METHOD, DEVICE, AND SYSTEM FOR THE AUTOMATED DETERMINATION OF OPTICAL DENSITIES OR OF THE CHANGE IN OPTICAL DENSITIES OF REACTION MIXTURES IN SHAKEN REACTORS

Abstract

The invention relates to a method, to a device, and to a system for the automated determination of optical densities or of the change in optical densities of reaction mixtures in shaken reactors during shaking operation. Methods and devices currently used therefor are often unreliable, are susceptible to environmental and process factors, or require interruptions to the shaking operation that impair the process control. The problem addressed by the invention is that of specifying a method and a device for the automated determination of optical densities or of the change in optical densities of reaction mixtures in shaken reactors during shaking operation that operate reliably under various environmental and process conditions. This problem is solved by means of a new measurement method, wherein the reaction mixture distribution, which periodically fluctuates because of the shaking action, is used to record measurement points (20/21) of transmission/scattered-light measurements, which measurement points fluctuate periodically as a result of shaking. All measurement points (20/21) of a measurement operation are combined into a measurement series (34), from which the optical density and/or the change in the optical density, and other process parameters, can be determined with high reliability by means of suitable mathematical methods. The invention is suitable in particular for biotechnological, pharmaceutical, chemical, and biochemical screening and optimization and process-monitoring applications.

Claims

1. Method for determining the optical density and/or the change in the optical density of a reaction mixture 2 in a shaken reactor 1, characterised in that light from at least one light source 4 enters the reaction mixture 2, and the light exiting the reaction mixture 2 is detected by at least one light sensor 5/6, and during the detection of the light by the at least one light sensor 5/6 the reactor 1 and the reaction mixture 2 are shaken, and the light is detected by the at least one light sensor 5/6 at a frequency such that the shaking frequency is not an integer multiple of the detection frequency, and at least two measurement points 20/21 detected by at least one light sensor 5/6 in a particular time interval are combined into a series of measurements 34.

2. Method according to claim 1, characterised in that the light is detected by the at least one light sensor 5/6 at a frequency higher than the shaking frequency, and/or at least one light path of at least one light source/light sensor pair 4-5/6 is orientated non-parallel to a shaking axis and/or shaking plane, and/or there is no relative movement between the reactor 1 and at least one light sensor 5/6.

3. Method according to either claim 1 or claim 2, characterised in that, from at least one series of measurements 34 determined in a particular time interval, the optical density and/or the change in the optical density and/or at least one fluid-mechanical parameter and/or at least one fluid-mechanical property of the reaction mixture 2 in the measured time interval is determined by means of at least one suitable mathematical/computing method 36 and/or at least one model function.

4. Method according to any of claims 1 to 3, characterised in that, from at least one series of measurements 34 and/or at least one time series 35 of a parameter of the reaction mixture 2 determined from at least one measurement point 20/21 and/or at least one series of measurements 34, at least one further parameter of the reaction mixture 2 is determined by means of at least one suitable mathematical/computing method 36 and/or at least one model function.

5. Method according to any of claims 1 to 4, characterised in that at least one model function is generated afresh and/or modified using at least one suitable mathematical method 36 and/or at least one suitable computing algorithm 36.

6. Method according to any of claims 1 to 5, characterised in that at least one metering system 12 for adding at least one substance is controlled as a function of the optical density and/or of the change in the optical density and/or of at least one other process parameter, and/or at least one further technical system which affects the process, in particular a system for temperature-controlling 25 the reaction mixture 2, is operated as a function of the optical density and/or of the change in the optical density and/or of at least one other process parameter.

7. Device for carrying out the method according to any of claims 1 to 6, characterised in that the device comprises at least one light source 4, which is positioned and orientated in such a way that the light emitted thereby can enter the reactor 1 and directly and/or indirectly enter the reaction mixture 2 under the conditions of at least one type of shaking, and the device comprises at least one light sensor 5/6, at least one light sensor 5/6 being positioned and orientated in such a way that it can directly and/or indirectly detect light emitted from the reactor 1 and/or the reaction mixture 2, and at least one light sensor 5/6 is positioned and orientated in such a way that the light detected thereby and emitted from the reactor 1 and/or the reaction mixture 2 has periodic changes due to shaking in at least one of the properties thereof, but in particular the intensity thereof, and the device comprises at least one processor 9/10 and/or is operated in combination with at least one processor 9/10, at least one processor 9/10 being used for at least one of the following purposes: recording the measurement values, storing the measurement values, processing the measurement values, representing the measurement values and/or processing results.

8. Device according to claim 7, characterised in that at least one light source 4 and/or at least one light sensor 5/6 are combined with at least one set of optics 38/39 consisting of at least one of the following components: a lens, an aperture, a prism, a diffraction grating, an optical gap, a polariser, an optical filter, an optical fibre, and/or an electronic system for changing the light source intensity 7 or the light sensor sensitivity 8, and/or at least one light sensor 24, which is positioned and orientated in such a way that it can only detect ambient light, but not any light exiting the reactor 1 and/or the reaction mixture 2.

