Water treatment method and equipment

Abstract

The present invention relates to a method for treating fluids, in particular water, comprising a renewal step (40) for restoring the adsorption capacity of an adsorbent bed (22), which comprises extracting (42) an adsorbent sample from the adsorbent bed (22), determining a target average age of the extracted adsorbent sample, in particular by rejuvenating at least a portion of the extracted adsorbent sample (44), where the extracted adsorbent has an actual reduction in pollutants at the target average age corresponding to a predetermined reduction target, and rejuvenating (46) the adsorbent bed (22) to the determined target average age. The present invention makes it possible to improve the treatment of fluids by adsorption and to take into account, in particular, emerging pollutants. The present invention also relates to a plant for treatment by the proposed method.

Description

[Technical Field]

[0001] The present invention relates to a method for treating fluids, in particular water, in particular drinking water (water that is to be made safe for drinking), but also municipal or industrial wastewater, which method comprises an adsorption step, more particularly the step of renewing the adsorbent with a mixed adsorbent of intermediate age determined as a function of a reduction target. [Background technology]

[0002] For the treatment of fluids, in particular for the production of drinking water or for the treatment of wastewater, it may be proposed to reduce the organic pollutants contained in the raw water or wastewater by an adsorption step of this material.

[0003] The increasing load of organic contaminants (natural organic matter and trace contaminants of anthropogenic or natural origin) observed in resources is forcing manufacturers of drinking water and wastewater purification equipment to retrofit treatment facilities that are no longer able to meet quality targets. This increase in organic contaminant load is also causing drinking water manufacturers to need to design new treatment equipment. Finally, wastewater purification equipment, whether for industrial or tertiary source wastewater prior to discharge into the natural environment or for wastewater (e.g., wastewater) that is to be directly or indirectly converted into drinking water, can also benefit from treatment that takes into account higher organic contaminant content. Taking into account this serious contamination by organic contaminants may particularly include the addition of purification equipment during the design or retrofit phase, particularly filtration and / or activated carbon, such as adsorption onto granular activated carbon (GAC) beds.

[0004] For example, from FR3003477, it is known to use activated carbon filters or activated carbon reactors to retain organic matter or other natural or artificial pollutants. FR3003477 in particular proposes the use of an upflow system without substantial expansion of the granular activated carbon bed, as well as a cleaning step for the filter formed by the activated carbon bed. The cleaning is carried out by substantial expansion of the bed. FR3003477 also proposes a periodic renewal step for the activated carbon bed to renew its adsorption capacity.

[0005] In this field of water treatment, consideration has yet to be given to the emergence of organic micropollutants of synthetic origin.

[0006] In particular, some of these emerging contaminants are poorly adsorbable, regardless of whether they are small, polar, or hydrophilic contaminants. They are primarily pesticide metabolites and can therefore be found downstream of adsorption processes, such as those using granular activated carbon. The levels of these emerging contaminants after treatment processes may exceed regulatory limits if they are specifically regulated, or at least represent anticipated risks for emerging contaminants that are not yet regulated.

[0007] However, activated carbon purification plants, even with periodic upgrades, are sized as a function of some traditional pollutants and are not designed to accommodate these poorly adsorbed emerging pollutants.

[0008] Therefore, there is a need for improved adsorption treatment of fluids such as water, with particular consideration of these poorly adsorbable emerging contaminants. Summary of the Invention

[0009] More specifically, the present invention relates to a method for treating a fluid, in particular water such as surface water or groundwater, or even wastewater, comprising a step of adsorption of contaminants contained in the fluid to be treated by an adsorbent bed, the method further comprising a renewal step for restoring the adsorption capacity of the adsorbent bed, the renewal being carried out by: extracting a sample of the adsorbent from the adsorbent bed; determining a target average age of the extracted sorbent sample, in particular by rejuvenating at least a portion of the extracted sorbent sample, at which target average age the extracted sorbent accounts for an actual reduction of contaminants corresponding to a pre-set reduction target; rejuvenating the sorbent beds to a determined target average age; The present invention proposes a method including:

