AU2008255622A1 - Determining chemical oxygen demand in water samples - Google Patents

Description

WO 2008/144808 PCT/AU2008/000735 DETERMINING CHEMICAL OXYGEN DEMAND IN WATER SAMPLES This invention relates to a method of determining chemical oxygen demand of water and in particular to a method of dealing with the analysis of samples containing difficult to oxidise organic compounds. 5 Background to the invention The analytical determination of organic material in water and wastewater represents one of the most important measurable parameters for water quality. Nearly all domestic and industrial wastewater effluents contain some level of 10 organic compounds. Even at low concentrations these compounds can cause a detrimental oxygen depletion (or demand) in waterways into which the effluents are released. This demand is due largely to the oxidative biodegradation of organic compounds by naturally occurring microorganisms, which utilize the organic material as a source of food. In this process, carbon is oxidised to C02, while 15 oxygen is consumed and reduced to water. Generally, oxygen demand is used to collectively represent the organic pollution level of a water sample. Chemical Oxygen Demand (COD) is one of the standard analytical methodologies that in widespread use for the determination of oxygen demand in waters and wastewaters. The standard COD method is the preferred 20 approach for assessing the oxygen demand of organic pollutants in heavily polluted or toxic water bodies and is the national standard for organic pollution evaluation in many countries. Despite this, the method has several drawbacks. It requires a time consuming (2-4 hrs) reflux process to allow for complete oxidation of organics. The method is incapable of quantitatively determining COD 25 concentrations below 10ppm. In addition, the method requires expensive (e.g. Ag 2 SO4), corrosive (e.g. concentrated H 2

SO

4 ) and highly toxic (Hg(II) and Cr(VI)) reagents. The latter is of particular environmental concern and has led to the Cr(VI) method being abandoned in Japan. Because of these drawbacks, the standard COD methodology is difficult to be incorporated into any on-line or automated 30 system for rapid monitoring applications. In order to meet the needs of online environmental monitoring and industrial processing control, a huge research effort has been directed to the development of rapid and accurate COD methods.

WO 2008/144808 PCT/AU2008/000735 2 Two different approaches have been widely employed to develop rapid COD methods. One is based on the modification of standard COD methods and the other is to develop totally new analytical principles. Various modifications of the standard method have been carried out by employing 5 alternative oxidising agents, oxidation methods, detection methods and system automation. Despite this extensive effort, modification of standard methods, to date, has achieved very little in a practical sense, which ha led researchers in the field to realise that the inherent problems of the standard methods cannot be easily overcome by simple modification. Since then, development of totally new analytical io principles for the determination of COD has become a main focus. The majority of these new developments are based on either the electrochemical or photocatalytic degradation principles. The electrochemical method employs electrochemical means to oxidise the organics at metal oxide electrodes such as PbO 2 , PbO 2 /Pt, PbO 2 /Ti, SnO 2 /Ti, 15 AgO/CuO solid composite and CuO/Cu electrodes. For these methods, the electrochemical current is measured as the analytical signal to quantify COD values. Advantages of the method include short assay time, wide linear range, easy automation and low instrumental cost. However, high and unstable background current has been shown to inhibit sensitive detection and to cause 20 problems for accuracy and reproducibility. In addition, limited oxidation power makes the method highly matrix dependent. Thus these methods are only capable of oxidising a small fraction of easily oxidisable organic compounds such as glucose. Application W02004/088305 discloses a photo-electrochemical method of 25 detecting chemical oxygen demand as a measure of water quality using a titanium dioxide nanoparticulate semiconductor electrode. Titanium(IV) oxide (TiO 2 ) has been extensively used in the photo-oxidation of organic compounds. TiO 2 is non photocorrosive, non-toxic, inexpensive, relatively easily synthesised in its highly active catalytic nano-particulate form, and is highly efficient in photo-oxidative 30 degradation of organic compounds. W02007/016740 discloses an improvement in this method to deal with interference created by the presence of chloride ions in a sample solution.