9. Device according to either claim 7 or claim 8, characterised in that the device comprises at least one metering system 12 and/or at least one connection point for at least one metering system 12 for the automated addition of at least one substance to the reaction mixture 2, and/or comprises at least one system and/or at least one connection point for at least one system for temperature-controlling 25 the reactor 1 and/or the reaction mixture 2, and/or communicates with at least one processor via radio and/or wiring, and/or draws electrical energy from at least one storage battery 26 and/or another electrical energy source.

10. System for taking and processing measurements of the optical density and/or the change in the optical density of a reaction mixture 2 in a shaken reactor 1, characterised in that the system comprises at least one device according to any of claims 7 to 9, the system using a method according to at least one of claims 1 to 6.

Description

EMBODIMENT

[0072] Hereinafter, the invention is described in greater detail with reference to the drawings.

[0073] FIG. 1 is a schematic drawing of a general embodiment of the method and of the device.

[0074] FIG. 2 shows by way of example, for a reactor 1 under orbital shaking, the dependency of the measured light intensities on the shape and distribution of the reaction mixture 2 in the reactor 1.

[0075] FIG. 3 shows by way of example, for a reactor 1 under rocker shaking, the dependency of the measured light intensities on the shape and distribution of the reaction mixture 2 in the reactor 1.

[0076] FIG. 4 shows an embodiment of the device for shaking flasks under orbital shaking.

[0077] FIG. 5 shows an embodiment of the device for T-flasks under rocker or orbital shaking.

[0078] FIG. 6 is a schematic drawing of a non-modular embodiment of the device.

[0079] FIG. 7 is a schematic drawing of a modular embodiment of the device.

[0080] FIG. 8 schematically shows the use of a sensor array as a light sensor 5.

[0081] FIG. 9 is a graphical legend explaining some of the recurring components of FIGS. 1 to 8.

[0082] FIGS. 10 to 14 show example measurement and evaluation results which were obtained by the method according to the invention.

[0083] Like or functionally equivalent elements are provided with like reference numerals in the drawings, only the reference numerals required for understanding the drawing, including in context with the other drawings, being used in each case. Therefore, doubled-up reference numerals in like or similar components of a drawing are largely omitted. Recurring components of FIGS. 1-8 are summarized in FIG. 9 and are therefore not additionally explained.

[0084] FIG. 1 illustrates the basic mode of operation of the method and device by way of a schematic embodiment. The reaction mixture 2 is located in a shaken reactor 1. Light 17 is shone into the reactor 1 and the reaction mixture 2 from at least one light source 4. The shone-in light 17 interacts with at least one component of the reaction mixture 2, and leaves the mixture and the reactor 1 again, in particular but not exclusively as backscattered light 18 and/or as forward-scattered or transmitted light 19. The light 18/19 exiting the reaction mixture 2 thereupon interacts with at least one light sensor 5/6, which converts the detected light intensity into an analogue electrical signal. This is in turn digitised by an evaluation and control unit 8 belonging to at least one light sensor 5/6, and stored as a measurement value 20/21 and/or passed to at least one processor 9. The firmware 30 running on the processor 9 controls the measurement value detection, in particular but not exclusively in terms of the measurement frequency, the sensor sensitivity, and the amplification of the analogue measurement signal. In addition, the firmware 30 running on the processor 9 controls the radiation intensity of the light source 4 by way of the light source control unit 7.

[0085] Both the light source 4 and the light sensor 5/6 can be equipped with optics. The optics 38 modify the light 17 emitted by at least one light source 4 and shone into the reactor 1 and reaction mixture 2, whilst the optics 39 modify the light 18/19 emitted by the reactor 1 and reaction mixture 2, which is thereupon detected by at least one light sensor 5/6. The components useable for the optics 38 and 39 include, in particular but not exclusively, optical filters, lenses, lens systems, apertures, aperture systems, optical gaps and shutters, polarisers, half-wave and quarter-wave plates, diffraction gratings, prisms and optical fibres. In some embodiments, the optics 38/39 can be controlled by processors 9 having firmware 30.

[0086] The processor 9 and firmware 30 communicate with a more powerful processor 10 (for example a PC) and the user software 31 running thereon via at least one radio module 28 or at least one wired communication module 29 or even directly. This communication includes, in particular but not exclusively, the transmission of measurement and analysis data and other information, such as control commands for components of the device which are controlled by the processor 9, firmware and/or software parameters, time and clock signals, firmware updates and licence information.

[0087] A plurality of measurement data detected sequentially in a particular time interval and at a particular measurement frequency are combined into a series of measurements 34 by the firmware 30 on the processor 9 and/or the user software 31 on the processor 10. Processing a series of measurements 34 using suitable mathematical and computing methods and algorithms 36 subsequently provides at least one parameter and/or at least one property 37 of the overall process running in the reaction mixture 2. Suitable mathematical and computing methods and algorithms 36 may be implemented both in firmware 30 and in user software 31. In an advantageous embodiment of the invention, and also as is shown in FIG. 1, the majority of the methods and algorithms 36 used are implemented and run by the user software 31, since the computing power of the processor 10 is usually much higher and thus better suited to more complex methods and algorithms 36 than that of the processor 9.

