AU2016202772A1 - Water quality sensor - Google Patents

Abstract

A method for detecting contamination of a water sample obtained from a water source is provided. The method includes the steps of (i) assessing fluorescent emission of a water sample in response to fluorescent excitation of said water sample; and (ii) comparing the result obtained in (i) with a reference, wherein a difference in fluorescent emission from the water sample as compared to the reference indicates contamination of the water sample. The water sample may be of potable water, and the contamination detected in the water sample may be contamination with recycled water. Also provided is a device and a system suitable for use in performing the method. 2697096vl 100101 1000 15-[-] ------ 300 500 320 120 600 -- 110 310 221 14 -------- -- ----1- Figure 1 2697119vl

Description

TITLE

WATER QUALITY SENSOR FIELD OF THE INVENTION

[0001] The present invention relates to water contamination. More particularly, the invention relates to a method for detecting contamination of water. The invention also relates to a device and system suitable for use in performing the method.

BACKGROUND OF THE INVENTION

[0002] With the introduction of dual-reticulation schemes, a considerable number of water quality incidents have occurred due to cross-connections and water system faults, posing risks to public health, as well as to the reputation and public perception of recycled water retailers. Water utilities across Australia have made multi-million dollar infrastructure outlays to distribute recycled water, and to mitigate the risk associated with cross-connections.

[0003] Water utilities currently focus on checking the plumbing connections of new dwellings (connected to a recycled water scheme) and an ongoing inspection program that inspects the plumbing of 5-20% of properties connected to recycled water annually. Whilst inspections methods are currently a workable solution, with the anticipated increases in numbers of customers with recycled water connections over the next decade, auditing targets may become difficult to achieve. Therefore the water industry continues to seek new and emerging technologies that may be able to provide a solution to the ongoing potential issues of cross-connections.

[0004] Electrical conductivity (EC) has previously been demonstrated to be a good basis for cross-connection detection. However in waters with unstable EC or where EC differences between potable and recycled water supplies are negligible, an alternative indicator is desired. Both UV absorbance and fluorescence may also vary between potable and recycled water supplies.

[0005] Industrial fluorescent equipment exists for detecting water contamination exists. However, such equipment measures absorbance spectra of a water sample at deep UV wavelengths, and due to the high cost of such equipment it is not routinely used in a household setting.

SUMMARY OF THE INVENTION

[0006] In a first aspect, the invention provides a method for detecting contamination of a water sample obtained from a water source, the method including the steps of: (i) assessing fluorescent emission of a water sample in response to fluorescent excitation of said water sample; and (ii) comparing the result obtained in (i) with a reference, wherein a difference in fluorescent emission from the water sample as compared to the reference indicates contamination of the water sample.

[0007] Preferably, the water sample is of potable water. Preferably, the contamination detected in the water sample is contamination with recycled water. In one preferred embodiment, said recycled water is Class A recycled water.

[0008] Preferably, the difference in fluorescent emission of the water sample as compared to the reference is an increase in fluorescent emission as compared to the reference.

[0009] Suitably, the degree or extent of the difference in fluorescence emission as compared to the reference is related to the degree or extent of contamination of the water sample.

[0010] Preferably, the reference is obtained by assessing fluorescent emission of a reference water sample (or “reference sample”) in response to fluorescent excitation of said reference sample. In a particularly preferred embodiment, said reference sample is trusted non-contaminated water from the same water source as the water sample that is assessed according to the method of this aspect.

[0011] Preferably the wavelength range of the fluorescent emission assessed according to the method of this aspect is 350 - 550 nm. More preferably said wavelength range is 400 - 500 nm.

[0012] Preferably, the wavelength range for fluorescent excitation of the water sample according to this aspect is 200 - 500 nm, or more preferably 250 nm - 400 nm. In one particularly preferred embodiment, said wavelength range is 310 - 370 nm. In alternative preferred embodiment said wavelength range is 270 - 300 nm.

[0013] In some preferred embodiments of the method of this aspect, the step of comparing with a reference the result obtained by assessing fluorescent emission of the water sample in response to fluorescent excitation of said water sample, includes calibration or adjustment of said result obtained and/or said reference.

[0014] Preferably, said result obtained is calibrated or adjusted based on the temperature of said water sample.

[0015] Preferably, in embodiments of this aspect wherein the reference is obtained by assessing fluorescent emission of a reference water sample in response to fluorescent excitation of said reference sample, said calibration or adjustment includes calibration or adjustment of the reference based on the temperature of the reference water sample.

[0016] In some preferred embodiments of the method of this aspect, the steps of (i) assessing fluorescent emission of the water sample in response to fluorescent excitation of said water sample; and (ii) comparing the result obtained in (i) with a reference, includes obtaining a plurality of individual readings of fluorescent emission as per (i), and/or comparing said readings to a plurality of individual references.

[0017] Suitably, in said preferred embodiments, said steps include a plurality of water sample to reference comparisons.

[0018] In some preferred embodiments, said method includes at least 3 water sample to reference comparisons. In other preferred embodiments, said method includes at least 6 water sample to reference comparisons. In still other preferred embodiments, said method includes at least 10 water sample to reference comparisons.

[0019] Preferably, the method of the first aspect is suitable for detecting contamination of the water sample at a minimum concentration of 10%, more preferably at a minimum concentration of 5%, or even more preferably at a minimum concentration of 2%, or even lower, such as a minimum concentration of 1 % or 0.5%.

[0020] Preferably, the water sample assessed according to the method of the first aspect is located within a pipe.

[0021] In preferred embodiments, said water sample is for domestic use. In said preferred embodiments, suitably, the water sample may be located within a pipe that is directly connected to a domestic residence or dwelling.

[0022] In preferred embodiments, the reference water sample according to the method of this aspect is located within a pipe. Suitably, the water sample assessed according to the method of this aspect does not flow, or does not substantially flow, into said pipe.

[0023] In some such embodiments wherein the water sample assessed according to the method of this aspect is contained within a pipe directly connected to domestic residence or dwelling, said pipe containing the reference sample may also be directly connected to the domestic residence or dwelling.

[0024] In some embodiments, the method of this aspect further includes the steps of (iii) assessing electrical conductivity of the water sample; and (iv) comparing the result obtained in (iii) with a reference, wherein a difference in electrical conductivity of the water sample as compared to the reference indicates contamination of the water sample.

[0025] In a second aspect, the invention provides a device for assessing fluorescence emission in a water sample, the device comprising: (i) a fluorescence excitation source; and (ii) a fluorescence detector, wherein (i) is adapted to induce fluorescent excitation of a water sample, and (ii) is adapted to detect fluorescence emission from the water sample in response to excitation from (i).

[0026] Preferably said device further includes (iii) an interface capable of transmitting and/or receiving signals from (i) and/or (ii).

[0027] Preferably said fluorescence excitation source and said fluorescence detector are radially oriented. In one particularly preferred embodiment said fluorescence excitations source and said fluorescence detector are oriented at approximately 90 degrees.

[0028] Preferably, the device is adapted for use on or in a pipe.

[0029] Suitably, said device is for use according to the method of the first aspect.

[0030] In a third aspect, the invention provides a system for detecting contamination of a water sample, the system comprising: (i) at least one device of the second aspect; (ii) a processor operatively linked to (i) and capable of determining if a change in fluorescent emission of the water sample in response to fluorescent excitation, as compared to a reference, has occurred; and (iii) an output operatively linked to (ii) and capable of notifying a user that contamination of the water sample has occurred, in response to detection of a change in fluorescent emission of the water sample by (ii).

