-rate mode using a flow injection system, resulting in 60 s for one sample analysis followed by a recovery time less than 10 min. Typical response curves of the sensor are shown in Fig. 4. Three different concentrations of a standard wastewater are tested, i.e. (O) 12.5 mg BODs -l^"1, (o) 25 mg BODs -l^-1, (Δ) 37.5 mg BODs -l^-1.

The sensor performance achieved showed wide detection linearity over a range of 5-700 mg BODs-l^-1 under the conditions of carrier flow as 0.6 ml-min^-1 and the sample injection volume as 500 μl (Fig. 5). A generally good agreement between the BOD values estimated by the biosensor and the conventional 5-day test was also achieved. Furthermore, the precision test was in the control range (i.e. repeatability < 1+7.5 %] , reproducibility < |±7.3 % | ). The sensor could be used over 1 week in continuous test, however the best performance was found within the first 24 hours where standard deviation of the sensor response was ±2.4 %.

In order to carry out the calibration procedure easily, the apparatus was designed to provide on-line analysis algorithms for estimating linear equation of the sensor response based on the slope of the initial linear fraction of the response curves. As demonstrated in Fig. 6, the linear equation estimated by the instrument was compared with the one made manually. The slopes, intercepts and R² values were 0.1565 nA-l-s^-1-mg^-1, 0.3732 nA-s^-1, 0.9986 and 0.1567 nA-l-s^-mg^-1, 0.3826 nA-s^-1, 0.9987, respectively. Both slopes and intercepts are very close. A similar linear equation could be obtained automatically by using the peak height of the response curves. The R² value was 0.9985.

The bio-receptor of the sensor could be easily renewed by injecting new bacterial paste without dismantling the sensor system. The replacement of immobilized cells and the sensor performance therefore were tested in turn for 8 times by renewal of cell paste, stabilization of the sensor signal, and analyzing a single diluted standard sample with BOD₅ of 25 mg-l^"1. As partly demonstrated in Fig. 7, the sensor was ready to use after the cells immobilization for about 30 min and a good reproducibility of the sensor response was obtained. The standard deviation was ±3.3 %. As illustrated in Fig. 8, the sensor signal was stabilized for about 30 min after the first cells immobilization and then calibrated by the diluted standard solution with BOD₅ in 10 mg-l^"1, 20 mg-l^"1 and 30 mg-l^"1, respectively. The used cell paste was then removed and a newly prepared cell paste was injected with help of a syringe. The newly prepared sensor was stabilized again for about 30 min and calibrated in the same manner. A good reproducibility of the sensor response was observed. The linear equations of these two calibration-curves are very close. The slopes and intercepts are 0.1536 nA-l-s^-1-mg^-1, 0.2777 nA-s^-1, and 0.1528 nA-l-s^-1-mg^"1, 0.2342 nA-s^-1, respectively.

The invention has now been described with reference to an embodiment. However, it is apparent to the one skilled in the art that further embodiments are available within the scope of the present, accompanying claims.

Claims

CLAIMS 1. A biosensor having exchangeable bio-part, whereby the bio-part in insoluble form is arranged to be introduced and replaced without having to dismantle the biosensor, thereby maintaining sensitivity and stability of the biosensor characterized by a construction allowing a fixed distance between the sensing part of a transducer (5), and the contact surface (6) with a surrounding medium that is to be monitored.

2. A biosensor according to claim 1, characterized by an arrangement for introduction of the bio-part in a restricted space (10) defined by the surface of an electrode (5) on one side and a membrane (6) that is in contact with the surrounding medium on the other side, the membrane (6) being arranged in such a way that the distance between the transducer surface (5) and the surrounding medium is adjustable, and whereby the biosensor is arranged to be kept mounted while replacing the bio-part.