WO 2008/144808 PCT/AU2008/000735 3 With these methods, the extent of degradation of organic pollutants in a water sample is measured by directly quantifying the extent of electron transfer at a TiO 2 nanoporous film electrode during an exhaustive photo-electrochemical catalytic degradation process. The photo-electrochemical method is a direct and absolute 5 method that does not require ongoing calibration. The method, in principle, measures the theoretical COD value, due to the extraordinary high oxidation efficiency and the accuracy of signal measurement. This new approach overcomes many problems associated with existing oxygen demand techniques (e.g., the matrix effect, one of the serious practical problems that most rapid COD methods 10 suffer due to insufficient oxidation efficiency). The method is environmentally friendly, robust, and easily automated. It requires only 0.5-5 min to complete an assay and consumes a very limited amount of reagent (electrolyte only). The method is capable of accurately determining as low as 0.2ppm COD. Although the method has been validated to be suitable for a wide spectrum of individual organic 15 compounds, including their mixtures and has been tested with large numbers of real samples collected from various natural and industrial sites, the determination of COD for complex organic compounds such as nicotinic acid and urea, which cannot be determined by the standard COD method, appear to be also difficult for the photo-electrochemical method, due to insufficient degradation efficiency. 20 Illuminated TiO 2 does possess superior oxidation power capable of mineralizing organic species via a photo-electrochemical catalytic degradation process. Thus when TiO 2 is illuminated with photons whose energy is equal to or greater than the band-gap energy (Eg), an photoelectron (e~) will be promoted from the valence band (VB) to the conduction band (CB), creating a photohole (h') in the valence 25 band: TiO+hv- h+ecb T2 vb cb This photohole is a very powerful oxidizing agent (+3.1V) that readily leads to the cession of an electron from the photohole scavenger (e.g. organics). In this way, direct oxidation of organic material can take place (Eqn 2). The attractiveness of 30 this form of degradation is that complete oxidation of organic contaminants can be achieved in a rapid, simple and efficient manner. In contrast, the photoelectron is a relatively weak reducing agent and has to be removed or consumed in a timely WO 2008/144808 PCT/AU2008/000735 4 fashion, otherwise recombination with a photohole will occur (see Eqn 3), leading to low photo-efficiency. With a photo-electrochemical system, the applied potential bias moves the photoelectron to the external circuit and then to the counter electrode where it is consumed via a reduction process. Considering that 5 photoelectrons which are drawn to the external circuit all originated from photocatalytic degradation of organics (and water), thus the measured photocurrent (or charge) is an effective analytical signal for the quantification of organic content. vb + R-H(ads)--R*(ads)+H, b + H2O(ads)--+ OH*(ads) + H+ (2) 10 h± + e --- heat (3) vb cb In theory, an illuminated TiO 2 nanoparticulate layer should be able to mineralize all dissolvable organic species found in natural waters and wastewaters. However, organic compounds such as nicotinic acid and urea that cannot be fully oxidised by 15 standard COD methods have also been found to be difficult to fully oxidise by means of photo-electrocatalysis at the TiO 2 surface. While the standard methods cannot fully oxidise these compounds because of insufficient oxidation power, for TiO 2 photo-electrocatalysis, incomplete oxidation must be due to other reasons than oxidation power as there is a 3.1 V oxidation potential at the photohole in 20 TiO 2 . In theory , this oxidation potential should thermodynamically oxidise nearly all dissolvable organic species. It is an object of this invention to provide a photo-electrochemical method that is able to deal with very difficult to oxidise organic compounds such as nicotinic acid or urea. 25 Brief description of the invention To this end the present invention provides a method of determining chemical oxygen demand in water samples that contain difficult to oxidise organic compounds in which the samples are diluted and a known quantity of an easily 30 oxidised organic substance is added to the diluted sample which is then subjected to an assay by a photo-electrochemical method using a titanium dioxide nanoparticulate semiconductor electrode and measuring the net charge to WO 2008/144808 PCT/AU2008/000735 5 determine chemical oxygen demand. The measurement of the chemical oxygen demand is preferably made in accordance with previously described method in W02007/016740. The net charge (Qnet) due to the photo-catalytic degradation of the difficult to 5 oxidise organic compounds, such as nicotinic acid or urea, can be obtained by deducting the Qadde from Qtotai; that is: Qnet = Qtotal - Qadded (Q 1) The COD value of the difficult to oxidise organic compounds in the original sample