A method for estimating dissolved organic matter in liquids

Abstract

The disclosure regards a computer implemented method for estimating the reactivity of dissolved organic matter in a liquid comprising the steps of obtaining a first fluorescence measurement from the liquid, wherein the first fluorescence measurement comprises a first emission wavelength and a first excitation wavelength, obtaining a second fluorescence measurement from the liquid, wherein the second fluorescence measurement comprises a second emission wavelength and a second excitation wavelength, determining a sample photosensitivity index based on the first fluorescence measurement, the second fluorescence measurement and a conversion factor specific to said liquid, and estimating the reactivity of dissolved organic matter in the liquid based on the sample photosensitivity index, wherein the conversion factor is a correlation between a series of base photosensitivity indexes and a series of fluorescence measurement ratios, obtained from a liquid of a same type as the liquid. The present disclosure further relates to a computer implemented method for calibrating a fluorescence setup in relation to a liquid, and to an optical measurement system for estimating the reactivity of dissolved organic matter in a liquid.

Description

[0001] A method for estimating dissolved organic matter in liquids

[0002] The present disclosure relates to method and a system for detecting and / or determining the reactivity of dissolved organic matter (DOM) in a liquid sample. The present disclosure further relates to a method for calibrating such a system in relation to a liquid.

[0003] Background

[0004] Dissolved organic matter (DOM) is found in most liquid samples, and is found everywhere in aquatic ecosystems, such as oceans and fresh water resources such as rivers and lakes. DOM is also found in water treatment systems and in most beverages manufactured from natural products, such as wine and beer.

[0005] The amount and composition of DOM can be informative about the processes in an ecosystem and be indicative of the quality of drinking water and other beverages. Thus, DOM is identified in many liquids to determine a plurality of characteristics of processes.

[0006] The composition of DOM determines if specific molecules or molecular assemblies can participate in chemical reactions. The term reactivity can be described as the tendency of an individual molecule or group of molecules within DOM to undergo changes.

[0007] Certain functional groups, such as aromatic structures, will be highly reactive toward one type of reaction (such as photooxidation) but far less reactive with regard to another type of reaction. The reactivity of its constituents in combination with the relative abundance of the constituent will then determine the impact of reactivity on the amount of DOM. Exposure to different reactions then leads to changes in amounts and compositions through the reactivity of the material.

[0008] It is generally beneficial to determine the character of DOM in a solution since the chemical composition determines e.g. how DOM interacts with treatment processes, or if DOM forms harmful byproducts during degradation. One such parameter is the photosensitivity of DOM. As sunlight will reduce the DOM’s ability to fluoresce, the change in fluorescence of a liquid over time can thus be recorded and used to determine the photosensitivity thereof. However, the state of the art in solutions related to the estimation of dissolved organic matter composition in liquid samples involve time consuming and highly expensive experimental setups, where a broad excitation spectrum is being irradiated on a sample from which a fluorescence spectrum is being measured for each excitation wavelength. Then, the same measurement needs to be performed at a different point in time to estimate the impact of e.g. sunlight degradation, in order to determine the change in fluorescence of the sample, which can then be correlated to the change of the dissolved organic matter in the sample. Such an experiment may take more than 20 hours, requires costly equipment and it cannot be performed in-situ in large water bodies such as lakes or oceans due to the time scale of the experiment.

[0009] As a result, the need of fast-paced experiments is required in order to allow researchers and companies to test the quality of liquid samples, such as drinking water or seawater, on-demand, allowing them to continuously measure the quality of water and perform actions when needed.

[0010] Hence, the need for a novel technique is required which can provide rapid experimental data estimating the reactivity of dissolved organic matter and the quality of a liquid sample.

[0011] Summary

[0012] One purpose of the present disclosure is to provide a method for rapid and reliable estimation of the reactivity of dissolved organic matter in a liquid, such as sea water or drinking water, without requiring extensive irradiation of the liquid. Any kind of liquid may have a number of organic compounds or small organic molecules, which can originate either naturally from plant material or bacteria, or from pollution sources. It is known that radiation from the sun can degrade such organic matter into smaller organic compounds, forming and / or chemically altering dissolved organic matter. For example, photodegradation of extended periods of time can lead to an increased utilization of the degraded dissolved organic matter by bacteria, effectively reducing the quality of a liquid.

[0013] The present disclosure relates to a computer implemented method for estimating the reactivity of dissolved organic matter in a liquid comprising the steps: obtaining a first fluorescence measurement from the liquid, wherein the first fluorescence measurement comprises a first emission wavelength and a first excitation wavelength and obtaining a second fluorescence measurement from the liquid, wherein the second fluorescence measurement comprises a second emission wavelength and a second excitation wavelength. Then, a sample photosensitivity index can be determined based on the first fluorescence measurement and the second fluorescence measurement, as well as a conversion factor specific to said liquid. Based on the photosensitivity index, the dissolved organic matter in the liquid can be estimated. In an embodiment, the conversion factor is a correlation between a series of base photosensitivity indexes and a series of fluorescence measurement ratios, obtained from a liquid of a same type as the liquid. Further details of that correlation are provided in the following sections of the present disclosure. As a result, the present disclosure provides a method of estimating the dissolved organic matter in a liquid by simply calculating two fluorescence measurements, and correlating them to a conversion factor, without requiring long irradiation experiments. I.e. the present disclosure can predict how much a liquid will be altered due to photodegradation, thereby providing valuable information.