[0031] In a preferred embodiment, the processor of the system of this aspect is operatively linked to at least two devices of the second aspect. Suitably, one of said devices is operatively connected with the water sample; and the other of said devices is operatively connected with a reference water sample. Preferably, one of said devices is located on or in a pipe containing the water sample, and the other of said devices is located on or in a pipe containing the reference sample.

[0032] Suitably, said system is for performing the method according to the first aspect.

[0033] In a fourth aspect, the invention provides a method of installing a device of the second aspect, said method including the step of operatively connecting a device of the second aspect to a water sample, or a reference water sample, to thereby install said device.

[0034] In a fifth aspect, the invention provides a method of installing a system of the third aspect, said method including the step of operatively connecting a system of the third aspect to a water sample and/or a reference water sample, to thereby install said system.

[0035] In a sixth aspect, the invention provides the use of a device of the fourth aspect or a system of the fifth aspect to detect contamination of a water sample.

BRIEF DESCRIPTION OF THE FIGURES

[0036] Figure 1 sets out a diagrammatic view of a preferred sensor of the invention.

[0037] Figure 2 sets out a diagrammatic view of a preferred system of the invention.

[0038] Figure 3 sets out UV-vis absorbance spectra for different concentrations of recycled water.

[0039] Figure 4 sets out a fluorescence excitation-emission map for 100% Class A recycled water.

[0040] Figure 5 sets out a fluorescence excitation-emission map for 100% potable water.

[0041] Figure 6 sets out a fluorescence excitation-emission map for 5% Class A water, 95% potable water.

[0042] Figure 7 sets out a fluorescence excitation-emission map for 10% Class A water, 90% potable water.

[0043] Figure 8 sets out a fluorescence excitation-emission map for 20% Class A water, 80% potable water.

[0044] Figure 9 sets out the setup of four in-line fluorescence sensors used for laboratory testing as described in Example 2. Figure 9A: schematic; Figure 9B photograph of testing setup.

[0045] Figure 10 sets out sensor readings for recycled water contaminated samples. Figure 10A shows results for 0 to 50% recycled water concentrations; Figure 10B shows results for 0 to 14% recycled water concentrations; Figure 10C shows results for 0 to 5% recycled water concentrations.

[0046] Figure 11 sets out a portable testing rig used for the field trials described in Example 3.

[0047] Figure 12 sets out fluorescence detection from a sensor of the invention placed in line with potable water (CC2) and one in line with the mixing tank (CC1), and output from an Electrical Conductivity sensor placed in line with the mixing tank (EC) for comparison, over the period of the field trial described in Example 3. Also shown is the variation in temperature during this period; fluorescence and EC values were not temperature corrected for this figure.

[0048] Figure 13 sets out a typical raw data set for potable and recycled water mixing cycles as per the field trial described in Example 3. CC1 = fluorescence sensor in mixing tank, CC2 = fluorescence sensor in the potable water line, EC = electrical conductivity sensor in mixing tank. Time step = 1 min.

[0049] Figure 14 sets out a diagrammatic summary of the approach used for inserting the simulated data points, as described in the examples. The original data is coloured in black and the inserted points in red. The range of three consecutive data points acquired from the same cross-connection ratio are analysed to find their minimum and maximum value. Then random numbers are selected so that they are uniformly distributed between these minimum and maximum values. Once selected, the desired quantity of such numbers is inserted below each original data range segment.

[0050] Figure 15 sets out the percentage of contamination events (‘cross-connections’) detected for different concentrations of recycled water vs the number of data points collected during each measurement at Facility 1.

[0051] Figure 16 sets out sample data used in the T-test calculation at the Facility 1 field trial (coloured cells indicate where cross-connections were detected over 99% of the time).

[0052] Figure 17 sets out recycled water quality data from Location 1 during the period of the Facility 1 trial. Recycled water data is a combination of measurements taken at Facility 1 and various places throughout the Location 1 reticulation system.

[0053] Figure 18 sets out potable water quality data from Location 1 during the period of the Facility 1 trial. Data was collected from numerous places throughout the Location 1 reticulation system.

[0054] Figure 19 sets out output from CC1, CC2 and EC for the duration of the field trial at Facility 2.

[0055] Figure 20 sets out (A) water quality data from Facility 2and the

Location 2 reticulation system during the period of the Facility 2 trial; (B) general potable water quality data from Facility 2and the Location 2 reticulation system.

[0056] Figure 21 sets out a typical raw data set for mixing cycles during the Facility 2 field trial. During the trial the ratio of recycled water to potable water was varied during a repetitive 45 min cycle comprising 30 min of testing together with a 15 min of cleaning and mixing. The figure demonstrates corrected data recorded during a 30 min cycle where mixtures of recycled water of 2.2, 4.3, 6.3, 8.2, 10.1, 11.9, 13.6, 15.2, 16.8, and 18.3% were created.

[0057] Figure 22 sets out a scatter plot of the CC1 fluorescence sensor for different CC1 concentrations for around 8,000 data points during the Facility 2 field trial, where stability in the signal measured by the sensors was consistent over time for each cross-connection percentage, as indicated by the clustering of the scatterplot.

[0058] Figure 23 sets out corrected Facility 2 CC1 sensor data used for the T-test statistical analyses.

[0059] Figure 24 sets out the percentage of contamination events (‘cross-connections’) detected for different concentrations of recycled water plotted against the number of data points collected during each measurement at Facility 2. Data is corrected data using additional modelled data points.

[0060] Figure 25 sets out sample data used in the T-test calculation at Facility 2 with an ongoing calibration (coloured cells indicate cross-connections were detected over 99% of the time).

[0061] Figure 26 sets out the estimated lifetime of LEDs for different operating conditions.

DETAILED DESCRIPTION OF THE INVENTION

[0062] The present invention is at least partly predicated on the discovery that exposing a water sample to fluorescent excitation, and assessment of a change in fluorescent emission in response to said fluorescent excitation, can be useful for detecting contamination in the water sample.

[0063] In one aspect, the invention provides a method for detecting contamination of a water sample obtained from a water source by assessing fluorescent emission of a water sample in response to fluorescent excitation of the water sample, and comparing the fluorescent emission of the water sample to a reference, wherein a difference between the fluorescent emission of the water sample and the reference indicates contamination of the water sample.

[0064] In embodiments herein described the water sample is of potable water and the contamination detected in the water sample is contamination with recycled water. It will be further appreciated that the invention may have application for assessing cross-contamination and mixing regimes of a range of water samples, including, without limitation, rainwater, groundwater, stormwater, seawater, and river water.

[0065] As used herein, “potable watef refers to water that is fit for human consumption. In some embodiments, potable water refers to water that meets relevant regulatory criteria. By way of example, it will be appreciated that, in some embodiments, potable water may be water that meets the ‘Australian Drinking Water Guidelines’ (Australian Drinking Water Guidelines 6, 2011, Version 3.2, Updated February 2016; incorporated herein by reference).

[0066] As used herein, “recycled watef refers to wastewater, e.g. sewage, although without limitation thereto, that is treated to remove impurities. Recycled water is used for a range of applications in commercial, industrial, agricultural and domestic or residential sectors.