can be readily obtained by substitution of the measured Qnet into the COD 10 Equation: COD (mg / L of 02) = net x32000 (Q 2) 4FV where F is a constant and V is known sample volume However the method of this invention utilises a synergetic photo-catalytic effect to improve the oxidation efficiency of the difficult to oxidise organic compound. A ratio 15 of the difficult to oxidise organic compound and the added organic compound is greater than one (1) preferably above 5 to provide sufficient synergetic degradation effect to enable photo-catalytic determination of COD for difficult to oxidise organic compounds. A preferred added organic compound is glucose. This invention is, in part, predicated on the insight that incomplete mineralisation of 20 the difficult to oxidize compound in the photo-electrochemical method could not have been caused by insufficient thermodynamic oxidative power. Therefore it has been hypothesized that the partial oxidation could be due to the impact of slow kinetics or to a different mechanistic pathway being involved Complete mineralization of the simplest organic compounds, such as formic acid 25 and oxalic acid requires the least number of electrons to be transferred (2e-). As an elementary reaction involving two electrons rarely occurs, therefore, the mineralization of any organic compound will involve multiple steps (i.e. elementary reactions) and produce intermediates that then require multiple electron transfer steps. Consequently, photo-catalytic mineralization of different organic compounds 30 can be markedly different from one to another due to the difference in the chemical structures of different compounds, the number of electrons required for the mineralization, the number of steps and intermediates, and the chemical WO 2008/144808 PCT/AU2008/000735 6 properties of each of the intermediates. In fact, an actual reaction pathway or mechanism is determined, to a large extent by the chemical, photochemical, thermodynamic and kinetic properties of the various intermediates produced during phot-electrochemical degradation of the original compound being 5 measured. The reaction intermediates are produced at greater concentration than the original molecules, as soon as photo-catalysis begins. As the chemical, photochemical, thermodynamic and kinetic properties of the intermediates can differ markedly from the original molecule, the subsequent reaction steps are then dictated by the properties of each of the photocatalytically generated intermediates 10 produced. Detailed description of the invention The drawings illustrate aspects of the invention. Figure 1 shows possible reaction pathways; 15 Figure 2 illustrates results from a standard dichromate COD determination of Nicotinic acid (a) comparison of measured COD value against the theoretical COD values in the original sample solutions and (b) correlation between the measured and theoretical COD values; Figure 3 illustrates COD determination for nicotinic acid by the method of this 20 invention (a) comparison of measured COD value against the theoretical COD values in the original sample solutions and (b) correlation between the measured and theoretical COD values; Figure 4 illustrates the COD determination of nicotinic acid by the method of this Invention in which NH 3 is assumed as the final degradation product, (a) 25 comparison of measured COD value against the theoretical COD values in the original sample solutions and (b) correlation between the measured and theoretical COD values; Figure 5 illustrates the COD determination of nicotinic acid by the standard dichromate method in the presence of glucose, (a) comparison of measured COD 30 value against the theoretical COD values in the original sample solutions and (b) correlation between the measured and theoretical COD values; Figure 6 illustrates the COD determination of nicotinic acid by the method of this invention in the presence of glucose(a) comparison of measured COD value WO 2008/144808 PCT/AU2008/000735 7 against the theoretical COD values in the original sample solutions and (b) correlation between the measured and theoretical COD values; Figure 7 illustrates results from a standard dichromate COD determination of urea, (a) comparison of measured COD value against the theoretical COD values 5 in the original sample solutions and (b) correlation between the measured and theoretical COD values; Figure 8 illustrates the COD determination of urea by the method of this Invention, (a) comparison of measured COD value against the theoretical COD values in the original sample solutions and (b) correlation between the measured 10 and theoretical COD values; Figure 9 illustrates the COD determination of urea by the standard dichromate method in the presence of glucose, (a) comparison of measured COD value against the theoretical COD values in the original sample solutions and (b) correlation between the measured and theoretical COD values; 15 Figure 10 illustrates the COD determination of urea by the method of this invention in the presence of glucose, (a) comparison of measured COD value against the theoretical COD values in the original sample solutions and (b) correlation between the measured and theoretical COD values. Figure 1 illustrates three possible reaction pathways, of which two would lead to 20 incomplete mineralization. Path I can lead to the complete mineralisation. Under this circumstance, the photo catalytically produced intermediates do not prohibit further photo-catalytic degradation and so the reaction can proceed to completion -i.e. the complete mineralisation. 