[0014] The presently disclosed approach is based on the novel insight that it is possible to determine a liquid specific conversion factor, in combination with wavelengths to be used in the estimation of fluorescence ratios, thereby “calibrating” the fluorescence setup used in the photosensitivity measurements. Once the setup is calibrated with respect to the specific liquid, this information can be utilized to predict a sample photosensitivity index in a new sample of the same liquid type. And once the photosensitivity index of the liquid sample is determined, the reactivity of dissolved organic matter in the liquid can be estimated. The presently disclosed approach can therefore prevent tedious photodegradation experiments and measurements of the excitation emission matrices. It is possible to estimate the amount, composition and as a result the quality of dissolved organic matter of a sample, by analyzing the reactivity of dissolved organic matter in the sample. Hence, the present disclosure enables a much faster estimation of the reactivity of dissolved organic matter in liquids, compared to the tedious measurements required without using such a conversion factor.

[0015] As described above, reactivity of DOM in a sample can be described as the tendency of an individual molecule or group of molecules within the DOM to undergo changes. Certain functional groups (e.g. aromatic structures) will be highly reactive toward one type of reaction (e.g. photooxidation) but far less (or equally or even more) reactive with regard to another type of reaction. The reactivity of its constituents in combination with the relative abundance of the constituent will then determine the impact of reactivity on the amount of DOM. Exposure to different reactions then leads to changes in amounts and compositions through the reactivity of the material. If the reactivity toward a reaction is known or can be estimated, the effectiveness of the reaction can be determined. In actual deployment, this would for example allow a water treatment plant to predict how treatment steps will affect the water quality and adjust the treatment accordingly to achieve the desired outcome. As the reactivity is linked to the amount, or composition, of DOM in a liquid, the presently disclosed approach can be used to estimate the amount, composition and / or quality, of dissolved organic matter in a sample.

[0016] For example, the first and the second fluorescence measurements can be obtained after exposing the liquid with two excitation signals of different wavelengths, which can be emitted by a light source such as an LED. A sensor (such as a photodiode) can be utilized to detect the emitted fluorescence from the liquid, and a filter can be employed to limit the detected fluorescence wavelength range to a preferred range.

[0017] The estimated DOM can be used to determine the susceptibility of dissolved organic matter to photobleaching. Alternatively or additionally, the estimated dissolved organic matter can for example be used to further determine the quality of dissolved organic matter.

[0018] The photosensitivity index provides information of how much has the liquid already been exposed to photodegradation, where the photodegradation changes the composition of the dissolved organic matter in the liquid. The compositional changes occur first in the compounds that absorb light (chromophores), since the direct effects of photodegradation necessitate the absorption of light. The susceptibility to direct photodegradation is therefore determined by the quantity and wavelength of the photon-source (sunlight) and the ability of a compound to absorb these photons at the emitted wavelengths. This process bleaches the dissolved organic matter, i.e. permanently decreases the ability to absorb light at the affected wavelengths. The present disclosure is based on the fact that a sample’s susceptibility to direct photodegradation decreases predictably and that the composition of chromophores can be used to make said predictions. For example, if a sample has high photosensitivity, that would mean that a substantial amount of dissolved organic matter is susceptible to future photodegradation. Whereas if a sample has low photosensitivity, its chromophores are less likely to degrade in the future. The benefits of knowing a sample’s photosensitivity especially comes from comparative measurements (in space and I or time). Such comparisons inform users about qualitative changes or differences in samples and facilitate application-dependent conclusions. While the photosensitivity is derived from the ratio of two fluorescence measurements with the help of a conversion process, the fluorescence intensities themselves also provide an abundance estimate of the dissolved organic matter at two spectrally distinct wavelengths. This simultaneously allows quantitative insights into the dissolved organic matter.

[0019] Determining the sample photosensitivity index based on the first fluorescence measurement and the second fluorescence measurement may comprise determining the ratio between the first fluorescent measurement and the second fluorescent measurement. This is typically done by dividing the two measurements. Determining the photosensitivity index is also based on a conversion factor. The conversion factor may in one embodiment comprise a conversion slope which can be a number that is multiplied to the two fluorescence measurements, typically the ratio of the two fluorescence measurements. In an additional or alternative embodiment the conversion factor further comprises a conversion offset which may be a constant value that is added to the final product.

[0020] The conversion factor can be extracted by determining the base photosensitivity index of a number of samples or of one sample at different time instances, using standard means, for instance by irradiating with a whole spectrum of radiation and measuring the fluorescence in a range of wavelengths, and performing such a measurement at a plurality of instances in time to measure the base photosensitivity index of the sample. Then, each of the base photosensitivity indexes is compared to a ratio of a first and a second fluorescence measurement, at a time instance. The conversion factor can be finally determined by fitting the correlation of the base photosensitivity indexes to the ratios, extracting a conversion slope and conversion offset such as an y-axis intercept. Further details and examples about the conversion factor is provided in the detailed description. The present disclosure further relates to a computer implemented method for calibrating a fluorescence setup in relation to a liquid for estimating the reactivity of dissolved organic matter in said liquid, the method comprising, obtaining a plurality of datasets comprising spectra of fluorescent radiation over a plurality of time instances of the liquid, and determining a number of base photosensitivity indexes for the liquid based on the spectra of fluorescent radiation, wherein each base photosensitivity index is determined over at least two time instances, namely before and after an exposure to light. For example, the spectra of fluorescent radiation can be excitation emission matrixes (100) as shown in Fig. 1 , which are measured at the “before” and “after” time points. Then, by estimating the average change in fluorescence between all the points of the excitation emission matrix relative to the same measurements before the exposure, it is possible to determine the base photosensitivity index for each matrix.