[0088] Measurement signals 20/21, series of measurements 34 and process parameters/properties 37 determined at various moments can be combined by the processor 9 comprising firmware 30 and/or the processor 10 comprising user software 31 into at least one time series 35 of the relevant data type. Processing a time series 35 using suitable mathematical and computing methods and algorithms 36 subsequently provides at least one further parameter and/or at least one further property 37 of the overall process running in the reaction mixture 2.

[0089] On the basis of at least one type of data determined from the optical measurements during the process sequence (process parameters/properties 37, time series 35, series of measurements 34, measurement signals 20/21), automated intervention in the reaction sequence of the reaction mixture 2 in the reactor 1 is possible during the process. For this purpose, substances are added to the reaction mixture 2 via a metering system 12. In addition, by way of a temperature control system 25, the temperature of the reactor 1 and the reaction mixture 2 can be set so as to influence the current process in a targeted manner.

[0090] FIG. 2 and FIG. 3 illustrate the functional principle of the method and device using the example of a flask under orbital shaking as a reactor 1 (FIG. 2) and using the example of a general reactor 1 under rocker shaking (FIG. 3). In the two embodiments shown, a light source 4, which shines light 17 into the reactor 1 and into the reaction mixture 2, is located below the reactor 1 and the reaction mixture 2. A light sensor 5 for detecting backscattered light 18 is also located below the reactor 1 and the reaction mixture 2. A light sensor 6 for detecting transmitted and forward-scattered light 19 is placed above the reactor 1 and the reaction mixture 2. The light sensor 5 provides the scattering signal 21, and the light sensor 6 provides the transmission signal 20.

[0091] FIG. 2 and FIG. 3 each show five example shaking positions (I. to V.), the distribution of the reaction mixture 2 in the reactor 1 being selected schematically to illustrate the functional principle of the invention. Neither of the drawings shows actually taken measurement results. For better illustration of the distribution of the reaction mixture 2 in the reactor 1, both a lateral section and a view from below are shown for the flask under orbital shaking in FIG. 2. The lateral section is made, projecting from the plane of the drawing, along the line A-A of the view from below, in such a way that the view of the observer extends along the line B-C. By contrast, FIG. 3 is merely a lateral section.

[0092] In the flask under orbital shaking in FIG. 2, during shaking, a characteristic distribution of the reaction mixture 2 in the reactor 1 occurs, the rotation of the reaction mixture 2 in the reactor 1 resulting in the height of the fill level above the light source 4 and light sensor 5, and thus also the height of the fill level between the light source 4 and light sensor 6, continuously changing periodically. For a constant wavelength and intensity of the shone-in light 17, for constant external light, under the assumption of a constant optical density of the reaction mixture 2 and in the absence of inhomogeneities such as bubbles or agglomerates of reaction mixture components, the transmission signal 20 and the scattering signal 21 are each a function of the fill level of the reaction mixture 2 in the reactor 1 brought about by the distribution. A detail from the correspondingly periodic signal progression is shown schematically in the graph in FIG. 2.

[0093] When the fill level above the light source 4 and light sensor 5 is at a maximum (FIG. 2, shaking position I.), the scattering signal 21 of light sensor 5 is also at a maximum, whilst the transmission signal 20 of light sensor 6 is at a minimum, since in this case the smallest fraction of the light passes through the reaction mixture by comparison with the other shaking positions. At lower fill levels above the light source 4 and light sensor 5 which are brought about by the movement of the reaction mixture 2 in the reactor 1, the scattering signal 21 of light sensor 5 is correspondingly lower and the transmission signal 20 of light sensor 6 is correspondingly higher than in shaking position I. When there is no reaction mixture above the light source 4 and light sensor 5, the scattering signal 21 of light sensor 5 is at a minimum, and the transmission signal 20 of light sensor 6 is at a maximum (FIG. 2, shaking positions III. and IV.).

[0094] The signal progressions of a reactor under rocker shaking turn out analogously in FIG. 3. When the fill level above the light source 4 and light sensor 5 is at a maximum (FIG. 3, shaking position I.), the scattering signal 21 of light sensor 5 is also at a maximum, whilst the transmission signal 20 of light sensor 6 is at a minimum. When there is no or almost no reaction mixture above the light source 4 and light sensor 5, the scattering signal 21 of light sensor 5 is at a minimum and the transmission signal 20 of light sensor 6 is at a maximum (FIG. 3, shaking position V.).

[0095] All of the signals 20 and 21 recorded at an optical density which is sufficiently constant for measuring purposes can be combined into series of measurements 34, which can be used in accordance with the explanations for FIG. 1 to determine the optical density and/or the change in the optical density of the reaction mixture 2 and to determine further process parameters and process properties. As a result of the periodicity within a series of measurements 34, outliers caused by disturbing factors in the curve progression can be efficiently identified and eliminated, greatly increasing the reliability of the determination of the optical density and further process parameters and process properties.