25 With Path II, some photo-catalytically produced intermediates (i.e. Rj) exhibit good photo-catalytic degradability and such fractions will proceed to complete mineralisation. However, some photo-catalytically produced intermediates (i.e. R;*) may occur at a certain stage of the photo-catalysis process and possess very slow kinetics. Once this occurs, further mineralisation of such a fraction will not occur 30 within the experimental time frame. As a result overall, a partial mineralisation is observed. Another possible scenario is illustrated by Path Ill. In the same manner as indicated for Path II, some photo-catalytically produced intermediates (i.e. R;*) are WO 2008/144808 PCT/AU2008/000735 8 readily photo-catalytically degraded to carbon dioxide and water (i.e. complete mineralisation). Nevertheless, some intermediates (i.e. Rj) produced at a certain stage of the photo-catalysis process may then undergo chemical or photo-chemical reactions that produce intermediates (or by-products) that then inhibit the TiO 2 5 photo-catalyst, leading to incomplete mineralisation. For example, it is possible that certain types of photo-catalytically produced intermediates (i.e. R;) may react with each other or may react with original molecules or with other intermediates triggering a chain reaction that produces polymeric by-products which are either difficult to further oxidise or which may poison the photo-catalyst surface (i.e. 10 inhibition). As a consequence, the overall reaction may be terminated (due to the inhibition effect of the polymeric by-products) or only partially proceed, and so result in an incomplete mineralisation. To overcome this problem the present invention proposes the synergetic oxidation to overcome the incomplete mineralisation problem. The synergetic principle is 15 illustrated below. As indicated above, reactions 1 b and 1 c may occur when the photo-catalytic oxidation of organic compound (A) alone, which may lead to incomplete mineralisation. RA + h* -+ RAi + RAj + h* -+ CO 2 + H 2 0 (Complete mineralisation) (1a) 20 RA + h* RA;* + h* (Slow kinetics) (1b) RA + h* - + R| -+ R-R; + Rj' - R-R-Rj' - -+ (Polymerisation) (1c) When photo-catalytic oxidation is conducted with a sample solution consisting of more than one organic compound , i.e. compounds (A) and (B). The primary photo-catalytic oxidation will occur as shown in Equation 2 and a new stream of 25 intermediates (RB,*, RB'... RBI) will be produced, in addition to those produced by photocatalytic oxidation of compound (A) (i.e. RAI, RAj). RA + RB + h+ - Ri' + RA;' + RBi+ + RB' +... + RBj (2) As most of these primary intermediates are highly reactive, reactions between the two stream of intermediates generated from (A) and (B) are likely to occur. In fact 30 this may provide a new reaction pathway, leading to the complete mineralisation. For example, a reactive intermediate, i.e. RBj', that is generated from photo catalytic oxidation of (B), can then react with the intermediates, RA;* (which otherwise have slow photo-catalytic oxidation kinetics, as indicated in Path II) or WO 2008/144808 PCT/AU2008/000735 9 with RAj (which otherwise would trigger a chain reaction to produce polymeric by products, as indicated in the Path 111) to produce RA-RBj or RBJ-RAj (see reactions 3a and 3b). RBj' + RA-+ RAr-RBj + h+ -- ...- CO 2 + H 2 0 (3a) 5 RBj + RAj-* RBj-RAj+ h-+ ...- * CO2 + H20 (3b) The properties of these intermediates (RA-RBj and RBJ-RAj) produced from the sample containing both (A) and (B) can be dramatically different to those of produced from the sample containing (A) only (i.e. RAi* and RA;). If the intermediates (RAr-RB and RBJ-RAj) produced by the subsequent reactions are 10 readily further photo-catalytically oxidised to carbon dioxide and water, then by adding a known amount of a different organic compound to a difficult to oxidise sample the synergetic photo-catalytic mineralisation of both can be achieved. In other words, a difficult to oxidise organic compound may be synergetically mineralised by simply adding another organic compound. 