[0021] Additionally, the method may comprise correlating the fluorescence loss of the liquid with respect to an excitation wavelength range and an emission wavelength range, and selecting a plurality of sets of first and second fluorescence measurements based on the correlation, each set corresponding to a distinct time instance, such as “before” and “after” time points, wherein each set of first and second fluorescence measurements comprises a first and a second emission wavelength, and a first and a second excitation wavelength, such that the plurality of sets of fluorescence measurements provide a maximum correlation between photosensitivity and fluorescence prior to an exposure. For example, by estimating a fluorescence coefficient in a 2D map of excitation versus emission spectra, it is possible to pinpoint areas of enhanced fluorescence loss after a certain photon dose. Such areas can be used to select fluorescence measurements for the estimation of fluorescence ratios.

[0022] Moreover, the method comprises the steps of determining a ratio for each set of the first and the second fluorescence measurements on each time instance, and determining a conversion factor by correlating the base photosensitivity indexes to the ratios, thereby calibrating the fluorescence setup by obtaining a liquid specific conversion factor to determine the sample photosensitivity index.

[0023] The present disclosure further relates to an optical measurement system for estimating the reactivity of dissolved organic matter in a liquid by calculating a photosensitivity index, the system comprising: at least a first light source configured to emit to the liquid a first excitation signal and a second excitation signal, a sensor configured to detect a first fluorescence signal and a second fluorescence signal from the liquid, and a processor configured to extract a first fluorescence measurement from the first fluorescence signal, and a second fluorescence measurement from the second fluorescence signal, and configured to determine the photosensitivity index based on the first fluorescence measurement, the second fluorescence measurement and a conversion factor based on the paragraph above, to estimate the reactivity of dissolved organic matter in the liquid.

[0024] The system can integrate a light source such as an LED that can emit a radiation of a preferred wavelength range to the liquid, and it can also comprise a sensor together with an optical filter, which can limit the detected fluorescence to a preferred wavelength range. A processor can be a microprocessor that can collect the fluorescence signals and convert them to fluorescence measurements, and the processor can analyze the data or transfer them to a second computer configured to analyze the fluorescence measurements.

[0025] As can be understood, the system disclosed herein provides a much simpler measurement system and method as known previously. By providing a computer implemented method as disclosed herein, it is possible to determine a first and a second fluorescence measurement which can be done at specific wavelength, which are distinct from each other at one, at one measurement time. Thus, a system may be provided which comprises a much smaller number of light sources and where the system can be made portable and independent of time-consuming measurements. In addition, the present disclosure relates to the use of a system for determining a photosensitivity index and / or the reactivity of dissolved organic matter of a liquid as described above, wherein at least a part of the system is submerged into the liquid, such as the first light source.

[0026] Thus, for example, instead of taking a water sample, e.g. from a body of water, and bringing it back to laboratory for time extensive analysis, it is possible to submerge the system as disclosed herein directly into the body of water and estimate the reactivity of dissolved organic matter in the body of water within seconds to minutes.

[0027] Such a system may also be installed directly at a location at a low cost. Such location may for example be at a water treatment plant where the instant estimation of the dissolved organic matter can be done and processes can be adjusted or actions taken according to specified rules.

[0028] Description of Drawings

[0029] Various embodiments are described hereinafter with reference to the drawings. The drawings are examples of embodiments and are intended to illustrate some of the features of the presently disclosed method for estimating dissolved organic matter in liquids and are not limiting to the presently disclosed system and method.

[0030] Fig. 1 illustrates an excitation emission matrix showing the fluorescence intensity as a function of excitation wavelengths and emission wavelengths for a liquid.

[0031] Fig. 2 shows a diagram illustrating the calibration of the detected fluorescence signals. Fig. 3 A, B shows the process of estimating a base photosensitivity index of a liquid. Fig. 4 shows the average dissolved organic matter fluorescence of a training dataset compared to the regression contribution of the same samples.

[0032] Fig. 5 A shows fitting of base photosensitivity index against fluorescence ratios, determining a conversion factor. Fig. 5 B shows a validation process for evaluating the error of the determined conversion factor.

[0033] Detailed description

[0034] As described in the summary, the present disclosure relates to a computer implemented method for estimating the reactivity of dissolved organic matter in a liquid comprising the steps: obtaining a first fluorescence measurement from the liquid, wherein the first fluorescence measurement comprises a first emission wavelength and a first excitation wavelength, obtaining a second fluorescence measurement from the liquid, wherein the second fluorescence measurement comprises a second emission wavelength and a second excitation wavelength, determining a sample photosensitivity index based on the first fluorescence measurement, the second fluorescence measurement and a conversion factor specific to said liquid and estimating the reactivity of dissolved organic matter in the liquid based on the sample photosensitivity index. In an embodiment, the conversion factor is a correlation between a series of base photosensitivity indexes and a series of fluorescence measurement ratios, obtained from a liquid of a same type as the liquid. For example, figure 1 shows an excitation emission matrix (100) showing on the y-axis the fluorescence emitted from the liquid (102) in response to excitation with light in the wavelength range shown on the x-axis (101). Each combination of excitation wavelength and fluorescence wavelength emitted from the liquid corresponds to a fluorescence intensity (104). The present disclosure does not require to conduct such measurements, as it is sufficient to only detect the fluorescence of two points (103) provided by two different excitations signals. The fluorescence intensity obtained by these two points are hereby called first fluorescence measurement and second fluorescence measurement. For example, the first fluorescence measurement is obtained from a fluorescence emission of 375 nm, caused by a first excitation signal of 300 nm, and the second fluorescence measurement is obtained from a fluorescence emission of 425 nm, caused by a second excitation signal of 350 nm. The process how the two fluorescence measurements are used in determining the photosensitivity index is described in the later sections of the description, where the conversion factor is introduced in more detail.