[0067] Recycled water can be classified according to purity and/or suitability for end use. The ‘Australian Guidelines for Water Recycling’ (.Australian Guidelines For Water Recycling, Managing Health And Environmental Risks (Phasel), 2006; incorporated herein by reference) provides detailed information regarding characteristics of recycled water that render it suitable for particular end use.

[0068] As will be understood by the skilled person, an A-D classification scheme of recycled water is sometime used, wherein categories A-D indicate suitability of the recycled water for end use. An example of such end use characteristics is as follows:

Class D: (lowest grade) suitable for irrigation of non-food crops such as turf, timber or flower lots.

Class C: suitable for irrigating of food crops, sporting facilities and parks.

Class B: suitable for livestock drinking, cattle grazing and wash-down water.

Class A: (highest grade) suitable for human contact, but not approved for drinking, bathing or swimming.

[0069] Suitably, said recycled water detected according to the method of this aspect is Class A recycled water.

[0070] According to embodiments of the method as herein described, a reference is obtained by assessing fluorescent emission of a reference water sample (alternatively referred to herein as a “reference sample”) in response to fluorescent excitation of said reference sample.

[0071] It will be appreciated that the properties of the fluorescent excitation used and fluorescent emission assessed (e.g. wavelength; intensity) for the reference sample as per the method of this aspect, are the same, or substantially the same, as those used and assessed, respectively, for the water sample.

[0072] In embodiments, the reference sample is obtained from the same water source as the water sample. By way of example, both the reference sample and the water sample may be obtained from the same water storage facility (e.g. the same dam or other reservoir). Suitably, the reference sample is a trusted, non-contaminated water sample obtained from the same water source as the water sample.

[0073] In some embodiments of the method of this aspect, the fluorescence emission from the water sample and/or the reference sample may be adjusted or corrected to account for one or more variables.

[0074] In an embodiment, the fluorescence emission from the water sample is adjusted or corrected based on the temperature of the water sample and/or the temperature of a reference sample. With reference to the examples, it will be appreciated that adjusting or correction fluorescence emission based on the relative temperature of the water sample and a reference water sample can increase accuracy and/or sensitivity of detection of contamination according to the method of this aspect.

[0075] In an embodiment, the fluorescence emission is adjusted or corrected in response to a decrease in excitation intensity to which the water sample and/or a reference sample is exposed. With reference to the examples, it will be appreciated that adjusting or correcting fluorescence emission in response to a decrease in excitation intensity to which the water sample and/or a reference sample is exposed can increase accuracy and/or sensitivity of detection of contamination according to the method of this aspect.

[0076] It will be further appreciated that the fluorescence emission from the water sample and/or the reference sample may be adjusted to account for differences or changes in water quality properties including, for example, particulate concentrations.

[0077] In embodiments herein described, the wavelength range of the fluorescent emission assessed according to the method of this aspect is 350 - 550 nm. More specifically, said wavelength range is 400 - 500 nm.

[0078] With reference to the examples and Figures 4-8, it will be appreciated that, upon suitable fluorescent excitation, the intensity of fluorescent emission of wavelengths within the abovementioned ranges was substantially increased in recycled water samples and potable water samples contaminated with various concentrations of recycled water, as compared to potable water.

[0079] The skilled person will appreciate that the wavelength range of 400 - 500 nm falls within the visual wavelength spectrum (i.e. about 380 nm - about 750 nm). Such assessment of fluorescent emission within the visual spectrum may be particularly advantageous for the invention. In this regard, as will be understood by the skilled person, assessment of fluorescent emission outside of the visual range may require equipment that is highly specialized and/or expensive. In contrast, assessment of fluorescent emission within the visual range may require less specialized and/or expensive equipment.

[0080] In embodiments herein described, the wavelength range for fluorescent excitation according to this aspect is 200 - 500 nm, or more preferably 250 nm - 400 nm. In one embodiment, the wavelength range is 310 - 370 nm. Alternatively, the wavelength range may be 270 - 300 nm.

[0081] With reference to the examples and Figures 4-8, it will be appreciated that fluorescent excitation using wavelengths within the abovementioned ranges resulted in particular fluorescent emissions that were substantially increased in recycled water samples and potable water samples contaminated with various concentrations of recycled water, as compared to potable water.

[0082] It will be appreciated that the wavelength range of 310 - 370 nm is within the ultra violet (UV) spectrum that is near to the visible light spectrum. Fluorescence excitation using a wavelength that is near to the visible light spectrum may be particularly advantageous for the invention. In this regard, as will be understood by the skilled person, fluorescent excitation using a wavelength that is far from the visible light spectrum (e.g. deep UV excitation) may require equipment that is highly specialized and/or expensive. In contrast, fluorescent excitation using a wavelength that is close to the visual range may require less specialized and/or expensive equipment.

[0083] In embodiments of the method of this aspect, the steps of assessing fluorescent emission of the water sample in response to fluorescent excitation of said water sample; and comparing the fluorescent emission with a reference, include a plurality of water sample to reference sample comparisons.

[0084] With reference to the examples and Figures 14-16 and 24-25, it will be appreciated that the use of a plurality of water sample to reference sample comparisons can result in increased accuracy and/or sensitivity for detection of contamination of a water sample according to the method of this aspect. As hereinbelow described, based on models developed using data collected in a field testing scenario, increasing the number of water sample to reference sample comparisons generally increased the percentage of potable water sample contamination with Class A recycled water events (‘cross-connections’) that were detected using one preferred embodiment of the method of this aspect.

[0085] In some preferred embodiments, the method of this aspect includes at least, 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, at least 19, or at least 20 water sample to reference comparisons.

[0086] In embodiments, the comparison of fluorescent emission of the water sample in response to fluorescent excitation with a reference, as per the method of this aspect, is a statistical comparison. In said embodiments, the difference in fluorescence emission in the water sample as compared to the reference is a statistically significant difference.

[0087] In some embodiments, said statistically significant difference is a p value of 0.05 or less, or preferably 0.01 or less, based on a paired T-test.

[0088] In embodiments, the method of this aspect is suitable for detecting contamination of the water sample at a minimum concentration of 10%, more preferably at a minimum concentration of 5%, or even more preferably at a minimum concentration of 2%, or even lower, such as a minimum concentration of 1 % or 0.5%.

[0089] With reference to the examples and Figures 14-16 and 24-25, it will be appreciated that, based on modelling conducted using data obtained in a field testing scenario, one preferred embodiment of the method of this aspect was capable of detecting contamination events (‘cross-connections’) of 2% contamination of potable water with Class A recycled water at a frequency of up to ~ 80%; contamination events of 4% contamination of potable water with Class A recycled water at a frequency of up to ~ 95%; and contamination events of 6% and above contamination of potable water with Class A recycled water at a frequency of ~ 100%.

[0090] Additionally, with reference to the examples and Figure 10, it will be appreciated that under controlled laboratory settings, it was demonstrated that the method has potential to detect levels of contamination as low as 0.5%, or even lower.

[0091] In embodiments, the water sample assessed according to the method of the first aspect is located within a pipe.

[0092] In the context of this invention a “pipe" refers to any hollow member suitable for conveying water. It will be appreciated that while pipes commonly have substantially circular cross section, as used herein a pipe need not necessarily have substantially circular cross section and may also have any other suitable cross-sectional shape, such as by way of nonlimiting example, square, rectangular, or triangular.