[0010] According to preferred embodiments, the invention includes one or more of the following features: The update process can be triggered (performed) periodically at a predetermined update frequency; The updating step may be triggered after a step of detecting a quality defect of the fluid to be treated downstream of the adsorption step, and preferably the detection step includes detecting the quality defect by comparing the level of contaminants or the number of contaminants between the upstream of the adsorption step and the downstream of the adsorption step; The detection of the quality defect is carried out when carrying out the treatment of the fluid to be treated, and preferably, the detection of the quality defect is carried out based on a measurement method selected from at least one of the measurement methods including chromatography, mass spectrometry and fluorescence spectroscopy; The extraction of the adsorbent is carried out during an interruption stage of the treatment of the fluid to be treated, preferably during a stage of washing the adsorbent, more preferably during a stage of washing the adsorbent by injection of air; The extraction of the sorbent is carried out by sampling at one or more points in the sorbent bed as the treatment of the fluid to be treated is carried out; In determining the target average age of the extracted sorbent samples, the actual reduction of contaminants is determined by sampling the fluid to be treated downstream of the adsorption step, and preferably the fluid to be treated sampled downstream of the adsorption step is doped prior to determining the actual reduction of contaminants; Determining the target average age of the sorbent sample is obtained by measuring the actual reduction of contaminants in a mixture of younger sorbent and at least a portion of the sorbent sample extracted from the sorbent bed, and rejuvenating the sorbent bed to the determined average age is obtained by at least partially replacing the sorbent bed with younger sorbent until the determined target average age of the sorbent bed is obtained; Determination of the target average age of the sorbent sample is obtained by measuring the reduction of contaminants in sorbent subsamples obtained by mixing the extracted sorbent with younger sorbent in various proportions, the collection of subsamples having an average age range intermediate between the age of the extracted sorbent and the age of the younger sorbent, and the average age of the subsample that shows the actual reduction of contaminants that best corresponds to the pre-set reduction target determines the target average age; The mean intermediate age of the sorbent subsamples is 2-10, preferably 3-5, and the intermediate ages are preferably evenly distributed between the ages of the extracted sorbent and the ages of the younger sorbents; The one or more measurements of contaminant reduction of the one or more mixtures of adsorbents are obtained by short bed adsorber measurements, the tests being preferably conducted on a bed having a volume of 100 mL or less, more preferably 50 mL or less, and even more preferably 20 mL or less; When determining the target average age of the extracted sorbent samples, the actual reduction of contaminants is determined for a group of contaminants selected from the group formed by desethylhydroxyatrazine, metaldehyde, aminotriazole, metazachlorine, metazachlorine ESA, metallochlorine, metallochlorine ESA, desethylatrazine, chlortoluron, atrazine, terbuthylazine, and optionally for at least one of the one or more contaminants identified as contributing to the quality defects of the fluid being treated that trigger the renewal process; The predetermined reduction target is defined as a minimum reduction value of the concentration of each pollutant in the group of pollutants to be tested, and the minimum reduction value of each pollutant is preferably 50% to 90%, more preferably 60% to 80% or 70% to 80%, and the minimum reduction value may be the same for each pollutant in the group of pollutants to be tested; The adsorbent bed comprises granular activated carbon, the adsorption step being carried out by passing the fluid to be treated, in particular water, through the adsorbent bed in an upflow manner, the method preferably comprising: at least one filtration / adsorption stage, in which the velocity of the fluid is sufficiently low so as not to cause a substantial expansion of the bed of activated carbon ensuring both the filtration and the adsorption of the substances contained in the fluid; at least one expansion stage in which a fluid circulates at a rate sufficient to substantially expand the bed of activated carbon, thereby subjecting the bed of activated carbon to washing with the fluid; Includes:

[0011] A fluid treatment facility using the proposed treatment method is also proposed, and this facility: a reactor for adsorbing contaminants contained in a fluid to be treated, the reactor having an adsorbent bed held therein and an orifice provided in the reactor for at least partially removing used adsorbent from the adsorbent bed; a mixer for mixing the extracted adsorbent sample from the adsorbent bed with young adsorbent; a measuring unit for measuring the actual reduction of contaminants in a mixture of young adsorbent and adsorbent extracted from the adsorbent bed, the measuring unit preferably being a short bed adsorber measuring unit, and the test being carried out on a bed having a volume of preferably 100 mL or less, more preferably 50 mL or less, and even more preferably 20 mL or less; Equipped with. [Brief explanation of the drawings]

[0012] [Figure 1] 1 illustrates an embodiment of a water treatment plant having a contaminant adsorption process.

[0013] [Figure 2] The scheme of the proposed method is shown below.

[0014] [Figure 3] This is the result of measuring the water discharged from the adsorption process by HPLC-HR&MS.