[0035] Fluorescence measurements and signal detection

[0036] In an embodiment, the step of obtaining a first fluorescence measurement may comprise the extraction of the first fluorescence measurement at a first fluorescence wavelength range, and the step of obtaining a second fluorescence measurement may comprise the extraction of the second fluorescence measurement at a second fluorescence wavelength range. For example, if the target for the first fluorescence measurement is to obtain the fluorescence intensity of a liquid at 350 nm, then the instrument that can detect that wavelength can detect measurements from a wavelength range around that value. For instance, the instrument can detect the fluorescence intensity from 345 to 355 nm. The wavelength range is determined by the technology of the instrument that detects the fluorescence. Moreover, the first fluorescence measurement can be derived from a first fluorescence signal, and the second fluorescence measurement can be derived from a second fluorescence signal. The term fluorescence signal highlights the experimental nature of the embodiment, as a fluorescence signal emitted by the liquid is detected by an instrument.

[0037] In another embodiment, the step of obtaining the first fluorescence measurement can comprise at least one sensor configured to obtain the first fluorescence measurement from a first fluorescence wavelength range, and the step of obtaining the second fluorescence measurement can comprise at least one sensor configured to obtain the second fluorescence measurement from a second fluorescence wavelength range. For example, a standard sensor configured to detect fluorescence signals can be employed, that can detect a certain range of wavelengths, and a fluorescence measurement can be estimated from the detected signal. Typical fluorescence sensors can be a fluorometer device that may detect signals from ultraviolet to near infrared wavelengths.

[0038] Moreover, it may be advantageous to limit the detected signal in order to collect more accurate fluorescence measurements. The at least one sensor described above, may comprise an optical filter, configured to limit the obtained fluorescence measurement to a preferred fluorescence wavelength range. For example, a sensor may be configured to detect signals from a broad range of wavelengths, such as 300 nm to 450 nm. The goal of an experiment might be to detect the fluorescence of a water sample at 350 nm and 400 nm. Therefore, a user can attach an optical sensor on a first sensor that can limit the detected wavelength to a narrower window, such as 340 nm to 360 nm. By doing that, the sensor will only be sensitive to detect signals having 340-360 nm wavelengths, providing a more accurate result. Another optical sensor can be attached on a second sensor, limiting the range of detected wavelengths to 390-410 nm. For example, the optical sensor can be a bandpass filter, allowing a specific range of wavelengths to pass while blocking other wavelengths. In another embodiment, the optical sensor can be a longpass filter, configured to transmit signals with wavelengths larger than a certain cutoff wavelength, or the optical sensor can be a shortpass filter, configured to transmit signals with wavelengths shorter than a certain wavelength.

[0039] Depending on the type of liquid that is tested, it can be beneficial to measure the fluorescence at different wavelengths. The first fluorescence measurement and the second fluorescence measurement can be extracted at distinct fluorescence wavelength ranges. In other cases, it can be advantageous the measure the fluorescence at the same wavelengths, excited by signals of different wavelength.

[0040] In another embodiment, the determination of the sample photosensitivity index can comprise the estimation of the ratio of the first fluorescence measurement to the second fluorescence measurement. As discussed in the summary of the application, the ratio of the two fluorescence measurements can provide insight of the photosensitivity index of a liquid, giving a prediction of the dissolved organic matter in the liquid, without requiring irradiating the sample with a whole spectrum of radiation for an extended period of time.

[0041] Furthermore, the first fluorescence measurement can be extracted from a wavelength preferably larger than 375 nm, preferably smaller than 425 nm, preferably around 400 nm. In addition, the second fluorescence measurement can be extracted from a wavelength preferably larger than 400 nm, preferably smaller than 450 nm, preferably around 425 nm. Naturally, depending on the type of the liquid that is tested, and the type of organic matter that may be present in the liquid, the preferred wavelength of fluorescence can vary from 300 nm to 600 nm. As described above, depending on the type of experiment, different sensors or optical filters can be utilized, to limit the detected fluorescence.

[0042] Depending on the type of liquid and the type of the information that a user wants to retrieve from the liquid, the precise timing of the first fluorescence measurement and the second fluorescence can be established. In an embodiment, the first fluorescence measurement and the second fluorescence measurement can be obtained within a time frame, the time frame is preferably less than 1 s. That time frame can be for example, the time difference between detecting the two fluorescence signals, or it may also be the time difference between a first excitation signal and a second excitation signal that are emitted to the liquid, with the purpose of triggering a first fluorescence signal and a second fluorescence signal from the liquid. In another embodiment, the first fluorescence measurement and the second fluorescence measurement can be obtained simultaneously. In certain scenarios, it can be beneficial to obtain the two fluorescence measurements simultaneously, as the time range where a fluorescence signal is emitted might be very small, and therefore its intensity might fade if a larger time frame is given. Moreover, if there is no interference between the two fluorescence signals, then the sensors can detect them simultaneously without compromising the accuracy of the measurements. It can also be advantageous that no photodegradation has occurred to the liquid in the time frame between the first fluorescence measurement and the second fluorescence measurement is obtained. In certain cases it can be important that no photodegradation has occurred to the liquid in between the two fluorescence measurements are obtained. In addition, a calibration process can also be performed to account for differences in the detection of fluorescence signals in the sensors. For example, a sensor configured to obtain the fluorescence signal and convert it to a fluorescence measurement functions by storing the fluorescence measurement as a voltage signal (200). As the sensor comprises of electronic components, it is natural that some calibration of the instrument is needed, so that the voltage corresponds to the actual measurement. Therefore, an ambient light voltage (201) can be subtracted by the signal voltage, such that it corrects the result. This results in accurate measurements in the presence of natural sunlight at various intensities. Moreover, a spectral correction (202) can also be performed on the final result. This ensures that varying intensities of LEDs or transparencies of filter materials or other optical components is accounted for and compensated.