[0093] In embodiments, the water sample assessed according to the method of this aspect is for domestic use. Suitably, the water may be located within a pipe that is directly connected to a domestic residence or dwelling.

[0094] Alternative or additionally, the water sample may be for industrial or commercial use.

[0095] Suitably, the reference sample according to the method of this aspect is contained within a pipe. In embodiments wherein the water sample assessed according to the method of this aspect is contained within a pipe directly connected to domestic residence or dwelling, the pipe containing the reference water sample may also be directly connected to the domestic residence or dwelling.

[0096] As hereinbefore described, in some embodiments the water sample assessed according to the method of this aspect, and the reference water sample, are obtained from the same water source. Suitable, the water sample assessed according to the method of this aspect does not flow into the pipe containing the reference sample. It will appreciated, however, that the reference sample may flow into the pipe containing the water sample.

[0097] In particular, it will be appreciated that the pipe containing the reference sample may be connected to the pipe containing the water sample, wherein water flows substantially only in the direction from the pipe containing the reference sample to the pipe containing the water sample.

[0098] By way of non-limiting example, the pipe containing the water sample may be located internally within a domestic residence or dwelling, such as a pipe directly connected to a faucet of a kitchen sink, and the pipe containing the reference water sample may be located nearer to the water source, e.g. a ‘mains’ water pipe which provides the water source to the domestic residence or dwelling.

[0099] In some embodiments, the method of this aspect includes the further steps of assessing electrical conductivity of the water sample; and comparing the electrical conductivity of the water sample with a reference, wherein a difference in electrical conductivity of the water sample as compared to the reference indicates contamination of the water sample.

[00100] With reference to the examples, it will be appreciated that the use of both fluorescent and electrical conductivity assessment according to embodiments of the invention which include the above steps may increase accuracy and sensitivity of detection of water contamination.

[00101] In another aspect, the invention provides a device for assessing fluorescence emission in a water sample.

[00102] With reference to Figure 1, a preferred embodiment of a device of the invention is described in detail as follows.

[00103] Device 10 comprises: interface 100; T-piece plumbing fitting 200; fluorescence excitation source 300; fluorescence detector 400; temperature sensor 500; and feedback photodiode 600.

[00104] Plumbing fitting 200 of device 10 is connected to interface 100. Fluorescence excitation source 300, fluorescence detector 400, temperature sensor 500, and feedback photodiode 600 are mounted to plumbing fitting 200.

[00105] As pictured, device 10 is attached to pipe 1. More specifically, plumbing fitting 200 of device 10 is located between sections 1A and 1B of pipe 1, which sections are substantially perpendicular. As represented by the arrows in sections 1A and 1B, water flows from section 1 A, through plumbing fitting 200, to section 1B. As pictured, pipe 1 contains a water sample of potable water. Flowever, it will be appreciated that pipe 1 may alternatively contain a reference sample.

[00106] Interface 100 of device 10 comprises microprocessor 110; memory 120; analogue to digital converter (ADC)-amplifier 130; real-time clock 140; and USB port 150.

[00107] Plumbing fitting 200 comprises side wall 210; end 220; and window 221. Window 221 is a Perspex UV transmission window sealed with an O-ring. Window 221 is oriented substantially perpendicularly to side wall 210.

[00108] Fluorescence excitation source 300 comprises current source 310, and ultraviolet radiation (UV) source 320. As pictured, UV source 320 is in the form of two UV 355 nm LEDs, however other suitable UV sources may be used. The UV 355 nm LEDs of UV source 320 protrude substantially perpendicularly through side wall 210 of plumbing fitting 200, such that UV light emitted from said LEDs is directed away from side wall 210.

[00109] It will be appreciated that the orientation of the 355 nm LEDs of UV source 320 with respect to end 220 may be particularly advantageous for device 10, as this avoids emitting UV from said LEDs directly towards fluorescence detector 400, which is located within end 220 as hereinbelow described.

[00110] Fluorescence detector 400 is located within end 220 of plumbing fitting 200. It will be appreciated that window 221 separates fluorescence detector 400 located within end 220 from the water sample flowing through plumbing fitting 200. Fluorescence detector 400 comprises two UV 400 nm long pass filters; and a polymethylmethacrylate spot beam lens with a mounted photodiode (not shown).

[00111] It will be appreciated that the location of window 221 and fluorescence detector 400 may be particularly advantageous for device 10. Specifically, where device 10 is oriented such that side 210 is inclined or vertical, and end 220 is located at the bottom of the device, the accumulation of air bubbles on or near to window 221 can be minimized. Such accumulation of air bubbles is undesirable for device 10, as it may inhibit or interrupt transmission of fluorescence to fluorescence detector 400.

[00112] Temperature sensor 500 is a digital temperature sensor.

Temperature sensor 500 is positioned to be immersed in water flowing through plumbing fitting 200.

[00113] Feedback photodiode 600 is mounted external to plumbing fitting 200, behind the UV LEDs of UV source 320.

[00114] In use, microprocessor 110 of interface 100 is capable of signalling fluorescence excitation source 300. Additionally, microprocessor 110 is capable of receiving fluorescence information from fluorescence detector 400, via ADC-amplifier 130. Microprocessor 110 is also capable of receiving fluorescence information from calibration photodiode 600, and temperature information from temperature sensor 500.

[00115] As pictured, the aforementioned signalling occurs electrically over wires, as represented by connecting lines in Figure 1. However, it will be appreciated that any suitable alternative signalling, such as wireless signalling, may also be used.

[00116] Optionally, in use, interface 100 may be capable of transmitting and/or receiving data to and/or from an external source, such as a processor of the system of the invention as hereinbelow described (not shown).

[00117] In use, upon relevant signalling from microprocessor 110, UV source 320 of fluorescence excitation source 300 emits UV light in the 355 nm range, and said UV light excites the water sample contained within pipe 1.

[00118] Upon excitation of the water sample contained within pipe 1 by fluorescence excitation source 300, fluorescence emission is transmitted from the water sample to fluorescence detector 400.

[00119] In use, fluorescence emission transmitted from the water sample travels through window 221 of plumbing fitting 200, through the UV 400 nm long pass filters of fluorescence detector 400, and onto the spot beam lens with mounted photodiode of fluorescence detector 400. Fluorescence detector 400 thereby detects fluorescence emission transmitted from the water sample in the greater than 400 nm wavelength range. Furthermore, in use, fluorescence detector 400 transmits data on said fluorescence emission detected to microprocessor 110 of interface 100.

[00120] In use, temperature sensor 500 detects temperature of the water samples contained within plumbing fitting 200, and transmits data on said temperature detected to microprocessor 110.

[00121] In use, feedback photodiode 600 detects UV intensity emitted by fluorescence excitation source 300, and transmits data on said emission to microprocessor 110. In response, microprocessor 110 may alter or change signalling to fluorescence excitation source 300 based on emission data received from feedback photodiode 600, to alter or change UV intensity emitted by fluorescence excitation source 300.

[00122] In another aspect, the invention provides a system for detecting contamination of a water sample.

[00123] With reference to Figure 2, a preferred embodiment of a system of the invention is described in detail as follows.

[00124] System 10 comprises: a first device of the invention 100A, operatively connected with a water sample; a second device of the invention 100B, operatively connected with a reference sample; processor 200; and output 300.