[0015] [Figure 4] This shows the results of measuring the decant water introduced into the adsorption step by HPLC-HR&MS.

[0016] [Figure 5] These are the results of measuring raw water upstream of the decantation process using HPLC-HR&MS.

[0017] [Figure 6] It represents a measure of the actual reduction of contaminants for subsamples of mixtures with three intermediate mean ages.

[0018] [Figure 7] It represents a measure of actual contaminant reduction for subsamples of mixtures with five intermediate mean ages.

[0019] [Figure 8] Represents the SBA test machine.

[0020] [Figure 9] It operates in an upflow manner and represents a bed of granular activated carbon, which represents the Carbazur UP process.

[0021] [Figure 10] Figure 1 shows a comparative reduction test between the SBA and Carbazur UP units at a pilot scale with 50,000 BVT of activated carbon.

[0022] [Figure 11] Figure 1 shows a comparative reduction test between the SBA and Carbazur UP units at a pilot scale with 80,000 BVT of activated carbon.

[0023] [Figure 12] 1 shows a diagram of the proposed fluid treatment facility. DETAILED DESCRIPTION OF THE INVENTION

[0024] The proposed treatment method is intended to remove contaminants contained in a fluid to be treated, which may be water, in particular water to be converted into drinking water, but also municipal or industrial wastewater (in particular leachate, which is liquid wastewater from waste storage facilities) before being discharged into the natural environment, or wastewater to be directly or indirectly converted into drinking water (such as wastewater from municipal wastewater).

[0025] Elsewhere in this specification, the term "pollutant" refers to both organic matter and micropollutants. Micropollutants can be defined as undesirable substances that can be detected in the environment at very low concentrations (micrograms / liter or even nanograms / liter). The presence of micropollutants in water is at least partly due to human activities (industrial processes, agricultural practices, or pharmaceutical or cosmetic residues). Micropollutants are characterized by the fact that, at these very low concentrations, they can affect living organisms due to their toxicity, persistence, and bioaccumulation, or due to organoleptic contamination (taste or odor, which is particularly relevant when water is treated for drinking water). Micropollutants are very numerous (more than 110,000 molecules are listed by European regulations) and diverse. The diversity of these contaminants makes it possible to classify them according to their origin, type, or their highly distinct chemical properties. For example, micropollutants can be of natural origin (e.g., geosmin, methylisoborneol - MIB, or compounds derived from soil decomposition including bacterial residues), plant origin (e.g., algal metabolites including microcystins), animal origin, or human origin. Micropollutants can be classified according to type, such as polar organic compounds abbreviated as POC or metal organic compounds abbreviated as MOC. Micropollutants can have very different chemical properties, such as detergents, metals, hydrocarbons, pesticides, cosmetics, or pharmaceuticals. Therefore, the proposed fluid treatment method is particularly applicable to pesticide-type compounds and related metabolites. This method is also particularly applicable to solvents. This method is further particularly applicable to pharmaceutical residues or residues from industrial activities. Therefore, all these categories of contaminants or micropollutants are particularly relevant to the proposed method.

[0026] The proposed fluid treatment method includes an adsorption step of contaminants contained in the fluid to be treated.

[0027] This adsorption step is carried out using a bed of adsorbent. The adsorbent (or adsorption medium) is, for example, activated carbon, especially granular activated carbon (GAC). The granular activated carbon used in the proposed method has, for example, a particle size of 400-1700 μm for at least 85-90% of the granules, or preferably a particle size of 800-1200 μm for at least the majority (50%) of the granules. The above dimensions refer to the equivalent particle diameters in the case of dry screening (sieving) or wet screening. Granular activated carbon can typically have an iodine value of more than 950 or 1000 mg / g in the fresh state (new condition). The iodine value is the amount of iodine adsorbed in milligrams per gram of activated carbon. This iodine value is used to quantify the adsorption capacity of a given adsorption medium. The measurement of the iodine value can be carried out, in particular, according to a protocol standardized by the standard ASTM D4607-14.