[0043] Signal excitation

[0044] As described in the previous section, an aspect of the present disclosure relates to exciting the liquid with certain wavelengths, capable of inducing fluorescence in the liquid, which can in turn be detected by certain sensors. In a preferred embodiment, the liquid can be excited with a first excitation signal and a second excitation signal. The first and second excitation signals can be emitted by any light source, such as a light emission diode, a laser, or a conventional light source. The wavelength of the excitation signal emitted by light source can be from 300 nm to 700 nm. Depending on the type of the liquid, or the type of the dissolved organic matter that is wished to be tested, different wavelengths of radiation can be chosen.

[0045] In another embodiment, at least one light source can be provided, said at least one light source can be used to excite the liquid with a first excitation signal and a second excitation signal. For example, a single light source can be configured to emit two different signals of different wavelengths, and a user can choose between the two, first emitting an excitation signal to trigger the first fluorescence measurement, and then emitting the second excitation signal to trigger the second fluorescence measurement. The single light source can be for example a multi-coloured light emission diode (LED), which can work by combining signals of different wavelength. For example, a processor can be configured to control the LED, adjusting the different signals to produce a final excitation signal of the preferred wavelength. Moreover, the light source can be configured such hat it has an emission frequency preferably larger than 1 Hz. As discussed in the previous paragraphs, it can be beneficial that the time frame between the two excitation signals is low enough, to not cause significant photodegradation of the liquid while the two fluorescence measurements are obtained.

[0046] Furthermore, the first excitation signal can be emitted at a wavelength preferably larger than 275 nm, preferably smaller than 325 nm, preferably around 300 nm. The second excitation signal can be emitted at a wavelength preferably larger than 325 nm, preferably smaller than 375 nm, preferably around 350 nm. The wavelength of the excitation signals can vary, and depend on the type of the liquid, as well as the conditions that are present such as pH, salinity or temperature of the liquid. For example, an excitation signal in a water sample taken from the sea can be around 350 nm, while for a different type of liquid such as a wine, the wavelength of the excitation signal can be different, depending on the types of organic compounds present in the liquid.

[0047] In addition, a calibration process can also be performed to account for differences in the emission of the light sources.

[0048] Types of liquids

[0049] The present disclosure can determine the quality of dissolved organic matter based on the photosensitivity index in a plurality of liquid types. Specifically, the liquid can be sea water, drinking water, or a liquid sample such as a natural freshwater sample, seawater sample, wastewater sample or beverage sample such as wine. By sea water or drinking water, we refer to the concept of taking fluorescence measurements in-situ, at an ocean or at a water mass, by loading a system that can conduct the photosensitivity measurements on a ship. The advantage of conducting such measurements is that it is not required for a user to collect a liquid sample and bring it to a lab before they can experimentally determine the photosensitivity of samples. Another advantage is the spatial resolution obtained when the present disclosure is mounted in the form of a battery-powered sensor on remotely operated equipment, or autonomous underwater gliders. At sensor measurement frequencies of 1 Hz or more, this allows a spatial resolution that cannot be achieved with grab samples and laboratory based measurements.. Here, the term grab sample refers to water samples taken at a specific location (latitude, longitude, and water depth) with the aim of conducting extensive measurements. In an embodiment, the volume of the liquid can preferably be less than 50 mL, more preferably less than 25 mL, even more preferably less than 5 mL, most preferably around 1.5 mL. In a controlled lab environment, it is advantageous to have a smaller volume, as that would allow the excitation signal to interact with the whole volume, triggering a fluorescence signal corresponding to the whole liquid sample. As a result, more accurate photosensitivity index measurements can be obtained.

[0050] In another embodiment, the liquid can have transparency of preferably higher than 50%, more preferably higher than 60%, even more preferably higher than 80%, most preferably higher than 95%. If the transparency of the liquid is low, the predicted photosensitivity becomes inaccurate since the assumed penetration depth of sunlight no longer matches the measured sample. Moreover, if the transparency of a liquid is low, for example 50%, that would mean that an excitation signal towards the liquid would only interact with a part of the liquid, as it would be either absorbed or reflected before interacting with the rest of the liquid. At low transparency, part of the emitted fluorescence is also reabsorbed by the organic matter in the liquid. As a result, measurements from such a liquid cannot be accurate without further correction. Therefore, it can be important that the transparency of the liquid is most preferably higher than 95%.

[0051] Moreover, the liquid can have a pH value preferably higher than 7, more preferably higher than 7.5, and preferably lower than 9, more preferably lower than 8.5, most preferably around 8. The fluorescence emission of certain compounded can be reduced and even quenched when the pH of the liquid is lower than 7, or higher than 9. Therefore, it can be important to maintain a stable pH during fluorescence measurements of a liquid. In addition, it can be advantageous that the liquid has a temperature higher than its freezing point, and a temperature lower than its boiling point.

[0052] In another embodiment, the method can comprise an additional step of identifying the type of the liquid by correlating the fluorescence measurement(s) of the liquid to historic data, wherein the historic data comprise fluorescence measurements and excitation signals of liquids. For example, the historic data can comprise past fluorescence measurements determining the photosensitivity index of various liquid samples of different types, and it can also comprise fluorescence measurements extracted from the literature. It is possible to analyze fluorescence measurements of a liquid and compare them with fluorescence measurement and fluorescence ratios of historic data. Depending on the correlation of the fluorescence ratios, it can be possible to determine the type and the origin of the liquid sample.