[0096] In addition, from the form of these series of measurements 34, qualitative and quantitative conclusions regarding fluid-mechanical parameters and properties of the reaction mixture 2 can be reached, since the shape and distribution of the reaction mixture 2 in the reactor 1 and thus the measurement signals 20/21 within a series of measurements 34 are dependent on the fluid-mechanical parameters and properties of the reaction mixture 2. One fluid-mechanical parameter of significance to the process is the viscosity of the reaction mixture 2, which can greatly affect the shape and distribution thereof in the reactor 1. The invention therefore makes it possible, on the basis of the measurement method thereof, to assess the viscosity of the reaction mixture 2, and this is of great value for monitoring and optimising a wide range of biotechnological and chemical processes in which the viscosity changes during the process (for example as a result of cell growth, filamentous growth, formation of gel-forming substances, formation of polymers, etc.).

[0097] FIG. 4 shows an embodiment of the device for carrying out the method of the invention for shaking flasks under orbital shaking. In this case, a typical Erlenmeyer flask is used as a reactor 1, which contains the reaction mixture 2 and can be clamped together with the housing 16 placed below it in a commercially available clamping device or a holder. In the housing 16 there is a light source 4 along with a control unit 7, a light sensor 5 having an evaluation and control unit 8, and a processor 9 connected to the control units 7/8. The processor 9, which in an advantageous embodiment of the invention is formed by a microprocessor, is connected to a storage battery 26 to provide electrical power. This also applies to all of the other electronic components of the device; however, for reasons of clarity, the corresponding connections are not shown. Further, a temperature control system 25, consisting for example of a temperature sensor, a Peltier element and a control unit, is connected to the processor 9.

[0098] A second part of the device in FIG. 4 is attached to the cover 3 of the reactor 1. For detecting transmitted and/or forward-scattered light, there is a light sensor 6 in the cover (optionally behind a glass or plastics material plate for reasons of sterility), which, like the light sensor 5 placed on the underside of the flask, is connected to the evaluation and control unit 8. On the cover, there is an electronic metering system 12, which is connected to the processor 9 and, under the control thereof, can pass a component 14 from the reservoir 13 via a supply line 15 into the reactor 1 and the reaction mixture 2. The opening around the supply line 15 is sealed in a sterile manner by a septum 23.

[0099] FIG. 5 shows an embodiment of the device for carrying out the method of the invention for culture flasks (T-flasks) under rocker or orbital shaking. In this case, the reactor 1 is formed by a T-flask which is filled with a reaction mixture 2 and which is fixed to the housing 16 by means of a clamping device 22. Analogously to FIG. 4, in the housing 16 there is a light source 4 along with a control unit 7, two opposite light sensors 5/6 having an evaluation and control unit 8, a processor 9 connected to the control units 7/8, and an electronic metering system 12, which is also connected to the processor 9 and, under the control thereof, can pass a component 14 from the reservoir 13 via a supply line 15 which penetrates through the septum 23 into the reactor 1 and the reaction mixture 2. The processor 9, which in an advantageous embodiment of the invention is formed by a microprocessor, is connected to a storage battery 26 to provide electrical power. This also applies to all of the other electronic components of the device; however, for reasons of clarity, the corresponding connections are not shown. Further, a temperature control system 25, attached below the reactor 1 and consisting for example of a temperature sensor, a Peltier element and a control unit, is connected to the processor 9.

[0100] The embodiments of the invention shown in FIG. 4 and FIG. 5 are non-modular and are functional individually in combination with a processor 10 comprising user software 31. However, it may be advantageous to modularise the device. In this regard, FIG. 6 is a schematic drawing of a non-modular embodiment of the device, and FIG. 7 is a schematic drawing of a modular embodiment of the device, so as to describe in more detail the interaction and modularity of the individual device components. The schematically shown electronic device components form functional units, which need not necessarily be in the form of individual electronic or other technical components. In some cases, a plurality of functional units can be combined in one electronic component (for example on a chip).

[0101] The non-modular embodiment shown in FIG. 6 consists of two primary components, a measuring station 40 and a processor 10 comprising user software 31, the measuring station further comprising, in addition to the functional units (4/5/6/7/8/24) relevant for measurements and the units (12/25) which affect the reaction mixture 2, a processor 9 comprising firmware 30 and, depending on the embodiment, various communication modules 28/29 and the electrical power supply 26/27. However, to carry out the parallelisation which is desired in most shaken reactor applications, it may be advantageous to keep each measuring station 40 which is in direct contact with a reactor 1 as small as possible. The modularised embodiment in FIG. 7 makes this possible by transferring some functional units out of the measuring stations 40 into a base station 41, which can be jointly used by a plurality of measuring stations 40, and by restricting each measuring station 40 to the functional units absolutely required at the reactor 1.

[0102] As is shown in FIG. 7, in particular the processor 9 comprising firmware 30 and the electrical power supply can be integrated into a base station 41 jointly used by a plurality of measuring stations 40, by way of a storage battery 26 or a mains connection 27 and the communication modules 28/29 potentially required for communication with the processor 10 and the user software 31. Each measuring station 40 is connected to the base station 41 to provide power and for communication, the communication between the processor 9 comprising firmware 30 and the functional units of the measuring station (including in the non-modular construction) taking place using serial communication protocols and standards (for example SPI, I.sup.2C, USB, CAN, Ethernet, IEEE 802 standards etc.) in an advantageous embodiment of the invention.