[0028] The treatment method including the adsorption step can correspond to the water treatment method shown in Figure 1. According to the method shown in this figure, The water 12 to be treated is first sampled in step 10. This water 12 is more broadly called source or raw water and is taken, for example, from a waterway as shown. This is a typical example of surface water. The source can also be sampled by drilling, in which case the water is called groundwater. It can also be wastewater of urban origin (such as wastewater also called residual municipal wastewater) or effluent of industrial origin. In this case, the sampled water may be subjected to a pretreatment step by clarification 14, for example comprising a settler or floater, for example by separation by flocculation to retain particulate or colloidal matter, and then optionally by filtration, in particular by sand filtration as shown. The type of pretreatment may depend on the origin of the fluid to be treated. If the source to be treated is groundwater, this pretreatment step may in particular be omitted, as in the case of surface water. For wastewater, it may be advantageous to carry out a pretreatment, for example upstream of clarification, which may also include biodegradation. The water then undergoes an adsorption step 18 , for example by a reactor 20 . After this adsorption step 18 and optional filtration step (not shown), the treated water may be distributed at a distribution step 80 . The proposed method can be schematized according to FIG.

[0029] The effectiveness of the adsorption process 18 is improved by the renewing process 40. The adsorption capacity of adsorbents, such as activated carbon, decreases with use to adsorb contaminants. The media renewing process allows the adsorption capacity of the adsorbent bed to be restored.

[0030] In the proposed method, this renewal step first involves extracting a sample 42 of the adsorbent from the adsorbent bed. This sample, like the rest of the bed, will lose some of its adsorption capacity with further use as a contaminant adsorbent. This loss of capacity is generally approached in terms of productivity, which is often considered to be partially equivalent to the age of the sample. Therefore, the age considered to be equivalent to the productivity of the sample can be quantified in terms of the treated bed volume BVT or bed volume BV. The treated bed volume corresponds to the volume of fluid, more specifically water, to be treated by the adsorbent relative to the volume of the adsorbent. Therefore, it can be inferred that the higher the productivity of the adsorbent, the larger the volume of fluid treated by the adsorbent, and the more worn or aged the adsorbent is, and therefore the adsorbent's adsorption capacity is decreasing.

[0031] Based on this extracted, and therefore aged, sample, updating in the proposed method involves determining 44 a rejuvenated age at which the sorbent sample will be restored to better contaminant reduction. This rejuvenated age is the target average age that should be achieved so that the sorbent exhibits better contaminant reduction than the sorbent sample at the time of its extraction. The target average age is defined as the age at which the extracted sorbent exhibits actual contaminant reduction corresponding to a pre-set reduction target.

[0032] To determine this target average age, at least a portion of the extracted sorbent sample can be subjected to rejuvenation.

[0033] In a preferred embodiment, this rejuvenation is, for example, in the form of a mixture of at least a portion of the extracted adsorbent sample with a young adsorbent. "Young adsorbent" refers to an adsorbent that is younger than the extracted adsorbent sample (simply mixing with this adsorbent does not result in rejuvenation). In particular, this young adsorbent may be a fresh adsorbent, i.e., an adsorbent that has not yet been used as an adsorbent after its production. This young adsorbent may also be a regenerated adsorbent, particularly in the case of activated carbon. A regenerated adsorbent corresponds to an adsorbent that, after a cycle of use as an adsorbent, has been treated, for example, thermally or chemically, to recover an adsorption capacity close to that of a fresh adsorbent. Fresh or regenerated adsorbents have an age that, when expressed in BVT, is considered to be zero. Therefore, as mentioned above, regeneration resets productivity to an age corresponding to zero. Even if regeneration allows for recovery of adsorption capacity, a regenerated adsorbent may nevertheless exhibit a more limited adsorption capacity than the same adsorbent in its fresh state. These more limited adsorption capacities after regeneration can be characterized, for example, by the iodine adsorption value or iodine value. The iodine value is the amount of iodine adsorbed in milligrams per gram of adsorbent and is used to quantify the adsorption power of the adsorbent medium. For example, for fresh adsorbents, the iodine value may be higher than 950 or 1000 mg / g (in the case of preferred activated carbon). Conversely, for used adsorbents, the iodine value may be less than 400 mg / g. Regeneration of the adsorbent can recover an iodine value of preferably more than 600 mg / g, or more preferably more than 700 mg / g. Regeneration may be carried out ex situ by the adsorbent supplier, particularly by reactivation, such as heat treatment above 800°C in the case of adsorbents in the form of activated carbon. The iodine value obtained in this regeneration case may be greater than 800 mg / g, or even greater than 850 mg / g. Regeneration may also be carried out in situ at the treatment facility by chemical or thermal treatment, in particular at temperatures lower than the reactivation at 800° C. This in situ regeneration advantageously makes it possible to recover part of the adsorption capacity (600-800 mg / g iodine number) without necessarily requiring the more restrictive ex situ reactivation.