15 Quantification of analytical signal The photoelectrochemical (PeCOD) determination of the COD value of a sample containing difficult to oxidise species such as nicotinic acid and urea via a synergetic oxidation process requires the addition of known amounts of easy oxidisable organics such as glucose into the original sample. In case of a normal 20 PeCOD assay, the Qnet is used to quantify the COD value. It can be obtained by deducting the background charge originated from the photo-catalytic oxidation of water from the total charge generated from photo-catalytic oxidation of both organics and water. However, such a Qnet quantification method cannot be used when the COD value of a difficult to oxidise sample has to be determined via a 25 synergetic oxidation process, due to the requirement for addition of easy oxidisable organic to the original sample to achieve an effective assay. Under such conditions, the analytical signal quantification can be achieved via the method illustrated in Figure 11: The photocurrent-time profile of the sample containing the known amount of easy 30 oxidisable organic compound (e.g., glucose) is firstly obtained by performing a PeCOD assay. The charge (Qadded) originating from photocatalytic oxidation of water and photocatalytic mineralisation of the easily oxidisable organic compound can be obtained by integrating the photocurrent with time, using the steady-state WO 2008/144808 PCT/AU2008/000735 10 (baseline) photocurrent as the integration reference line (see Figure 1 1a). Thus the integration reference line is obtained by offsetting the steady-state photocurrent to zero. The photocurrent-time profile of the original sample containing the difficult to oxidise organic compounds (e.g., nicotinic acid or urea) with the added easily 5 oxidisable organic compound of known amount (e.g., 20ppm COD equivalent glucose) is then obtained. The total charge, Qtotai, originating from photo-catalytic oxidation of water, the photo-catalytic mineralisation of easy oxidisable organic compound (i.e. glucose) and the oxidative photo-catalytic degradation of the difficult to oxidise organics (e.g., the nicotinic acid or urea) can be obtained by 10 integrating the photocurrent with time, using the steady-state (baseline) photocurrent as the integration reference line (see Figure 1 1b). The Qnet obtained, which is simply due to the photo-catalytic degradation of the difficult to oxidise organic compounds, such as nicotinic acid or urea, can then be obtained by deducting the Qaddde from Qtotai; that is: 15 Qnet = Qtotal - Qadded (O 1) Thus the COD value of the difficult to oxidise organic compounds in the original sample can be readily obtained by substitution of the measured Qnet into the COD Equation: COD (mg / L of 02) = net x32000 (Q 2) 4FV 20 where F is a constant and V is known sample volume Example 1 COD Determination of Pure Nicotinic acid Standard method The COD values of samples containing known concentrations of nicotinic acid 25 were firstly analysed by the standard dichromate method. It was found that the measured (determined) COD values fluctuated between 5.1ppm and 10.7ppm when the theoretical COD value was increased from 19.2ppm to 96ppm (see Figure 2a). It has been well established that the accuracy and reliability of the standard method is highly questionable when the sample COD value is below 30 1Oppm. As the determined COD values for all samples investigated were close to or below 1 Oppm and by considering the accuracy of the standard method at these low concentrations, the obtained data suggests that the measured COD values are WO 2008/144808 PCT/AU2008/000735 11 not indicative of the original concentration of nicotinic acid. The percentages of the determined COD value are plotted against the theoretical COD values and shown in Figure 2b. It is notable that the percentage of the determined COD decreased almost linearly with the concentration of nicotinic acid in the original sample 5 (theoretical value), as the total amount of oxidised nicotinic acid was independent of original concentration. Based on this data, it can be concluded that the standard dichromate method is incapable of accurately determining a COD value for a water sample containing nicotinic acid due to insufficient oxidation efficiency. Method of W020041088305 10 The COD values of samples containing known concentrations of nicotinic acid were then analysed by the method of W02004/088305 (see Figure 3). The measured COD values linearly increased with the theoretical COD values of the original samples as the concentration of nicotinic acid is increased from 4.8ppm to 48ppm (see Figure 3a). This observation differs from the case of the standard 15 method where the measured COD values are independent of the theoretical COD values of the original samples. A comparison of the percentages of the determined COD value against the theoretical COD values is shown in Figure 3b. The measured COD percentage decreased from 99.5% to 66% as the concentration of nicotinic acid was increased from 4.8ppm to 48ppm equivalent COD. In addition, 20 the measured COD percentage is noted to be much higher than for the standard method, indicating a higher oxidation efficiency. At low concentration, nearly 100% oxidation is achieved. It should be recognized that the theoretical COD values of the original samples shown in Figure 4 were calculated based on Equations 4 and 5, on the basis that 25 the full oxidation of 1 mole of nicotinic acid will consume 7.5 moles of 02 or the loss of 30 moles of electrons, assumimg that the oxidation product of N in nicotinic acid is NO 3 ~. 4 H + 02 + 4e- -* 2H 2 0 (4) 2C 6