3. A biosensor according to claims 1, and 2, characterized in that the biosensor is used in conjunction with an integrated flow-cell (1), the latter being constructed such that the bio-part (3) is arranged to be introduced and removed via a set of at least two thin channels (7) while the flow system is present in an arrangement for applying the biosensor in analysis of samples that are taken from a medium to be analyzed.

4. A biosensor according to claims 1, 2, and 3, characterized in that the sensitivity and response time is controlled by using spacers of varying thickness and membranes of different permeabilities at the tip of the electrode, thereby controlling the thickness of the layer of the bio-part and at the same time controlling the flux of metabolites that will be analyzed by the bio-part.

5. A biosensor according to claims 1, and 2, characterized by being designed to be used by immersion in the reaction mixture to be analyzed.

6. A biosensor according to claims 1, and 2, characterized by being based on a mixed culture of organisms operating as a metabolic consortium.

7. A biosensor according to claims 1, and 2, characterized by being based on a mixture of two or more well defined microbial species/strains that cooperate metabo cally

8. A biosensor according to claim 1, and 2, characterized by being based on a bio-part (3) that is active by binding the target molecule and thereby causing a signal that can be picked up by the transducer.

9 A biosensor according to claims 1, 2, and 8, characterized by being based on a microbial organism, either alone or in combination with another bio-part, that will change its metabolism as a result of a chemical stimulus from external sources.

10. A biosensor according to one or more of claims 1-9, wherein the sensor comprises en oxygen probe, an electrode, an optical sensor or an electrochemical sensor.

11. A biosensor system with easily replaceable bio-part that is an integrated part of a flow injection system which is capable of performing sequence operations automatically including preparations of sample in a desired concentration range, sample loading, sample injection, signal recording, data processing and self- cleaning of the system.

FIGURE LEGENDS

Fig 2. Block diagram representation of the major components in the BOD biosensor system. Sample preparation unit (block A), FIA unit (block B), electronic control unit (block C), BOD biosensor (block D), and data acquisition unit (block E). Large shaded ^' arrows indicate analog electrical connections, whereas small arrows specify the digital electrical connections. Large unshaded arrows designate fluidics connections. Fig 3. Block diagram representation of the major analysis steps. Large shaded arrows indicate the sequence of analysis procedure controlled by the microcontroller, whereas large unshaded arrows represent the data acquisition and analysis procedure managed by PC. Small arrow with solid fine line specifies the digital electrical signal for triggering the data acquisition. Small arrows with dashed bold line designate the recycle of analysis procedure. Fig 4. Full sensor responses for diluted standard solution in various concentrations. Fig 5. A calibration curve of the BOD biosensor. The carrier (0.01 M KH₂P0₄-K₂HP0₄, pH

7.1) flow rate was 0.6 ml-min-1 and the sample injection volume was 500 ml. The relationship of sensor response and the BOD5 of the standard solution was linear over a range of 5-700 mg-l-1. Fig 6. Calibration curves of the BOD biosensor, (o) Computer-generated calibration curve, (o) manual-made calibration curve.

Fig 7. A fraction of the time-scale for a series of bio-layer renewal, signal stabilization and sample analysis of a single diluted standard solution (25 mg BOD5-I-1) in turn. The recorded signal was the current amplitude between the initial base line current and the sensor output signal after a sample injection. The length from zero point to the first dashed line in X-axis and the length between the second and third dashed lines show that it took about 30 min to be ready for sample analysis after the biocatalyst renewal. The standard deviation of the sensor response was ±3.3 %.

Fig 8. Time-scale for biocatalyst renewal, stabilization of the sensor signal and sample analysis of the standard solution in 10, 20 and 30 mg BOD5-I-1. Length of the solid bars and dashed bars represents the time required for sensor signal recording (1 min) and recovery (10 min), respectively, whereas their positions in Y-axis indicate the amplitude of the sensor response. The length from zero point to the first dashed line in X-axis and the length between the second and third dashed lines show that it took about 30 min to be ready for sample analysis after the bio-layer renewal.