[0034] After rejuvenating the extracted sample and for the purpose of determining a target average age, the proposed method can include measuring the contaminant reduction by the extracted sorbent rejuvenated by blending. This measurement of contaminant reduction can be performed directly, in particular by comparing contaminant concentrations upstream and downstream of the fluid treatment using the rejuvenated extracted sorbent. Contaminant reduction can also be measured indirectly, for example by measuring the contaminant level using iodine value measurement (as described above), or by chromatography (e.g., as described below, thus in particular HPLC, HPLC-HR, or HPLC-HR&MS), mass spectrometry, or fluorescence spectroscopy. The contaminant level thus determined can then be correlated with the actual concentration of the contaminant, for example, using a predetermined alignment chart for each contaminant.

[0035] This measurement of contaminant reduction by the rejuvenated sorbent bed samples is then compared to a pre-established reduction target, and if this target is achieved, the rejuvenated sorbent bed samples have an average age that forms the rejuvenation target achieved for the entire sorbent bed.

[0036] Once the target average age is determined, the proposed method then includes rejuvenating 46 the adsorbent beds to the determined target average age.

[0037] Rejuvenating an adsorbent bed may include, among other things, at least partially replacing the adsorbent bed with younger adsorbent until a determined target average age for the adsorbent bed is reached. The amount of younger adsorbent added is calculated, for example, using the arithmetic average of the amount of spent adsorbent remaining in the bed and the amount of younger adsorbent to be added to the bed. Typically, the amount of remaining spent adsorbent is reduced before adding the younger adsorbent so that the adsorbent bed has an approximately constant volume.

[0038] This rejuvenation, performed by at least partially replacing the sorbent bed with younger sorbent, can be carried out, in a particularly preferred embodiment, by determining a target average age by mixing a sample with younger sorbent. In this case, the rejuvenation performed to determine the target average age is a smaller-scale reproduction of the rejuvenation of the entire sorbent bed. This example has the advantage of allowing for more precise control of the recovery of adsorption capacity obtained with the rejuvenated sorbent bed.

[0039] In the proposed method for determining a target average age at which to renew a carbon bed, the contaminant reduction of the rejuvenated extracted sorbent is referred to as the “actual reduction of the contaminant.” The reduction is called “actual” in that it is determined based on a sample of sorbent that is effectively used in a treatment method.

[0040] The proposed method differs in this determination of actual reduction from the theoretical evaluation of contaminant reduction by adsorbents proposed in the prior art.

[0041] Some prior art methods propose updating adsorbent beds by partially and periodically replacing them with fresh adsorbent, such as those described in the aforementioned FR3003477 or FR2874913. In these methods, before industrial implementation, a theoretical operational limit for the treated bed volume is determined, which corresponds to a critical mean age above which the theoretical reduction of pollutants is no longer considered satisfactory. Programming the periodicity of the renewal allows for a good theoretical reduction of pollutants, taking into account this limit determined before industrial implementation of the method. The theoretical reduction can be determined, for example, using a homogeneous surface diffusion model (HSDM), especially when the adsorbent is activated carbon.

[0042] However, for the same treated bed volume, two adsorbent beds may exhibit different contaminant reduction capacities depending on the actual variation in contaminant concentration and type actually treated by each of these adsorbent beds. Furthermore, the theoretical adsorption capacity per treated bed volume is generally determined by the contaminant, but does not take into account competition between organic matter and micropollutants for adsorption sites, or competition between micropollutants themselves (also known as the "cocktail effect"). For example, highly adsorbent micropollutants tend to saturate the adsorption sites before less adsorbent micropollutants can be adsorbed. Competition between contaminants for adsorption sites is therefore particularly true for compounds with low affinity for adsorption (low-polarity molecules), such as pesticide metabolites. Therefore, theoretically determining adsorption capacity as a function of treated bed volume may tend to overestimate the remaining adsorption capacity of an actual adsorbent bed.

[0043] In the method proposed herein, an age target for targeted reduction is obtained based on a sample extracted from an industrially operating adsorbent bed. This age target is therefore not theoretically bound by prior testing or modeling of comparable worn (used) adsorbents (e.g., measured in units of bed volume processed). In contrast, in the proposed method, this sample therefore represents the actual wear of the adsorbent bed, taking into account, in particular, the results of competition between contaminants in the fluid for adsorption sites actually processed by the adsorbent. Therefore, determining the target average age from this sample allows for optimal rejuvenation of the adsorbent bed, taking into account the actual history of the adsorbent bed.