H

5

NO

2 + 1502 -> 12 CO 2 + 4H20 + 2HN0 3 (5) 30 However, with the standard dichromate method, the oxidation product of N in nicotinic acid is NH 3 and therefore the oxidation half reaction of nicotinic acid during the COD detection should be given as: 2C 6

H

5

NO

2 + 1102 -+ 12 CO 2 + 2H 2 0 + NH 3 (6) WO 2008/144808 PCT/AU2008/000735 12 Under this circumstance, the oxidation of 1 mole of nicotinic acid will consume 5.5 moles of 02 or, in other words, will require the loss of 22 moles of electrons. Figure 4 shows the relationship between the theoretical and measured COD values when the theoretical COD values are calculated based on NH 3 as the final 5 degradation product of N in the original sample. It appears that the measured COD values were higher than that of theoretical COD values when the nicotinic acid theoretical concentration was below 18ppm COD (see Figure 4a). At a higher concentration, however, the measured COD value was found to be lower than that of theoretical COD values (see Figure 4a). This can be clearly demonstrated when 10 the percentage of the measured COD is plotted against the theoretical COD values. A measured COD greater than 100% of theoretical is observed when the theoretical concentration was below 18ppm while over 90% was observed at higher concentrations (see Figure 4b). This data suggests that ammonia (or ammonium) is not the solitary final degradation product of the nitrogen in the 15 original sample. This was confirmed by charge balance calculation. At very low concentration (i.e. 3.5ppm COD), the nitrogen in the original sample is stoichiometrically oxidised to nitrate (see Equation 5) rather than to ammonium (see Equation 6). In the medium concentration range (i.e. between 8 to 20ppm COD), the nitrogen degradation products are a mixture of nitrate and ammonium 20 as other experiments have proved that nitrite can be quantitatively oxidised to nitrate should it produced as an intermediate species during the photo-catalytic degradation). However, at high concentration, ammonium was found to be the major nitrogen oxidation product. In summary, the standard dichromate method has been confirmed to be incapable 25 of accurately determining COD values of pure nicotinic acid due the relatively low degradation efficiency of the method. The degradation efficiency is markedly improved when the method disclosed in W02004/088305 is used. However, while the method is capable of directly determining the COD of low concentration nicotinic acid, it is unsuitable for higher concentrations, due to incomplete 30 degradation and to the different oxidation products of nitrogen that occur as a function of nicotinic acid concentration. Example 2 COD Determination of Mixed Glucose/Nicotinic acid WO 2008/144808 PCT/AU2008/000735 13 As proposed above, a difficult to oxidise organic compound may be synergetically mineralised by adding another organic compound (see Equations 3a and 3b). In this section, glucose was selected as a suitable synergetic organic compound to be added into a water sample containing nicotinic acid. The concentration ratios 5 between the added glucose and nicotinic acid (in terms of equivalent COD) were varied in order to enable an evaluation of the synergetic degradation effect. Standard method Figure 5 shows the results of COD determination for nicotinic acid in a solution mixture with glucose measured by the standard dichromate method. The data were 10 obtained from various solution compositions (i.e. at different theoretical COD ratios between nicotinic acid and glucose) with different original theoretical COD concentrations of nicotinic acid of 19.2ppm, 24ppm and 48ppm. It was found that when a 19.2ppm nicotinic acid original theoretical COD concentration was used, a change in the ratio between nicotinic acid and glucose from 1:0 to 1:5 resulted in a 15 decrease in the percentage of the undetermined nicotinic acid from 77% to 64%. However, when a 24ppm nicotinic acid original theoretical COD concentration was used, a change in the ratio between nicotinic acid and glucose from 1:0 to 1:10 showed no significant positive effect on the percentage of the undetermined nicotinic acid. With a 48ppm nicotinic acid original theoretical COD concentration, 20 no significant positive effect on the percentage of the undetermined nicotinic acid was observed when the ratios between nicotinic acid and glucose were changed from 1:0 to 1:5. Based on these observations, it can be concluded that with the standard COD method, glucose addition has no significant effect on the nicotinic acid degradation efficiency. 25 Method of this invention Figure 6 shows the results of COD determination for nicotinic acid in a solution mixture with glucose measured by the method of this invention. The data was obtained from various solution compositions (different theoretical COD ratios between nicotinic acid and glucose) with different original theoretical COD 30 concentrations of nicotinic acid of 4.8ppm, 12ppm and 24ppm. As mentioned above it was found that nearly 100% of nicotinic acid can be determined without the need for adding glucose when the nicotinic acid original theoretical COD concentration is low (i.e. 4.8ppm). With 12ppm nicotinic acid original theoretical WO 2008/144808 PCT/AU2008/000735 14 COD concentration, a change in the ratio between nicotinic acid and glucose from 1:0 to 1:10 results in a decrease in the percentage of the undetermined nicotinic acid from 7% to 1% ( i.e. 99% nicotinic acid being determined). However, with a 24ppm nicotinic acid original theoretical COD concentration, when the ratio 5 between nicotinic acid and glucose was changed from 1:0 to 1:10, the percentage of the undetermined nicotinic acid was initially increased at 1:2 and then decreased to 3% at 1:10. These results clearly demonstrated a synergetic effect of glucose on the photocatalytic degradation of nicotinic acid, especially at higher concentrations of 10 nicotinic acid in water. It can be concluded that when an appropriate ratio of glucose is added to a water sample containing nicotinic acid, a close to 100% of the nicotinic acid concentration can be determined. Example 3 COD Determination of Pure Urea In this work, urea is also selected as a testing compound due to its unique 15 chemical composition. It is an organic compound having with a very high nitrogen to carbon ratio (2:1). This relatively unique feature makes it a very interesting organic compound warranting a detailed investigation. 20 Standard method Figure 7 summarises the COD data for urea as obtained by the standard dichromate method. The measured COD values for different urea concentrations (theoretical COD values based on nitrate as the final oxidation product of nitrogen in the original samples) were all around 5ppm, regardless of the original 25 concentration of urea (see Figure 7a). The percentages of the measured COD values were found to decrease as the concentration increased (see Figure 7b). This data confirmed that a urea sample can be partially degraded by the standard method although the percentage of degradation was very low (i.e. low measured COD values). 30 Urea has been generally regarded as an undeterminable organic compound by the standard dichromate method. This is because urea can dissociate into carbon dioxide and ammonia through a hydrolysis reaction that involves no net electron transfer (see Equation 7).