[00125] Processor 200 is capable of transmitting and receiving signals between both device 100A and device 100B. As depicted in Figure 2, said signals are transmitted wirelessly as represented by connecting lines. However, any suitable alternative signalling may be also used, as desired, such as wired signalling.

[00126] Output 300 is capable of receiving a signal from processor 200. Additionally output 300 is capable of notifying or alerting a user in response to a relevant signal from processor 200. As pictured, output 300 comprises an audible alarm for notifying a user, however this can be varied as desired.

[00127] In use, processor 200 transmits data to respective interfaces of device 100A and device 100B. In response, the respective interfaces signal the respective fluorescence excitation sources as hereinabove described.

[00128] In use, processor 200 receives and compares data from device 100A and device 100B regarding fluorescence detected in the 400 - 500 nm wavelength range in response to excitation in the water sample, and reference sample, respectively.

[00129] In use, processor 200 can calibrate or adjust the respective fluorescence data received from device 100A and device 100B, as required, prior to comparison. Suitably, processor 200 receives data from device 100A and device 100B regarding temperature of the water sample and the reference sample, respectively (as detected by the respective temperature sensors of the devices as hereinabove described), and performs calibration of the respective data accordingly.

[00130] In some embodiments processor 200 may further perform calibration based on internal fluctuations in fluorescence in the water sample and/or the reference sample detected by respective devices 100A and 100B.

[00131] In use, based on suitable criteria, processor 200 determines if a relevant difference between fluorescence in the 400 - 500 nm wavelength range of the water sample and the reference sample exists. In some embodiments processor 200 performs comparisons between multiple corresponding fluorescence signals obtained from device 100A and device 100B, and determines if a statistically significant difference exists between fluorescence of the water sample and fluorescence of the reference sample, based on these multiple comparisons.

[00132] In use, upon detection of a relevant difference between fluorescence of the water sample and the reference sample, processor 200 transmits a signal to output 300.

[00133] In use, upon receipt of a signal from processor 200 transmitted in response to detection of a relevant difference between fluorescence of the water sample and the reference sample, output 300 raises an audible alarm suitable for notifying a user.

[00134] The invention also provides a method of installing a device or system for detecting water contamination, for example the device or system as described above with reference to Figures 1 and 2, respectively. It will be appreciated that installation according to these aspects may be by any of the range of suitable means that will be readily apparent to the skilled person.

[00135] To assist the skilled person to further understand and put the invention into practical effect, the following non-limiting examples are provided.

EXAMPLES EXAMPLE 1. Analytical Characterisation of Water Samples.

[00136] Potable and Class A recycled water was obtained from several water supplies and the fluorescence emission was monitored with a Fluorescence Spectrophotometer to determine fluorescence emission characteristics. Additionally, recycled water was mixed in different “contamination” ratios with potable water from the same source to provide practical reference standards for sensor development.

[00137] UV-vis absorption spectra for all samples was obtained using using standard 1 cm path length quartz cuvettes. The UV-vis absorption spectra from the majority of water samples were virtually featureless in the range from 240 to 800 nm, with only an absorption peak in each spectra centred at around 210 nm (Figure 1). With current technologies available, building a sensor for monitoring wavelengths in the range from 200 to 240 was not considered to fall within desirable guidelines for the invention of being affordable, as it would involve costly light sources and detectors, and nor would utilizing such wavelengths in the UV spectrum be desirable due to power usage and material suitability and degradability.

[00138] Excitation emission matrices (EEMs) were also collected for each of the potable-recycled water combinations. Despite the absorption data not being capable of providing a clear indication of recycled water contamination, the fluorescence emission in response to excitation was capable of providing a clear distinction of recycled water from all sites. Fluorescence emission maps for different concentrations of recycled water from each site showed two distinct regions of fluorescence excitation, the first being in the region of 250-270 nm and the second from 310-370 nm (Figure 4-8). Without being bound by theory, the 254 nm absorption band is typically a measure of the aromatic character of dissolved organic matter, whereas the aliphatic molecules absorb in the range of 300-370 nm (see e.g. Hambly, ‘Fluorescence as a portable tool for cross-connection detection within dual reticulation systems’. 2013. PhD Thesis, University of New South Wales).

[00139] In view of the results set forth above, it was appreciated by the inventors that strong emission signals suitable for detecting water contamination could be obtained using excitation sources with wavelengths in the visual range (e.g. 320-370 nm). This could obviate the need for expensive and hazardous UV-B (290-320 nm) or UV-C (200-290 nm) excitation sources. EXAMPLE 2. Laboratory Assessment of Contamination Detection.

[00140] Sensors of the invention as herein described with reference to Figure 1 were used for laboratory assessment of detection of contamination of potable water with recycled water. For these experiments, based on the results described in Example 1, ‘challenge’ samples from the site in which the Class A recycled water exhibited the most similar fluorescence characteristics to the corresponding potable water were used, in order to robustly assess the accuracy and sensitivity of the sensors. This recycled water was produced at a facility using ozonation, biological media filtration, ozonation, UV disinfection and chlorination.

[00141] Prior to testing, potable and recycled water was allowed to equilibrate at room temperature overnight. Samples containing concentrations of 0 - 50% recycled water were then produced by mixing appropriate amounts of recycled water into a reservoir tank. Water from the reservoir was continually pumped through four replicate inline sensors as set forth in Figure 9, and the readout from their respective optical detectors was displayed in real time.

[00142] Figure 10 shows an exemplary response of the sensor to levels of contamination from 0% to 50% of potable water with the recycled water. Numerous replicates of the test were performed and a linear relationship between the sensor output and the amount of recycled water mixed contamination observed. It will be appreciated based on the results set forth in Figure 10B and Figure 10C, that detection of concentrations of recycled water contamination of 1%, or even lower such as 0.5%, was possible. EXAMPLE 3. Field Assessment of Contamination Detection [00143] For field assessment of sensors of the invention, the portable rig set forth in Figure 11 was used to allow for precise mixing of potable and recycled water for simulating different cross-connection ratios. The system includes waterproof housings for electronics, data logging and external communication to a web server, circuits, mixing tank and flow sensors. The field trial rig accommodates two of the fluorescence sensors of the invention as herein described with reference to Figure 1: one in line with the potable water and one in line with the mixing tank to measure the fluorescence of the different cross-connection mixtures. The field trial rig also contains an Electrical Conductivity sensor, placed in line with the mixing tank, for comparative assessment.

Trial 1. Facility 1 Water Recycling Facility [00144] A first field trial was conducted at Facility 1 and ran from the 2 March 2015 to 8 April 2015. During this time, the desired contamination ratio was achieved by periodically dosing the mixing tank with a known volume of recycled water. Figure 12 provides an overview of the fluorescence signal from the sensor placed in line with mixed recycled and potable water circulated from the mixing tank (CC1) and the sensor placed in the potable water line (CC2), and the signal from EC sensor (placed in line with CC1) for comparison, over the period of the trial. Figure 12 further sets out the fluctuations in temperature during the period of the trial - the fluorescence and EC data in Figure 12 is not corrected for temperature changes, however, a linear correction was subsequently applied to all data so that a direct comparison could be made at a constant temperature.