[0044] Therefore, the proposed fluid treatment method addresses new constraints arising from novel, difficult-to-adsorb contaminants by controlling the age of the adsorbent to ensure permanently efficient removal of target or non-target contaminants, which is made possible by continuous adsorbent renewal.

[0045] Renewal of the adsorbent, and particularly GAC, can be achieved by extracting a portion of the adsorbent and replacing it with younger adsorbent, which, with respect to GAC in particular, can be fresh or regenerated GAC.

[0046] The continuous renewal of the adsorbent allows for an adapted change in the renewal volume and / or renewal frequency as a function of the variations in the quality of the raw water, in other words, the proposed method makes it possible to control the adsorption treatment method by adjusting the amount of adsorbent to be renewed by monitoring the actual saturation state of the adsorbent.

[0047] The proposed method therefore allows for better treatment of fluids such as water, taking into particular account new adsorbents that are difficult to adsorb.

[0048] In addition, the fact that the sample is extracted from the sorbent bed also makes the proposed method particularly effective in determining the actual reduction of the sample in the fluid to be treated. Thus, in a preferred embodiment, when determining the target average age of the extracted sorbent sample, the actual reduction of contaminants is determined by a sample of the fluid to be treated taken downstream of the adsorption process. The fluid to be treated is also called the matrix, especially in the field of wastewater treatment. The proposed method determines the target average age at which the sorbent bed must be rejuvenated, and this target average age is determined by sorbent and fluid samples, which therefore accurately correspond to the future use of the sorbent bed with the determined target average age.

[0049] Determining the target average age by extracting a sample of the adsorbent also allows for taking into account the actual matrix sample, allowing for further optimized rejuvenation of the adsorbent.

[0050] This optimization is particularly advantageous when compared with prior art methods that determine the remaining adsorption capacity based on a preset model using a matrix that is sampled only during model design and remains fixed thereafter. In particular, EP 3153475 proposes a treatment method using a reactor with an activated carbon bed, in which bed renewal is controlled according to the UV reduction yield, which correlates with trace pollutant reduction, based on a model determined using a matrix sampled before the reactor begins operation. In the above-mentioned prior art methods, the rejuvenation rate of the adsorbent bed is evaluated based on a matrix that no longer corresponds to the matrix actually treated by the adsorbent. Therefore, in the case of emerging pollutants, the above-mentioned methods may no longer be suitable for determining actual pollutant reduction. This incompatibility with emerging pollutants is particularly serious when these emerging pollutants are barely adsorbable, such as pesticide metabolites. In contrast, the method proposed herein, which allows for the determination of a target average age based on the current matrix, allows for the adsorption renewal to be adapted to these pollutants that may be found in the matrix.

[0051] Therefore, the proposed method for the renewal of the sorbent bed, controlled in particular by the extraction of samples of the sorbent, provides a better treatment of fluids such as water, especially taking into account emerging pollutants that can hardly be adsorbed.

[0052] In a preferred embodiment, the renewal process can be triggered upon detection of a quality defect in the fluid to be treated. The proposed method can therefore include a step of detecting a quality defect in the fluid to be treated, after which the renewal process is triggered. This step can be performed downstream of the adsorption process. This downstream detection makes it possible to determine whether or not the adsorbent bed has "breakthrough." The terms bed breakthrough or filter breakthrough are used. In particular, in the field of water treatment, filter breakthrough occurs when the filter allows contaminants or elements that it is intended to retain to pass through. In other words, it is proposed to renew an adsorbent, such as granular activated carbon, by monitoring a quality parameter of the fluid to be treated, such as water, at the outlet of the adsorption filter. When a limit value is reached, renewal of the adsorbent bed is triggered.

[0053] In addition to detecting quality defects, this trigger allows for better consideration of emerging contaminants that are barely adsorbable. The detection of quality defects may also be carried out for target (or predetermined) contaminants. In this case, the detection process specifically targets some predetermined contaminants to be monitored. However, this detection process can also be carried out for non-target contaminants. The fluid to be treated, in particular water, can be subjected to measurements that allow for the determination of the presence of impurities or contaminants without a prior specification of the type of contaminant.