WO 2008/144808 PCT/AU2008/000735 15

CN

2

H

4 O + H 2 0 -* Co 2 +2 NH 3 (7) Thus with the standard dichromate method, ammonia is generally regarded as the final oxidation product of the nitrogen element in the original compound. As ammonia so produced cannot be oxidised by the standard dichromate method, 5 consequently, urea cannot be determined by that method, because no dichromate is consumed (i.e. there is no electron transfer occurring). If this is true, then all urea samples investigated should give a Oppm COD result. However this does not correspond with the experimental observation shown in Figure 8. It is reasonable to assume that a small fraction of nitrogen in the urea molecule has 10 been oxidised into nitrate rather than ammonia (see Equation 8).

CN

2

H

4 0 + 402 -+ Co 2 + H 2 0 + HNO 3 (8) In this process, one urea molecule loses 16 electrons or consumes 4 oxygen molecules. Hence the non-zero COD results of Figure 8 Method of W020041088305 15 Figure 8 also provides the results of COD determination for urea by the method of W02004/088305. It can be observed in figure 8 that the measured COD values increased as the theoretical COD values of the original urea sample increased (see Figure 8a). This is very different to the case of the standard dichromate method where the measured COD values are independent of the original urea 20 concentration. However, when the theoretical COD concentrations of urea was increased from 4.8ppm to 48ppm, the percentage of the measured COD values initially increased and then level off at a value close to 30% (see Figure 8b). This indicates that around 30% of urea iss reacted in accordance with the reaction shown by Equation 8, where nitrate is the final degradation product. 25 Example 4 COD Determination of Mixed Glucose/Urea Urea,was also investigated as another test case of synergetic degradation of a difficult to oxidise organic compound,. Again, glucose was selected as the synergetic organic compound to be added into the urea sample. The concentration ratios between the added glucose and urea were varied to enable evaluation of the 30 effectiveness of synergetic degradation in te case of urea. Standard method Figure 9 shows the COD data for urea in a solution mixture with glucose, measured by the standard dichromate method. The data were obtained from WO 2008/144808 PCT/AU2008/000735 16 various solution compositions (i.e. for different theoretical COD ratios between nicotinic acid and glucose) with different original theoretical COD concentrations of urea (12ppm, 24ppm, 48ppm and 96ppm). The data of figure 9 reveals that with samples containingl2ppm urea original theoretical COD concentration, an increase 5 in the glucose to urea concentration ratio results in a decrease in the percentage of urea determined with the unmeasured percentage of urea increasing from 64% (1:0) to 75% (1:5). With samples containing 24ppm urea, the undetermined urea slightly increased from 79% to 83% when the ratio was increased from 1:0 to 1:10. With higher urea concentrations such as 48ppm and 96ppm, no significant 10 influence was observed when the urea to glucose ratio changes from 1:0 to 1:5. However, lower urea determined percentages observed for higher concentration urea samples. These observations suggest that with the standard COD method, glucose addition does not facilitate further degradation of urea. Method of this invention 15 Figure 10 shows the COD data obtained by the method disclosed in W02004/088305 from urea solutions containing various concentrations of added glucose. With a 4.8ppm urea original theoretical COD concentration, a change in the ratio between urea and glucose from 1:0 to 1:5 results in a decrease in the percentage of the undetermined urea from 82% to 12% (i.e. 88% urea being 20 determined). With a 12ppm urea original theoretical COD concentration, when the ratio between urea and glucose was varied from 1:0 to 1:5, a decrease in the percentage of the undetermined urea from 71% to 5% (i.e. 95% urea being determined). At the higher 24ppm urea original theoretical COD concentration, 100% urea is determined with a urea to glucose ratio of 1:5. This experimental data 25 demonstrated again that the synergetic effect of glucose on the photocatalytic degradation of a difficult to oxidise organic compoundprovides a useful method for COD determination. When an appropriate ratio of glucose is added to the urea sample, a close to 100% of urea concentration in a water sample can be photo catalytically determined. 30 Conclusions (i) The standard dichromate method has been proven to be ineffective in determining the COD of a water sample containing either nicotinic acid or urea, due to insufficient degradation efficiency of the method.