[0053] Calibrating the fluorescence setup (Conversion factor)

[0054] One of the important features of the present disclosure is the selection of the emission and excitation wavelengths for obtaining the fluorescence measurements for a liquid sample, as well as the extraction of the conversion factor for the estimation of the sample photosensitivity index of a liquid. Such a conversion factor allows a user to take advantage of historic data in order to predict the photosensitivity of a liquid without exposing it to lengthy irradiation experiments.

[0055] Specifically, the present disclosure relates to a computer implemented method for calibrating a fluorescence setup in relation to a liquid for estimating the reactivity of dissolved organic matter in said liquid, the method comprising, obtaining a plurality of datasets comprising spectra of fluorescent radiation over a plurality of time instances of the liquid, determining a number of base photosensitivity indexes for the liquid based on the spectra of fluorescent radiation, wherein each base photosensitivity index is determined over at least two time instances, correlating the fluorescence loss of the liquid with respect to an excitation wavelength range and an emission wavelength range, selecting a plurality of sets of first and second fluorescence measurements based on the correlation, each set corresponding to a distinct time instance, wherein each set of first and second fluorescence measurements comprises a first and a second emission wavelength, and a first and a second excitation wavelength, such that the plurality of sets of fluorescence measurements provide a a maximum correlation between photosensitivity and fluorescence prior to an exposure, determining a ratio for each set of the first and the second fluorescence measurements on each time instance, determining a conversion factor by correlating the base photosensitivity indexes to the ratios, thereby calibrating the fluorescence setup by obtaining a liquid specific conversion factor and optimal emission and excitation wavelengths to determine the sample photosensitivity index. Analyzing the above method step by step, the determination of the base photosensitivity indexes is done by calculating an emission excitation matrix at a given time instance (300), as seen in Fig. 3 and estimating an average fluorescence at every emission-excitation combination Xbefore (301). At a second time instance, the emission excitation matrix can be calculated (302), providing an average second fluorescence for every emission-excitation combination Xafter(303). Then, a base photosensitivity index for the corresponding two time instances can be calculated as seen in the equation in Fig. 3 (304). In the context of the present disclosure, the photosensitivity index may be defined as the arithmetic mean of the fluorescence loss between excitation wavelengths e.g. between 280 and 400 nm, and emission wavelengths e.g. 380 and 550 nm after an exposure to a fixed photon dose of natural sunlight relative to the unexposed sample. The fluorescence loss is calculated by subtracting the fluorescence of the photobleached sample Xafter (303) from the unexposed sample Xbefore (301). Thus, a more pronounced fluorescence loss results in a higher photosensitivity index. Since the photosensitivity index as calculated here (304) is a ratio of two measurements, it is unitless and represents the measured or predicted fraction of fluorescence lost due to photobleaching.

[0056] For certain liquid types, choosing wavelengths in the vicinity of specific areas can be advantageous, as the fluorescence loss in these areas is maximum. Fig. 4 shows the average dissolved organic matter fluorescence of the training dataset (401) compared to the regression contribution (402) of the same samples. The regression contribution can be determined using N-way partial least squares. Lower value regression areas highlight those areas that contribute to smaller photosensitivity indices, while higher value regression areas highlight areas that contribute to wavelengths that result in higher photosensitivity indices. I.e. if a sample has more fluorescence in the lower value regression area relative to the higher value regression area, its photosensitivity will be smaller. As seen in Fig 4, the areas labelled as C and M (400), produce the highest relative change in fluorescence in a liquid, when irradiated with a specific photon dose. Therefore, such points can be selected as fluorescence measurements to be used for the ratio calculation. For example, a plurality of samples can be collected, from which two fluorescence measurements are extracted in the vicinity of the areas C, M. Using that method it is possible to acquire a number of fluorescence ratios which can be used to calibrate the system and determine the conversion factor. An additional way of collecting fluorescence ratios, is by using a liquid sample that can be irradiated for 60 seconds, and every 5 seconds, two fluorescence measurements can be obtained, in the vicinity of the areas C, M. Then, 12 sets of first and second fluorescence measurements can be collected, which can then be used to estimate the fluorescence ratios. The base photosensitivity indexes can be plotted against the fluorescence ratios, where the fluorescence ratios (500) are plotted on the x-axis and the base photosensitivity indexes (501) are plotted on the y-axis. Such a plot is shown in Fig. 5A. By fitting those data points using a linear function (502), it is possible to extract a slope and an intercept, which are the conversion factor. The physical meaning of that fit, is that by multiplying the slope to a fluorescence ratio, and adding the intercept value, it is possible to get a prediction of the photosensitivity index. The fit furthermore indicates that the fluorescence composition of a sample implies its future susceptibility to photobleaching by natural sunlight. The fluorescence fingerprint of a sample allows the prediction how much of the sample will be bleached in the future. A low value implies a low susceptibility. A sample that starts out with a high photosensitivity will be affected less when exposed to sunlight (i.e. it is bleached). This results in a lower photosensitivity; a repeated exposure would further lower the photosensitivity, leading to less photobleaching. For example, if a sample has high photosensitivity, that would mean that a substantial amount of dissolved organic matter is susceptible to future photodegradation. Whereas if a sample has low photosensitivity, its chromophores are less likely to degrade in the future.

[0057] Finally, the main objective of the above method is that after conducting the tedious emission excitation matrix measurements for a specific liquid type to determine the conversion facto it is no longer required to conduct such extensive measurements. It would be sufficient to measure a fluorescence ratio and use the conversion factor to estimate the photosensitivity index, which can be linked to the dissolved organic matter in the liquid.