[0054] This non-targeted detection of contaminants in water (or the fluid to be treated) can be carried out by at least one of the following measurement methods: chromatography, mass spectrometry, and fluorescence analysis (also called 3D fluorescence). In a particularly preferred manner, the measurement method can be high-performance liquid chromatography (HPLC), in particular high-resolution liquid chromatography, abbreviated as HPLC-HR, and more particularly combined with mass spectrometry (MS). Figures 3, 4, and 5 correspond to graphical representations of the measurement results obtained with the latter measurement methods, HPLC-HR and MS. Each point in the graph corresponds to a molecule or a type of molecule reacted in the chromatography. Thus, each point corresponds to an imprint of a contaminant or type of contaminant. In other words, the graphs are also called imprints. For each of these points in the graphs, the X-axis of these figures represents the retention time of the molecule corresponding to the point, measured in minutes. The Y-axis of these figures indicates the mass-to-charge ratio of each molecule corresponding to the point. This value is measured in m / z, i.e., mass per ionic charge. Finally, each point on these figures represents the intensity of a chromatographic response - a black dot for high intensity, a gray dot for medium intensity, and a white dot surrounded by a black circle for low intensity - which is related to the concentration of the impurity or contaminant represented by that dot.

[0055] FIG. 3 shows the results of a measurement performed by fluid adsorption on a fluid downstream of a treatment process, such as point 33 shown in FIG. 1. This graph allows the presence of impurities in the fluid being treated, particularly contaminants in water. This evidence is not targeted, i.e., if a point appears in the graph, it corresponds to the presence of an impurity or contaminant, without necessarily identifying the exact type of contaminant or attempting to specifically detect this contaminant. Thus, if FIG. 3 shows a point at time t that was not present in the preceding measurement, breakthrough of the sorbent bed by the contaminant corresponding to this point, and therefore in particular a non-target contaminant, can be suspected. However, breakthrough by emerging contaminants, such as pesticide residues, can be detected in particular with the non-targeted measurement methods described above, while targeted measurement methods generally target more traditional contaminants.

[0056] In a preferred embodiment, the detection of quality defects is carried out by comparing the quality of the fluid being treated between upstream and downstream of the adsorption step, as shown in particular in Figures 3, 4 and 5. The measurements in Figures 3, 4 and 5 were carried out by treating water samples from the Seine River in one and the same treatment line. FIG. 4 shows the results of a measurement taken after a first fluid treatment step, such as point 34 shown in FIG. 1, in this example after a decanting step of the fluid to be treated. FIG. 5 shows the results of measurements taken at sampling points of the fluid being treated, such as points 35 shown in FIG.

[0057] In particular, the methods associated with these figures allow for the comparison of the number of contaminants upstream and downstream of the adsorption process. If a point appears downstream of the adsorption process but is abnormally maintained upstream, the adsorbent no longer retains this contaminant. Therefore, it may be advantageous to trigger an adsorbent bed update. Similarly, if a point appears downstream of the adsorption process but is not present upstream, the adsorbent bed is releasing a previously adsorbed contaminant. In this case, it may also be advantageous to trigger an adsorbent bed update. Alternatively or additionally, it may be possible to compare the contaminant levels upstream and downstream of the adsorption process, particularly by mass spectrometry or other measurement methods. Figures 3, 4, and 5 particularly demonstrate medium or high concentrations of impurities according to the intensity of the displayed points. Based on this information about the contaminant concentration or level, it may be advantageous to trigger an adsorbent bed update, especially if a point changes from low or medium intensity to high intensity. In this preferred embodiment, which compares the quality of the fluid being treated upstream and downstream of the adsorption step, the measurement method can be carried out alternately upstream and downstream using a single analytical sensor shared by measurements at these two points.

[0058] Preferably, the detection of quality defects of the fluid to be treated, whether upstream or downstream of the adsorption, can be carried out inline during the treatment method. In this case, the monitoring may be carried out by an inline analyzer, in particular based on a fluorescence or chromatographic method, whether targeted or non-targeted (imprint) as described above. This inline monitoring allows the detection of quality defects during the treatment of the fluid to be treated, optimizing the volume (quantity) of the fluid to be treated. This inline detection step may be carried out continuously or quasi-continuously, for example with an inter-detection period shorter than the transit time of a bed volume through the adsorbent bed. Furthermore, this detection step may be carried out upon request from an operator. In any case, it is not necessary to interrupt the production of the fluid to be treated in order to determine whether a quality defect has been revealed and whether a bed renewal step must be triggered.