[00145] Figure 13 demonstrates typical data recorded during a 30 min cycle where mixtures of recycled water with potable water of 2.2, 4.3, 6.3, 8.2, 10.1, 11.9, 13.6, 15.2, 16.8, and 18.3% were created. Aside from two times when the EC sensor failed to respond, the majority of the data was collected without significant drift or complication. These data demonstrate that the EC and fluorescence sensors have a similar ability to distinguish between water quality types.

[00146] Within the portable field trial rig, the two prototype sensors are operated in a paired configuration for monitoring the fluorescence of both potable and recycled water (i.e. data is collected simultaneously from both CC1 and CC2). This approach lends itself to the utilisation of a paired T-test, which can provide a continuously updated statistical evaluation as to whether the readings from the two sensors are correlated. In practice, this would be akin to one sensor operating within a potable reticulation system and having the other sensor located within a premises where the presence of a cross-connection was a possibility.

[00147] The ability of a paired T-test comparison under the circumstances described above to accurately and sensitively detect contamination will depend at least in part, on the number of sensor readings compared. To specifically assess the effect of number of sensor readings on detection accuracy and sensitivity additional simulated data points were added to the field trial data to assess improved differentiation.

[00148] Figure 14 illustrates the approach used for inserting the simulated data points. The original data is coloured in black and the inserted points in red. The range of three consecutive data points acquired from the same CC ratio are analysed to find their minimum and maximum value. Then random numbers are selected so that they are uniformly distributed between these minimum and maxim values. Once selected, the desired quantity of such numbers is inserted below each original data range segment.

[00149] The original data consisted of three consecutive sensor readings from the same cross-connection ratio, the simulated data points were determined by selecting a random number that was uniformly distributed between the minimum and maximum sensor readings of those three consecutive readings. In doing so, it was possible to add as many of such points as desired to the field trial data. A random portion of the raw field trial data consisting of 12,500 data points was selected for this investigation.

[00150] The statistical improvement made in accurately confirming contamination event cross-connection through the use of a greater number of fluorescence measurements is illustrated in Figures 15 and 16. The plots show the percentage of cross-connections detected as a function of the number of data points collected during each measurement for different concentrations of recycled water. For the case of 2% contamination, there is a clear asymptotic increase in the percentage of contamination events detected with increasing number of samples, which tends to level out at around 79% for 20 samples per measurement. For a cross-connection ratio of 4%, around 82% of cross-connections are detected for 3 samples which increases asymptotically to around 96% for 20 samples. When the cross-connection ratio was at 6%, the detection limit approaches 100% much more rapidly and once the cross-connection ratio is above this value then virtually all cross-connections are detected regardless of the number of samples acquired. The highlighted cells in Figure indicate that more than 99% of cross-connections were detected.

Trial 2. Facility 2 [00151] A second field trial was conducted at Facility 2. As was the case during the Facility 1 field trial (Trial 1), sensors CC1 and EC were exposed to the different cross-connection ratios in the mixing tank and sensor CC2 was placed in-line with the potable water sample at Facility 2.

[00152] Figure 20A and 20B provide a summary of water quality measurements from the Facility 2 (‘Class A’ outlet and at the customer tap) as well as data from the local potable supply (Location 2). It was also noted that the pH of the potable water sample, along with turbidity and total dissolved solids can fluctuate over a yearly cycle, yet appear to do so in a relatively controlled manner as levels did not change significantly during the trial period. There was a notable difference in the colour readings of Class A recycled water and potable water, which is suspected to be related to the organic components of the water such as humic acids. This suggests that fluorescence sensing approaches may be particularly well-suited to detect cross-connections in the system at Facility 2.

[00153] Figure 19 presents the output data for each of the sensors corrected to a 22°C temperature. The raw temperature of the water tank is also reported for completeness. As for Trial 1, CC2 and EC were exposed to the repetitious blending of waters ranging from potable water to 18.3% cross-connection over a five-week period (from April 8 2015 to May 18 2015).

[00154] Several observations can be made directly from this figure. Firstly the EC sensor went offline on four occasions. Secondly, the EC value of the potable water at Facility 2 is almost an order of magnitude greater than that of Facility 1, yet the fluorescence baseline value was not significantly changed. An additional point of relevance is that CC1 experienced a reduction in its baseline value over the duration of the trial. This was most likely due to a slow burn out of the CC1 excitation source, the reason for which, as well as the significance, is discussed in subsequent below. There was also a brief power outage around the 30th of April and hence data is absent for that period.

[00155] Owing to the slow decrease in the background fluorescence for CC1, further analysis was performed using data was corrected to provide an ongoing calibration. In practice, this calibration procedure would not be available for a specific sensor and implications are discussed further below. However, for the purpose of demonstrating a scenario in which this artefact was not present, a portion of 10,000 corrected samples, were used to assess the accuracy of cross-connection detection.

[00156] Figure 23 demonstrates corrected data, taking into account the abovementioned reduction in CC1 baseline value over the duration of the trial.

[00157] Figure 21 demonstrates data recorded during a 30 min cycle where mixtures of recycled water of 2.2, 4.3, 6.3, 8.2, 10.1, 11.9, 13.6, 15.2, 16.8, and 18.3% were created.

[00158] Figure 22 shows a scatter plot of the of the CC1 fluorescence sensor for different CC1 concentrations for around 8,000 data points, where stability in the signal measured by the sensors was consistent over time for each cross-connection percentage, as indicated by the clustering of the scatterplot.

[00159] Contamination of potable water with recycled water was assessed using paired T-test calculation based on a moving average of the previous three differences in the CC1 and CC2 sensor reading, as per the previous test at Facility 1.

[00160] Figure 24 shows a plot of the percentage of cross-connections that were detected for different concentrations of recycled water as a function of the number of sample points acquired. It is seen clearly that for 2% contamination, the curve once again increases logarithmically and tends to asymptote to around 90%. For virtually all other scenarios, the cross-connection scenario was virtually always detected, regardless of the number of samples acquired. A full list of results can be found Figure 25. The highlighted cells indicate that more than 99% of cross-connections were detected.

[00161] Without being bound by theory, it appears that the greater sensitivity and accuracy with which contamination was detected in the Facility 2 trial (Trial 2), as compared to the Facility 1 trail (Trail 1), taking into the abovementioned correction of the CC1 baseline, can be largely attributed to the increased dynamic range of the fluorescence from CC1 sensor data, seemingly due to increased organics in the recycled water.

[00162] [00163]

Calibration [00164] As noted above, in view of the field data collected, it is apparent that ongoing calibration of sensors can be important for enabling an accurate assessment of a cross-connection. In conducting a running paired T-test between the readings of any two sensors, the critical factor is the baseline value as it represents the “null-hypothesis” in the T-test (i.e. it responds to the difference between the mean values of two datasets at their respective baseline values).

[00165] There are several factors that could affect the baseline value, the main ones being biofouling of optics components and degradation of the sensor component. The calibration is also complicated by the dependence of fluorescence on ongoing temperature and water quality changes. Temperature changes can be accounted for by compensation through simultaneous measurement. The degradation of components can in some cases also be accounted for (which is addressed in the following subsection). A change in quality of the supplied water can be slightly more challenging.

[00166] In regard to potential changes in water quality, throughout the field trials set forth herein, the portable system was able to observe clear steps and changes in the recorded sensor data for both potable and recycled ‘Class A’ water. Upon noticing this, the water suppliers were contacted and it was found that at the time points in question, the treatment facility had undergone changes. These changes included a change in chlorination regimes in the facility and outages or changes to the purification system. Although it was positive to note that such changes in the water treatment system were spotted directly from the sensor data, at the same time, such changes result in a step change in either the recycled or potable water, and this change could interfere with an ongoing baseline calibration of the sensors.