[0058] Elements of the disclosed computer implemented method for calibrating a fluorescence setup can be adjusted according to any parts of the present application. For example, the types of liquid used -and their features such as transparency- for calibrating a fluorescence setup can be the same used in the process of estimating the reactivity of dissolved organic matter in a liquid. In addition, the emission and excitation wavelength ranges disclosed for the computer implemented method for estimating the reactivity of dissolved organic matter in a liquid can also be applied for the method for calibrating a fluorescence setup.

[0059] Figure 5B shows a validation process, where sample photosensitivity indexes (503) are plotted against base photosensitivity indexes (504) for the same liquid. Such a plot proves that the process of estimating the sample photosensitivity index produces reliable results for unknown samples of the same type, as the sample photosensitivity indexes are comparable to the measured values of base photosensitivity index.

[0060] The computer implemented method for calibrating a fluorescence setup in relation to a liquid can be configured, such that the method is conducted on a plurality of samples of the same liquid type. Using a plurality of different samples can be beneficial, as it can remove any systematic errors that could be created from using the same sample at different time instances. In addition, a combination of different samples measured at different time instances could also be used, in order to determine an even more robust conversion factor.

[0061] System for estimating dissolved organic matter

[0062] The present disclosure further relates to an optical measurement system for estimating the reactivity of dissolved organic matter in a liquid based on a photosensitivity index, the system comprising at least a first light source for emitting to the liquid at least a first excitation signal comprising a first excitation wavelength and at least a second excitation signal comprising a second excitation wavelength, at least one sensor for detecting from the liquid at least a first fluorescence signal comprising a first emission wavelength and at least a second fluorescence signal comprising a second emission wavelength, and a processor configured to execute the method as described in the previous paragraphs of the application.

[0063] For example, such an optical measurement system can comprise a number of LEDs, which can emit signals of customizable wavelengths, depending on the type of the liquid. In addition, such a system can have a number of sensors (such as photodiodes), configured to detect emission signals caused by the interference of the excitation signals to the liquid. Such sensors can comprise filters to optimize the detected signals. The processor can be any computing unit such as a microprocessor, which can either collect the data and determine the sample photosensitivity index of the liquid, or it may also transfer the data to another processing unit, which can analyze the data and determine the sample photosensitivity index.

[0064] In an embodiment, the system comprises a liquid volume, such as a liquid sample. For example, the system can be integrated to function with a liquid sample that is stored in a chamber of the system. In another embodiment, the system can be transported to a tank, wherein a specific liquid can be tested. The sample photosensitivity index can be used to estimate the quality of dissolved organic matter in the liquid, which in turn can be used to estimate the quality of the liquid type or the effect of sunlight radiation on the liquid. For example, if a sample has a large concentration of dissolved organic matter, then that is an indication that it has been exposed to sunlight radiation.

[0065] In another embodiment, the system comprises a temperature sensor, configured to measure the temperature of the liquid. Such a feature can be advantageous, as the temperature of the liquid may in certain cases affect the fluorescence, and therefore affect the sample photosensitivity index. Having such a sensor can be important in order to calibrate the fluorescence measurements.

[0066] The present disclosure further relates to the use of a system for determining a photosensitivity index and / or the reactivity of the dissolved organic matter of a liquid, wherein the system comprises a first light source configured to emit a first and second excitation signal, and at least a first sensor for receiving a first and a second fluorescent measurement, wherein a part of the system is submerged into the liquid. For example, the part including a set of sensors and / or radiation emitters can be submerged into the liquid, in order to enhance the amplitude of the received signal, and also to excite the liquid directly, without allowing the radiation to interact with a intermediate interface, such as air between the system and the liquid. The system can be configured according to the embodiments described in the segments above.

[0067] In an embodiment, the liquid is natural freshwater sample, seawater sample, wastewater sample or beverage sample such as wine.

Claims

Claims1. A computer implemented method for estimating the reactivity of dissolved organic matter in a liquid comprising the steps:• obtaining a first fluorescence measurement from the liquid, wherein the first fluorescence measurement comprises a first emission wavelength and a first excitation wavelength,• obtaining a second fluorescence measurement from the liquid, wherein the second fluorescence measurement comprises a second emission wavelength and a second excitation wavelength,• determining a sample photosensitivity index based on the first fluorescence measurement, the second fluorescence measurement and a conversion factor specific to said liquid, and• estimating the reactivity of dissolved organic matter in the liquid based on the sample photosensitivity index, wherein the conversion factor is a correlation between a series of base photosensitivity indexes and a series of fluorescence measurement ratios, obtained from a liquid of a same type as the liquid.

2. The method according to claim 1 , wherein the step of obtaining a first fluorescence measurement comprises the extraction of the first fluorescence measurement at a first fluorescence wavelength range, and the step of obtaining a second fluorescence measurement comprises the extraction of the second fluorescence measurement at a second fluorescence wavelength range.

3. The method according to any one of the preceding items, wherein the step of determining the sample photosensitivity index comprises estimating a ratio of the first fluorescence measurement to the second fluorescence measurement.

4. The method according to any one of the preceding claims, wherein the first fluorescence measurement is derived from a first fluorescence signal, and the second fluorescence measurement is derived from a second fluorescencesignal.

5. The method according to any of the preceding claims, wherein the first fluorescence measurement and the second fluorescence measurement are extracted at distinct fluorescence wavelength ranges.

6. The method according to any one of the preceding claims, further comprising the step of exciting the liquid with a first excitation signal and a second excitation signal.

7. The method according to any one of the preceding claims, wherein the liquid can be sea water, drinking water, or a liquid sample such as a natural freshwater sample, seawater sample, wastewater sample or beverage sample such as wine.

8. The method according to claim 6, further configured to provide at least one light source used to excite the liquid with a first excitation signal and a second excitation signal.