[0103] The following statements apply both to non-modular (for example FIG. 6) and to modular (for example FIG. 7) embodiments of the invention.

[0104] The functional units communicating with the processor 9 and the firmware 30 are, in particular but not exclusively, the control unit 7 for at least one light source 4 and the control and evaluation unit 8 for at least one light sensor 5/6, it also being possible for a measuring station 40 to contain a plurality of control/evaluation units 7/8 and a plurality of light sources 4 and light sensors 5/6. The primary components of the control unit 7 for at least one light source 4 may, in particular but not exclusively, be digital potentiometers, diode drivers, laser drivers and microprocessors, which can be used to modify and set properties such as the intensity or spectral ranges of the light 17 emitted by the light source 4.

[0105] The purpose of the control and evaluation unit 8 for at least one light sensor 5/6 is to digitise the analogue measurement signals generated by the light sensor 5/6, to amplify and filter the analogue measurement signals generated by the light sensor 5/6 prior to digitisation, and to pass on the digitised measurement data to the processor 9 and the firmware 30. Primary components of the control and evaluation unit 8 for at least one light sensor 5/6 may be, in particular but not exclusively, digital potentiometers, operational amplifiers, analogue-digital converters, suitable frequency filters and microprocessors.

[0106] In some embodiments, light sources 4 and light sensors 5/6 may be displaceable individually or jointly or jointly with the associated control/evaluation unit 7/8, and this can be carried out, in particular but not exclusively, by electric motors controlled by the processor 9 comprising firmware 30 or other electrical magnet systems. The purpose of the displaceability is to adapt the positions of light sources 4 and light sensors 5/6 so as to adapt the device to the relevant process and measurement requirements in an automated manner.

[0107] Further, if required, at least one temperature control system 25 and at least one metering system 12 may be integrated into the measuring station, which are both controlled by way of the processor 9 comprising firmware 30 so as to intervene in the process with feedback to the measurement data and the further parameters and properties of the process determined by the processor 10 and user software 31. Primary components of the temperature control system 25 may be, in particular but not exclusively, temperature sensors, Peltier elements, analogue-digital converters and microprocessors. Primary components of the metering system 12 may be, in particular but not exclusively, electrically controllable valves and pumps, temperature sensors, liquid sensors, pressure sensors, Peltier elements, analogue-digital converters and microprocessors.

[0108] Both in at least one measuring station 40 and in at least one base station 41, at least one ambient light sensor 24 comprising a corresponding control and evaluation unit 8 may be integrated so as to detect effects on the optical measurements on the reactor(s) 1 and the reaction mixture(s) 2 and be able to eliminate or reduce corresponding interferences in the measurement data.

[0109] In an advantageous embodiment of the invention, the communication between the processor 10 and the user software 31 and the processor 9 comprising firmware 30 also takes place using serial communication protocols and standards (for example SPI, I.sup.2C, USB, CAN, Ethernet, IEEE 802 standards etc.) Any communication modules 28/29 required for this purpose are located on the processor 10 and on the processor 9 of the measuring station 40 or base station 41, depending on the embodiment.

[0110] In an advantageous embodiment of the invention, the hardware 32 of the processor 10 is much more powerful than the processor 9, in such a way that much more complex and more resource-intensive mathematical methods and computing algorithms can be implemented in the user software 31 than in the firmware 30, which is primarily for controlling the electronic components on the measuring station and/or base station 40/41.

[0111] FIG. 8 illustrates the use of sensor arrays as light sensors 5/6 using the example of a flask under orbital shaking as a reactor 1. The light 17 shone into the reactor 1 and the reaction mixture 2 by a light source 4 is scattered on at least one component of the reaction mixture 2. The scattering is continuous over a particular scattering angle range, but usually in all spatial directions. The intensity of the scattered light 18 is dependent on the scattering angle. This dependency can be exploited so as to take very exact and reliable measurements in a wide range, covering several orders of magnitude, of the optical density of the reaction mixture 2. For this purpose, many individual light sensors 5 are combined into a light sensor array 33, which simultaneously detects the scattered light intensity at various scattering angles. The comparison and the correlation of the angle-dependent scattered light intensities recorded at different times ensure reliable measurements over several orders of magnitude of the optical density. This reliability can be achieved, in particular but not exclusively, by targeted selection of the optimum light sensor 5 in the light sensor array 33 in each case and by comparing and weighting the values of all of the light sensors 5 in the light sensor array 33 by suitable mathematical methods. Light sensor arrays 33 of this type may also be used as transmission sensors, in particular but not exclusively so as to detect the transmission in different spatial directions simultaneously.

[0112] In particular but not exclusively, integrated one-dimensional and two-dimensional light sensor arrays, such as CCD chips or CMOS APS chips, and other one-dimensional and two-dimensional arrays of photodiodes, photoresistors and phototransistors may be used as a light sensor array 33.