WO 2008/144808 PCT/AU2008/000735 17 (ii) A small fraction of nicotinic acid or urea can be degraded (determined) by the standard dichromate method with or without the presence of glucose. This demonstrates that ammonia/ammonium may be not the solitary final degradation product of the nitrogen in the original sample of nicotinic acid and urea. Especially 5 for urea, the small fraction of determined COD confirms that at least a small fraction of ammonia/ammonium is further oxidised to nitrate. (iii) The degradation efficiency of the standard dichromate method towards nicotinic acid and urea is not improved by the use of synergetic degradation using an added different organic compound (i.e. glucose). 10 (iv) The method of.W02004/088305 possesses higher degradation efficiency towards nicotinic acid and urea, but alone , it remains insufficient for the effective COD determination of water samples containing nicotinic acid or urea. (v) The degradation efficiency of the method ofW02004/088305 towards nicotinic acid and urea can be dramatically improved by the use of synergetic 15 degradation comprising the addition of a different organic compound (i.e. glucose) to a water sample containing a difficult to oxidise organic compound such as nicotinic acid or urea. (vi) Photocatalytic determination of difficult to oxidise organics, such as nicotinic acid and urea, is feasible when an appropriate ratio of a suitable organic 20 compound such as glucose is added into the original sample. (vii) Under most of circumstances, an effective ratio of 1:5 between the difficult to oxidise organic compound and added(easy to oxidise) organic compound has been shown to provide sufficient synergetic degradation to enable the photo-catalytic determination of COD for difficult to oxidise organic compounds. 25 (viii) With the photo-catalytic COD determination, nitrate has been identified as a final degradation product of the nitrogen in the original sample of nicotinic acid and urea. From the above it can be seen that the present invention provides a unique and simple method of making COD measurements for difficult to oxidise organic 30 compounds. Those skilled in the art will realise that this invention may be implemented in embodiments other than those described herein without departing from the core teachings of this invention.

Claims (6)

1. A method of determining chemical oxygen demand in water samples containing organic compounds that are difficult to oxidize, in which the 5 samples are diluted and a known quantity of an easily oxidized organic substance is added to the sample which is then subjected to an assay by a photo-electrochemical method and the net charge is measured as a measure of the chemical oxygen demand. 10

2. A method as claimed in claim 1 in which chemical oxygen demand is calculated using the formula COD (mg / L of O2) Qnet x 32000 4FV Where Qnet is the total charge measured less the charge due to the added 15 easily oxidized organic substance and F is a constant and V is known sample volume

3. A method as claimed in claim 1 or claim 2 in which the ratio of the difficult to oxidise organic compound to the added organic substance is greater than 20 one.

4. A method as claimed in claim 3 in which the ratio of the difficult to oxidise organic compound to the added organic substance is above five. 25

5. A method as claimed in claim 3 or 4 in which the added compound is glucose.

6. A method as claimed in claim 1 or 2 in which a titanium dioxide nano particulate semiconductor electrode is used as the photo-anode. 30