9. The method according to any one of the preceding items, wherein the conversion factor comprises a first value that is multiplied to the ratio of the two fluorescence measurements and a second value that is added to the product of the multiplication of the first value and the ratio of the two fluorescence measurements.

10. The method according to any one of the preceding claims, wherein the determination of the reactivity of dissolved organic matter is used to quantify the effect of sunlight radiation on the liquid.

11. The method according to any one of the preceding claims, wherein the at least one light source has an emission frequency preferably larger than 1 Hz.

12. The method according to any one of the preceding claims, wherein the step of obtaining the first fluorescence measurement comprises at least one sensor configured to obtain the first fluorescence measurement from a firstfluorescence wavelength range, and the step of obtaining the second fluorescence measurement comprises at least one sensor configured to obtain the second fluorescence measurement from a second fluorescence wavelength range.

13. The method according to any one of the preceding claims, wherein the at least one sensor comprises an optical filter, configured to limit the obtained fluorescence measurement to a preferred fluorescence wavelength range.

14. The method according to any one of the preceding claims, wherein the at least one light source is a light emission diode.

15. The method according to any one of the preceding claims, wherein the step of exciting the liquid with a first excitation signal, comprises emitting the first excitation signal at a wavelength preferably larger than 275 nm, preferably smaller than 325 nm, preferably around 300 nm.

16. The method according to any one of the preceding claims, wherein the step of exciting the liquid with a second excitation signal comprises emitting the second excitation signal at a wavelength preferably larger than 325 nm, preferably smaller than 375 nm, preferably around 350 nm.

17. The method according to any one of the preceding claims, wherein the first fluorescence measurement is extracted from a wavelength preferably larger than 375 nm, preferably smaller than 425 nm, preferably around 400 nm.

18. The method according to any one of the preceding claims, wherein the second fluorescence measurement is extracted from a wavelength preferably larger than 400 nm, preferably smaller than 450 nm, preferably around 425 nm.

19. The method according to any one of the preceding claims, wherein the volume of the liquid is preferably less than 50 mL, more preferably less than 25 mL, even more preferably less than 5 mL, most preferably around 1.5 mL.

20. The method according to any one of the preceding claims, wherein the first fluorescence measurement and the second fluorescence measurement are obtained within a time frame, the time frame is preferably less than 1 s.

21. The method according to any one of the preceding claims, wherein the first fluorescence measurement and the second fluorescence measurement are obtained simultaneously.

22. The method according to any one of the preceding claims, wherein no photodegradation has occurred to the liquid in the time frame between the first fluorescence measurement and the second fluorescence measurement is obtained.

23. The method according to any one of the preceding claims, wherein the liquid has a transparency of preferably higher than 50%, more preferably higher than 60%, even more preferably higher than 80%, most preferably higher than 95%.

24. The method according to any one of the preceding claims, wherein the liquid has pH value preferably higher than 7, more preferably higher than 7.5, and preferably lower than 9, more preferably lower than 8.5, most preferably around 8.

25. The method according to any one of the preceding claims, wherein the liquid has a temperature higher than its freezing point, and a temperature lower than its boiling point.

26. The method according to any one of the preceding claims, further comprising the step of identifying the type of the liquid by correlating the fluorescence measurement(s) of the liquid to historic data, wherein the historic data comprise fluorescence measurements and excitation signals of liquids.

27. A computer implemented method for calibrating a fluorescence setup in relation to a liquid for estimating the reactivity of dissolved organic matter in said liquid, the method comprising,obtaining a plurality of datasets comprising spectra of fluorescent radiation over a plurality of time instances of the liquid, determining a number of base photosensitivity indexes for the liquid based on the spectra of fluorescent radiation, wherein each base photosensitivity index is determined over at least two time instances, correlating the fluorescence loss of the liquid with respect to an excitation wavelength range and an emission wavelength range, selecting a plurality of sets of first and second fluorescence measurements based on the correlation, each set corresponding to a distinct time instance, wherein each set of first and second fluorescence measurements comprises o a first and a second emission wavelength, and o a first and a second excitation wavelength, such that the plurality of sets of fluorescence measurements provide a maximum correlation between photosensitivity and fluorescence prior to an exposure, determining a ratio for each set of the first and the second fluorescence measurements on each time instance, determining a conversion factor by correlating the base photosensitivity indexes to the ratios, thereby calibrating the fluorescence setup by obtaining a liquid specific conversion to determine the sample photosensitivity index.

28. The method according to any one of the claims 1-26, wherein the conversion factor is determined according to claim 27.

29. The method according to claim 27, wherein the method is conducted on a plurality of samples of the same liquid type.

30. An optical measurement system for estimating the reactivity of dissolved organic matter in a liquid based on a photosensitivity index, the system comprising: at least a first light source for emitting to the liquid at least a first excitation signal comprising a first excitation wavelength and at least a second excitation signal comprising a second excitation wavelength,at least one sensor for detecting from the liquid at least a first fluorescence signal comprising a first emission wavelength and at least a second fluorescence signal comprising a second emission wavelength, and a processor configured to execute the method of any one of the claims 1 - 28.

31. The system of claim 30, comprising a liquid sample.

32. The system according to any one of the claims 30-31 , wherein the reactivity of dissolved organic matter is used to quantify the effect of sunlight radiation on the liquid.

33. The system according to any one of the claims 30-32, further comprising a temperature sensor, configured to measure the temperature of the liquid.

34. Use of a system for determining a photosensitivity index (or the dissolved organic matter) of a liquid according to claims 30-33, wherein the system is submerged into the liquid.

35. Use according to any one of the claims 33 and 34, wherein the liquid is natural freshwater sample, seawater sample, wastewater sample or beverage sample such as wine.

Images