[0113] FIGS. 10-14 show example measurement results which were determined by the method according to the invention. The measurements were taken using a device according to FIG. 2 in Erlenmeyer flasks, under orbital shaking at 160 revolutions per minute, as a reactor 1. Suspensions of baker's yeast cells (S. cerevisiae) in LB medium were used as a reaction mixture 2, relative flask fill volumes of 20% being used for the measurements of FIGS. 10-12 and relative flask fill volumes of 10% being used for the measurements of FIGS. 13-14. An LED having a peak wavelength of 528 nm was used as a light source 4, and was placed non-centrally below the flask (similarly to FIG. 2). Photodiodes were used as light sensors 5/6, a photodiode being placed below the flask next to the light source 4 as a light sensor 5 for scattered light and a photodiode being placed in the cover 3 of the flask as a light sensor 6 for transmitted light in each case. The light intensity I/.sub.0 before traversing the reactor and the reaction mixture was not measured. The formulae and equations cited in the following, in particular in reference to FIGS. 10-14, for the relationships between biomass concentration and signal intensities or other results of optical measurements apply in an identical or modified form to many other components of possible reaction mixtures 2 which interact with shone-in light 17, in particular but not exclusively to concentrations of organisms, proteins, nucleic acids (for example DNA, RNA), lipids, sugars, biopolymers, plastics materials, and other organic and/or inorganic particles, molecules, ions and, generally, substances.

[0114] FIG. 10 illustrates the correlation between the intensity of the signal of the light sensors 5/6 and the biomass concentration in the reaction mixture 2. In each case, a signal intensity value represents the average of a series of transmission or scattering measurements 34/20 or 34/21, as shown by way of example in FIG. 11 for the signal intensities from FIG. 10 at biomass concentrations of 0.33 g/L, 0.99 g/L and 2.04 g/L. As a result of the detection according to the invention of the light by the light sensors at a frequency such that the shaking frequency is not an integer multiple of the detection frequency, according to the invention the periodically fluctuating fill level results in a sensor signal which fluctuates periodically as a result of the shaking. For the example measurement shown in FIGS. 10-13, a measurement frequency of 200 Hz was selected, in such a way that a shaking period is represented by 75 measurement points.

[0115] A clear relationship between the signal intensity determined as the average of a series of measurements and the biomass concentration of the reaction mixture can be seen from FIG. 10, both for the transmission values and for the scattering values. In the measurement range shown, the represented signal intensities of the transmission and scattering light are each in an exponential relationship with the biomass concentration, as is also generally known for the absorption of light in homogeneous diluted solutions. The corresponding Beer-Lambert law for the transmission of light is

[00003]$E={\log }_{10}\left(\frac{I_0}{I}\right)=\mathrm{.Math.}.Math.c.Math.d$

[0116] where E is the optical density, I.sub.0 is the light intensity before traversing the medium, I is the light intensity after traversing the medium, E is the extinction coefficient, c is the concentration of the absorbing component, and d is the distance traversed by the light, which in this case is formed by the (optionally weighted) average of all of the light distances as a result of the periodic light distance fluctuations according to the invention.

[0117] The linear relationship between the optical density and the absorbing component of the reaction mixture also applies to the logarithm of the signal intensities and the biomass, as can be seen from FIG. 12, since it can be assumed that I.sub.0 is constant for all measurements within an experiment. If I.sub.0 is unknown, not only the change in the optical density and the biomass concentration can be determined from the changes in the signal intensity, but also the absolute biomass concentration, if the biomass concentration for one signal intensity value is known. For transmission signals, this results from the difference in the optical densities and the rearrangement of the Beer-Lambert law.

[00004]$\Delta .Math..Math.E=E_2-E_1={\log }_{10}\left(\frac{I_1}{I_2}\right)=\mathrm{.Math.}.Math.d.Math.\left(c_2-c_1\right)$$c_2=\frac{{\log }_{10}\left(\frac{I_1}{I_2}\right)}{\mathrm{.Math.}.Math.d}+c_1$

[0118] To determine absolute biomass concentrations in this manner, ε the extinction coefficient and d the distance traversed by the light must be known, and can be determined from calibration series. If ε and d are unknown, absolute biomass concentrations can still be determined if the biomass concentration for one signal intensity value is known and if I.sub.0 is known (in particular but not exclusively in that I.sub.0 is determined during the experiment, or in that required I.sub.0 values are determined by the product manufacturer and stored in the appliance). This results from the change in the relative biomass concentration c.sub.2/c.sub.1.