[00167] In view of the above, it will be appreciated that periodic calibration of the installed sensors is desirable in order to ‘reset’ any offset in the ongoing sensitivity of the device such that the ‘null-hypothesis’ for the calculation can be established on precise credentials. One possibility for achieving this is to use the data values themselves as a means for calibration. During the field trials this was relatively straightforward, as a known quantity of each type of water could be dosed to all the sensors, and hence the ‘null-hypothesis’ condition could be reassessed. However achieving this in a real scenario could be quite different. One means of approaching this may be an inbuilt calibration solution that doses a precise amount of a fluorescent compound on demand, or could involve an evacuation of water from the sensor in order to run a periodic baseline check.

Lifetime of the excitation source in the device [00168] As set forth above, from the fluorescence data presented from the Facility 2 field trial, a substantial decay in the baseline value was observed.

This decay, although not obvious on a day-to-day basis, could either be attributed to biofouling of the optics or a slow decay in the efficiency of the excitation source. Upon inspection of the optics of CC1 at the completion of the Facility 2 trial there were no signs of biofouling, nor was there a measured increase in signal following cleaning. Therefore, the slow decay in signal was attributed to a loss in the LED excitation source’s emission intensity over time. This particular excitation source had been used continuously for approximately six month of laboratory experiments and both field trials.

[00169] Importantly, in a real application scenario, the LED would not be required for continuous 24/7 operation (as was the case in the field trial conducted for this example) but rather they would have a much lower ‘duty cycle’ (i.e. the ratio of time on to time off). For example, for a duty cycle of 1 minute in 20 minutes (5% duty cycle), the energy consumption drops to around 75%, under which conditions the lifetime of a suitable, currently available LED excitation source would be expected to be over 3 years. This is illustrated graphically in Figure 29.

[00170] Furthermore, the current that flows through the LED can be varied, and thus the power (hence, the output light intensity) depends on the current flowing through it, i.e. the more the current, the brighter the light. During the field and lab trials, the LED excitation sources were operated in a constant current mode, i.e. the electrical current passing through the LED is maintained at a constant level, in a continuous manner 24 hours per day. As per the manufacturers data sheets of the LEDs used, the upper level for the current flow is around 25 mA, which results in a high output emission (beyond that required for this project). Decreasing to the current from 25 mA to a more appropriate value of around 5 mA could see the lifetime increase from 63 days to 313 days.

[00171] Realistically, the full-time utilisation of the LED excitation source in the present work would equate to a 10-year lifetime at 5% duty and as such calibration may be required on an annual or six-monthly basis. On these timeframes, it may be biofouling or other materials degradation issues that may lead to concern, rather than LED degradation.

[00172] In order to compensate for the slow burn out and decay in the LED excitation source, which was appeared to be the major contribution to error in the present work, one strategy would be to place a secondary light detector (photodiode) in the vicinity of the excitation source. This detector could monitor the LED’s emission and then provide feedback to the driving circuits. Should the LED’s output decrease, the feedback detector could initiate a response to appropriately adjust the current levels through the LED to ensure a constant excitation intensity. Ultimately, this would be achieved through the utilisation of an LED excitation source with a built-in photodiode that is capable of assessing its own UV intensity, though this may add to the cost of the device assembly. EXAMPLE 3. Comparison of Fluorescence and EC Measurement Approaches [00173] Work carried out in parallel to the current project has demonstrated that EC is able to detect cross-connections down to 2% in selected recycling/potable water systems (Jayaratne et al., 2015). When undertaking similar measurements in the current project, the fluorescence prototypes have been demonstrated to achieve similar detection levels, <1% in the laboratory under controlled conditions and <4% in field conditions. Both EC and fluorescence are able to easily distinguish between difference potable water types and can therefore be used to provide information on mixing regimes within potable reticulation systems. The work undertaken for the current invention has however highlighted that a means of calibration of sensors, both fluorescence based as described herein, and EC, can be particularly beneficial for the accuracy and sensitivity of contamination detection.

[00174] The complementary nature of EC and fluorescence sensors is demonstrated in the below data, wherein the ratio of sensor outputs in recycled and potable water are presented.

[00175] It will be evident that at Facility 1, there was a significant difference in the EC levels in recycled and potable water, 10.5 : 1, as opposed to fluorescence differences of approximately 2.2 : 1. Hence in this situation, using an EC sensor to detect cross-connections may be more accurate than fluorescence sensors. In contrast, the system at Facility 2 shows that fluorescence sensors are likely to be more accurate due to an increased ratio in signal from recycled water when compared to potable, 2.6 : 1, and EC measurements, whilst being higher at Facility 2, only have a twofold difference between recycled and potable supplies.

[00176] It follows that a combination fluorescence and EC sensing may be particularly powerful detection of contamination of potable water with recycled water in at least some circumstances. EC is determined by dissolved salt concentrations whereas fluorescence is dependent almost independently on organic carbon contents within the water. Such a device may have particular flexibility to be accurate in a wide range of water quality types, could allow benefits in allowing a water retailer to change treatment, blend or supplement potable water with particular confidence. Such a dual detection could also potentially enable additional robustness of detection whereby a cross-connection event alert from one sensing element could be either supported or challenged by the second independent sensing element. Suitably, the same statistical approaches for water sample to reference sample comparisons as described herein with respect to fluorescence comparisons could be applied using EC data in conjunction with fluorescence data. Where differences between water sample and reference sample were detected based on both fluorescence and EC data, this may provide a particularly robust indication of water contamination. EXAMPLE 4. Summary remarks in view of laboratory and field results [00177] Embodiments of this invention deliver a ‘low-cost’ fluorescence-based sensor that can measure the mixing of relatively clean waters such as potable and recycled water.

[00178] Embodiments of the exemplified methods, sensor, and system described herein may be particularly useful for assessing water parameters for ultimate use within consumer premises. For instance, the method could compare water quality between the point of supply (e.g. smart water meter) and point of use (e.g. kitchen sink), with the intention of communicating the finding and thus detect cross-connections.

[00179] For this invention, various potable and recycled ‘Class A’ recycled waters sourced from different suppliers were characterised. The ability of spectroscopic methods to distinguish between water quality types was evaluated. It was found that UV fluorescence, as opposed to UV absorbance, was considerably more sensitive in being able to distinguish between potable and recycled water.

[00180] The preferred sensor prototypes described herein, with reference to Figure 1, were developed with the view of being affordable in-home devices, as opposed to expensive, analytical instruments. The total cost of the components for the prototype device were kept to under AU$200, and with production scaling would be expected to the further reduced in cost. The physical assembly of fluorescence sensor prototypes involved appropriate selection of UV LED excitation, filtering, microprocessor integration, signal amplification and conditioning, and temperature control.

[00181] The sensitivity of the fluorescence sensor prototypes was demonstrated in controlled laboratory tests, showing that recycled water contamination less than 1% could be detected. The practical limits of the sensor were subsequently explored in two field trials. From the data acquired during both field trials (refer to Table 2 and Table 5), it is seen that for recycled water contamination > 6.3 %, over 99% of all cross-connections were detected using the paired T-test calculation. The data analysis revealed that the number of sampling points does have an impact on the valuation of the statistical outcomes, particularly for differentiating between waters of different qualities. Consequently, a minimum of 10 data sampling points are including in statistical calculations can increase the overall efficacy for determining cross-connections, particularly at lower levels of recycled water contaminations.