[00005]$\frac{c_2}{c_1}=\frac{\frac{{\log }_{10}\left(\frac{I_1}{I_2}\right)}{\mathrm{.Math.}.Math.d}+c_1}{c_1}=\frac{{\log }_{10}\left(\frac{I_1}{I_2}\right)}{\mathrm{.Math.}.Math.d}.Math.\frac{1}{c_1}+1=\frac{{\log }_{10}\left(\frac{I_1}{I_2}\right)}{\mathrm{.Math.}.Math.d}.Math.\frac{\mathrm{.Math.}.Math.d}{{\log }_{10}\left(\frac{I_0}{I_1}\right)}+1=\frac{{\log }_{10}\left(\frac{I_1}{I_2}\right)}{{\log }_{10}\left(\frac{I_0}{I_1}\right)}+1$$.Math.c_2=\left(\frac{{\log }_{10}\left(\frac{I_1}{I_2}\right)}{{\log }_{10}\left(\frac{I_0}{I_1}\right)}+1\right).Math.c_1$

[0119] The Beer-Lambert law cannot be used to determine the biomass concentration from scattering signals. However, as can be seen from FIGS. 10/12, the dependency between the scattering signal and the biomass concentration can also be modelled exponentially in the shown measurement range, in such a way that, analogously to the above equations, the scattering intensity Is can be expressed as

I.sub.s=I.sub.0.Math.10.sup.d.Math.c.Math.V

[0120] where I.sub.0 is the light intensity before traversing the medium, σ is the scattering coefficient, c is the concentration of the scattering component, and V is the reaction volume generating detectable scattered light, which in this case, because of the periodic light distance fluctuations according to the invention, is taken as the (optionally weighted) average of all values of V. As a result of multiple scattering, the light detected by a scattered light sensor 5 does not necessarily originate from the entire volume traversed by the light. Rather, the reaction volume V generating detectable scattered light is dependent on the relevant biomass concentration, in particular at higher biomass concentrations, in such a way that overall, for corresponding measurement conditions, there is a linearly approximate relationship between the scattering intensity I and the biomass concentration, which can be expressed as

I.sub.s=I.sub.0.Math.σ.sub.lin.Math.c

[0121] where I.sub.0 is the light intensity before traversing the medium, σ.sub.lin is the linear scattering coefficient, and c is the concentration of the scattering component. At low biomass concentrations, analogously to the transmission measurement, the biomass concentration is

[00006]$c_2=\frac{{\log }_{10}\left(\frac{I_{s.Math..Math.2}}{I_{s.Math..Math.1}}\right)}{\sigma .Math.V}+c_1$$\mathrm{or}$$c_2=\left(\frac{{\log }_{10}\left(\frac{I_{s.Math..Math.2}}{I_{s.Math..Math.1}}\right)}{{\log }_{10}\left(\frac{I_{s.Math..Math.1}}{I_0}\right)}+1\right).Math.c_1$

[0122] For higher biomass concentrations, the biomass concentration can be approximated as

[00007]$c=\frac{I_s}{I_0.Math.{\sigma }_{\mathrm{lin}}}$$c_2=\frac{I_{s.Math..Math.2}}{I_{s.Math..Math.1}}.Math.c_1$

[0123] For correct evaluation of the series of measurements 34 which fluctuate periodically according to the invention, further mathematical methods are available in addition to the method of averaging applied above, in particular but not exclusively direct fitting of periodic model functions to the series of measurements 34 or statistical evaluations of each series of measurement 34 using classifications, fitting of distribution functions, correlation analyses between different measurement series of a light sensor 5/6 and/or different light sensors 5/6 etc. A further mathematical method, which is suitable in particular for robust evaluation of the series of measurements at low biomass concentrations, is discrete folding of two series of measurements determined at different moments, which in each case results in a discrete folding function F for the folded pair of measurement series. This folding function can now be mathematically evaluated further and correlated with the biomass. One possible type of evaluation involves taking the average of all of the elements of a discrete folding function. For transmission measurements, FIG. 13 shows the correlation between the base-ten logarithm of the quotients of the folding averages and the biomass concentration. Analogously to the above statements on the evaluation of the series of transmission measurements, this relationship is in accordance with the equation

[00008]$\Delta .Math..Math.E_F={\log }_{10}\left(\frac{F_1}{F_2}\right)={\mathrm{.Math.}}_F.Math.d.Math.\left(c_2-c_1\right)$

[0124] where ΔE.sub.F is the change in the folding-specific optical density, F.sub.1 and F.sub.2 are the averages of the folding functions F.sub.1 and F.sub.2, ε.sub.F is the folding-specific extinction coefficient, c is the concentration of the absorbing component, and d is the distance traversed by the light. If there is at least one known concentration value, this results in a biomass concentration of

[00009]$c_2=\frac{{\log }_{10}\left(\frac{F_1}{F_2}\right)}{{\mathrm{.Math.}}_F.Math.d}+c_1$

[0125] FIG. 13 further shows that the invention makes robust measurements possible both in the exclusion of and in the presence of ambient light (in particular daylight and room lighting) even in low biomass concentration ranges. The measurement data under ambient light were recorded in daylight and additional room lighting, whilst the measurements without ambient light were taken by darkening the entire measurement construction using a black, opaque fabric. Both measurements show a linear relationship between biomass concentration and the base-ten logarithm of the folding average quotients even at low biomass concentrations (the linear relationship also applies to the signal averages which are not shown). Only the gradient of the linear functions is different; under ambient light, the processed signal increases more slowly, since the level of the periodic fluctuations according to the invention in the transmission signal 20 turns out lower than for the darkened measurement as a result of the higher background light intensity. However, the gradient can be appropriately scaled to the ambient light conditions by using at least one ambient light sensor 24 which is read in parallel with the measurement on the reactor 1 and reaction mixture 2, in such a way that robust measurements are possible even in varying ambient light.