[00182] It was demonstrated that Electrical Conductivity (EC) and fluorescence can be complementary approaches for detecting contamination of potable water with recycled water, in view of the varying water quality parameters of the potable water from the two different field trial sites. That is, EC was found to be particularly effective at Facility 1 whereas fluorescence was found to be particularly effective at Facility 2. Given that molecules and dissolved organic matter that alter the fluorescence and EC will vary in their concentration across different systems, a combination of such approaches may be potentially advantageous in some circumstances.

[00183] The data obtained suggested that ongoing calibration of cross-connection devices (both EC and fluorescence) could be particularly beneficial but the extent and frequency of calibration will depend upon various factors including variation in temperature and water quality, and decay in the excitation source for fluorescence measurements.

[00184] Notably, the fluorescence sensors did not experience any catastrophic failures or experience problems associated with biofouling. However, following the use of a fluorescence sensor for over six months of continuous operation, its excitation intensity was found to decrease slowly. In practical scenarios, the sensor would not operate continuously and measures can be input into the device to ensure constant excitation intensity (through a photodiode feedback loop), as for the preferred device of the invention described herein with reference to Figure 1. There are also various means to provide an ongoing calibration to the device if required, including a metered injection of a fluorescent standard or via evaluation of the sensor to provide a regular operational baseline.

[00185] In general, the invention has great potential to provide an excellent means of differentiating water quality over long periods of time at low cost.

[00186] In this specification, the terms ‘comprises’, ‘comprising’, ‘includes’, ‘including’, or similar terms are intended to mean a non-exclusive inclusion, such that a method, system or apparatus that comprises a list of elements does not include those elements solely, but may well include other elements not listed.

[00187] The above description of various embodiments of the present invention is provided for purposes of description to one of ordinary skill in the related art. It is not intended to be exhaustive or to limit the invention to a single disclosed embodiment. As mentioned above, numerous alternatives and variations to the present invention will be apparent to those skilled in the art of the above teaching. Accordingly, while some alternative embodiments have been discussed specifically, other embodiments will be apparent or relatively easily developed by those of ordinary skill in the art. The invention is intended to embrace all alternatives, modifications, and variations of the present invention that have been discussed herein, and other embodiments that fall within the spirit and scope of the above described invention.

[00188] The disclosure of each patent and scientific document, computer program and algorithm referred to in this specification is incorporated by reference in its entirety.

Claims (26)

1. CLAIMS:

   1. A method for detecting contamination of a water sample obtained from a water source, the method including the steps of: (i) assessing fluorescent emission of a water sample in response to fluorescent excitation of said water sample; and (ii) comparing the result obtained in (i) with a reference, wherein a difference in fluorescent emission from the water sample as compared to the reference indicates contamination of the water sample.
2. 2. The method of claim 1, wherein the water sample is of potable water.
3. 3. The method of claim 1 or claim 2, wherein the contamination is contamination with recycled water, or preferably Class A recycled water.
4. 4. The method of any one of claims 1-3, where the difference in fluorescent emission of the water sample as compared to the reference is an increase in fluorescent emission as compared to the reference.
5. 5. The method of any of the preceding claims wherein the reference is obtained by assessing fluorescent emission of a reference water sample in response to fluorescent excitation of said reference water sample.
6. 6. The method of claim 5, wherein the reference water sample is obtained from the same water source as the water sample.
7. 7. The method of any of the preceding claims, wherein the wavelength range of the fluorescent emission assessed is 350 - 550 nm, or preferably 400 - 500 nm.
8. 8. The method of any of the preceding claims, wherein the wavelength range for fluorescent excitation is 200 - 500 nm, 250 nm - 400, or preferably 310 - 370 nm or 270 - 300 nm.
9. 9. The method of any of the preceding claims wherein the steps of (i) assessing fluorescent emission of the water sample in response to fluorescent excitation of said water sample; and (ii) comparing the result obtained in (i) with a reference, includes a plurality of water sample to reference comparisons.
10. 10. The method of claim 9, wherein said method includes at least 3, at least 6, or at least 10 water sample to reference comparisons.
11. 11. The method of any of the preceding claims, wherein said method is capable of detecting contamination of the water sample at a minimum concentration of 10%, 5%, 2%, 1% or 0.5%.
12. 12. The method of any of the preceding claims, wherein the water sample is located within a pipe, preferably wherein said pipe is directly connected to a domestic residence or dwelling.
13. 13. The method of claim 12 wherein the reference water sample is located within a pipe wherein the water sample does not flow into said pipe, preferably wherein said pipe is directly connected to a domestic residence or dwelling.
14. 14. The method of any of the preceding claims, wherein the step of (ii) comparing the result obtained in (i) with a reference, includes calibration or adjustment of said result obtained and/or said reference.
15. 15. The method of claim 14, wherein said calibration or adjustment includes calibration or adjustment based on the temperature of said water sample and/or the temperature of a reference water sample.
16. 16. The method of any of the preceding claims, wherein said method includes the further steps of (iii) assessing electrical conductivity of the water sample; and (iv) comparing the result obtained in (iii) with a reference, wherein a difference in electrical conductivity of the water sample as compared to the reference indicates contamination of the water sample.
17. 17. A device for assessing fluorescence emission in a water sample, the device comprising: (i) a fluorescence excitation source; and (ii) a fluorescence detector, wherein (i) is adapted to induce fluorescent excitation of a water sample, and (ii) is adapted to detect fluorescence emission from the water sample in response to excitation from (i).
18. 18. The device of claim 17, further comprising an interface operatively connected to (i) and (ii).
19. 19. The device of claim 17 or claim 18, wherein said fluorescence excitation source and said fluorescence detector are radially oriented, preferably wherein said fluorescence excitation source and said fluorescence detector are oriented at approximately 90 degrees.
20. 20. The device of claim 19, wherein said device is adapted for use in or on a pipe.
21. 21. A system for detecting contamination of a water sample, the system comprising: (i) at least one device of any one of claims 17-20; (ii) a processor operatively linked to (i) and capable of determining if a change in fluorescent emission of the water sample in response to fluorescent excitation, as compared to a reference, has occurred; and (iii) an output operatively linked to (ii) and capable of notifying a user that contamination of the water sample has occurred, in response to detection of a change in fluorescent emission of the water sample by (ii).
22. 22. The system of claim 21, wherein the processor of the system is operatively linked to at least two devices of the second aspect, and wherein: one of said devices is operatively connected with a water sample; and another of said devices is operatively connected with a reference water sample.
23. 23. The system of claim 22, wherein the one of said devices is located in or on a pipe containing the water sample, and the another of said devices is located in or on a pipe containing the reference water sample.
24. 24. A method of installing the device of any one of claims 17-20, said method including the step of operatively connecting the device to a water sample, or a reference water sample, to thereby install the device.
25. 25. A method of installing the system of any one of claims 21-23, said method including the step of operatively connecting said system to a water sample and/or a reference water sample, to thereby install the system.
26. 26. Use of the device of any one of claims 17-20, or the system of any one of claims 21 -23, to detect contamination of a water sample.