[0126] FIG. 14 shows the relationship between biomass concentration and a modified base-ten logarithm of the folding average quotients of series of transmission measurements over a wide range of concentrations up to more than 40 g/l. Results are shown for two measurements on unbaffled flasks, one measurement each with and without ambient light, and a measurement on a baffled flask without ambient light. The measurement frequency on the unbaffled flask was 200 Hz, whereas series of measurements for the baffled flask were recorded at 450 Hz so as to be better able to detect the turbulent distribution of the reaction mixture 2 in the reactor 1.

[0127] Over the biomass concentration range shown in FIG. 14, the relationship between the transmission intensity and the biomass concentration can no longer be modelled using a simple exponential function, and so the Beer-Lambert law is no longer valid, and has to be modified. This relationship can be optimally described using the sum of at least two exponential functions having different coefficients in the exponent, for example

I=I.sub.0.Math.(a.Math.10.sup.−ε.sup.1.sup..Math.c.Math.d+b.Math.10.sup.−ε.sup.2.sup..Math.c.Math.d)

[0128] Since functions of this type usually cannot be solved analytically for the concentration c, the biomass concentration has to be determined using an analytically solvable approximation function which sufficiently accurately represents the progression of the sum of a plurality of exponential functions. For this purpose, the function on which the Beer-Lambert law is based can for example be modified to

I=I.sub.0.Math.10.sup.−ε.Math.c.sup.z.sup..Math.d

[0129] resulting in a biomass concentration c of

[00010]$c=\sqrt[z]{\frac{{\log }_{10}\left(\frac{I_0}{I}\right)}{{\mathrm{.Math.}}_z.Math.d}}$

[0130] where I.sub.0 is the light intensity of the medium, I is the light intensity after traversing the medium, ε.sub.z is the corrected extinction coefficient, z is the correction exponent and d is the distance traversed by the light. Analogously to the above equations for the transmission, the biomass concentration can subsequently also be calculated as

[00011]$c_2=\sqrt[z]{\frac{{\log }_{10}\left(\frac{I_1}{I_2}\right)}{{\mathrm{.Math.}}_z.Math.d}+c_1^z}$

[0131] As was shown for the above example measurement, the folding averages can also be used for the intensities and the corrected folding-specific extinction coefficients can also be used for the corrected extinction coefficients. The application of a Beer-Lambert law modified in this manner makes it possible to determine the biomass concentration over a wide measurement range exclusively by using transmission data. By way of a suitable combination of the measurement data of a plurality of light sensors 5/6, the robustness and exactness of the measurements can additionally be improved.

[0132] FIG. 14 additionally demonstrates how the invention can be used on continuously shaken baffled systems, which are characterised by turbulent flows, inhomogeneous reaction mixture distributions in the reactor, and disrupting factors such as bubbles and foam. The technologies known in the art cannot take robust measurements on systems of this type during shaking operation. This is only possible by recording, according to the invention, a multiplicity of measurement values, at a frequency such that the shaking frequency is not an integer multiple of the detection frequency.

LIST OF REFERENCE NUMERALS

[0133] 1 Reactor

[0134] 2 Reaction mixture

[0135] 3 Cover

[0136] 4 Light source

[0137] 5 Light sensor for scattered light

[0138] 6 Light sensor for transmitted and/or forward-scattered light

[0139] 7 Control unit for at least one light source 4

[0140] 8 Evaluation and control unit for at least one light sensor 5/6

[0141] 9 Processor (for example comprising central control unit as a microcontroller or SoC)

[0142] 10 Processor (for example in the form of a PC or server)

[0143] 11 Shaking axis orthogonal to the shaking plane in orbital shaking

[0144] 12 Metering system

[0145] 13 Reservoir

[0146] 14 Component to be added

[0147] 15 Supply line

[0148] 16 Housing

[0149] 17 Beam of light emitted by light source

[0150] 18 Beam of backscattered light

[0151] 19 Beam of transmitted and/or forward-scattered light

[0152] 20 Signal of 6 (“transmission signal”)

[0153] 21 Signal of 5 (“scattering signal”)

[0154] 22 Clamping device

[0155] 23 Septum

[0156] 24 Ambient light sensor

[0157] 25 Temperature control system

[0158] 26 Storage battery

[0159] 27 Mains-connected power supply

[0160] 28 Radio module

[0161] 29 Wired communication module

[0162] 30 Firmware

[0163] 31 User software comprising graphical interface

[0164] 32 Processor hardware

[0165] 33 Light sensor array

[0166] 34 Series of measurements

[0167] 35 Time series

[0168] 36 Mathematical and/or computing methods/algorithms

[0169] 37 Parameters and properties of the process

[0170] 38 Optics for modifying the light of light source 4

[0171] 39 Optics for modifying the light from the reaction mixture 2 and reactor 1

[0172] 40 Measuring station

[0173] 41 Base station