[0059] Alternatively or additionally, the renewal process may be triggered periodically at a preset renewal frequency. The renewal period can be defined to limit the risk of sorbent bed breakthrough and rejuvenate the sorbent according to the proposed method. The reduction target preset according to the proposed method can be defined as a nominal contaminant reduction, at which the reduction of residual contaminants after the renewal period is determined to be sufficient to reach the quality standard required by the operator.

[0060] Preferably, the predetermined reduction target is defined as a minimum reduction value for the concentration of each contaminant in the group of contaminants being tested. This reduction of a contaminant is usually expressed as the ratio of the amount of retained contaminant to the amount of intruding contaminant. It is therefore a relative value. Nevertheless, this relative value allows the post-treatment contaminant concentration to meet regulatory thresholds for predictable intruding contaminant concentrations. For each contaminant, this minimum reduction value can be 50% to 90%, preferably 60% to 80%, or even 70% to 80%. The minimum reduction value may be defined differently depending on the contaminant, or it may be the same for all contaminants being tested. In particular, in this embodiment, and more generally in the proposed method, the actual reduction of a contaminant can be determined separately for each contaminant in the group of contaminants being tested.

[0061] All contaminants for which the actual reduction of the contaminants is determined may be selected from the group formed by desethylhydroxyatrazine, metaldehyde, aminotriazole, metazachlorin, metallochlorin, desethylatrazine, chlortoluron, atrazine, terbuthylazine, in particular. It is particularly advantageous to include contaminants identified as causing quality defects in the group of contaminants for which the actual reduction of the contaminants is determined, especially if a detection step for the target contaminants is carried out downstream of the sorbent bed.

[0062] As mentioned above, to determine the actual reduction achieved by the extracted rejuvenated adsorbent sample, a sample of the fluid being treated is preferably taken, particularly downstream of the adsorption step. More preferably, the sampled fluid is doped with the contaminant before determining the actual reduction of the contaminant. This doping allows for a shorter time required to determine the actual reduction. For example, doping can be achieved by increasing the concentration of the contaminant under consideration in the sampled fluid. Typically, deshydroxyatrazine, metaldehyde, aminotriazole, metazachlorin, or metallochlorin can be doped by increasing their concentrations by 1 to 10 μg / L, preferably 1 to 2 μg / L, and more preferably 1 to 2 μg / L. Similarly, desethylatrazine, chlortoluron, atrazine, and terbuthylazine can be doped by increasing their concentrations by 1 to 10 μg / L, preferably 9 to 10 μg / L.

[0063] As mentioned above, with or without the use of a sample of the fluid to be treated and with or without doping of this sampled fluid, the determination of the target average age of the sorbent sample is preferably obtained by measuring the actual reduction of contaminants in a sorbent mixture, which corresponds in particular to a mixture of a young sorbent, such as a fresh or regenerated sorbent, and at least a portion of the sorbent sample extracted from the sorbent bed.

[0064] In a further preferred embodiment, the actual reduction is measured in several subsamples obtained from the extracted sorbent sample. These subsamples are obtained by mixing the extracted sorbent with a younger sorbent. The mixing to obtain the subsamples is performed in a variable ratio to obtain a set of subsamples with a range of average ages. This range of average ages allows for an intermediate age distribution between the age of the extracted sorbent and the age of the younger sorbent. The intermediate average age range allows for as many actual reduction measurements as possible and ensures that the extracted sorbent is closest to the optimal age at which it should be rejuvenated, so that it exhibits an actual reduction that best corresponds to the preset reduction target. Among these subsamples, the sample with the actual contaminant reduction that best meets the preset reduction target is the sample used to determine the target average age of the proposed method. In particular, the target average age can be interpreted as the age of the mixture forming this subsample. Alternatively, the target average age can be interpreted as the average or extrapolation of the age of the mixture of this subsample and the age of the subsample immediately above it within the intermediate age range. In addition to determining the target average age, rejuvenation of the adsorbent bed is performed until the target average age is achieved.

[0065] The number of intermediate ages in the set of subsamples may be 2 to 10, or preferably 3 to 5. A reasonable number of subsamples allows for the estimation of a target average age that is sufficiently close to the optimal age. By considering the number of subsamples, it is preferable that the intermediate ages are evenly distributed between the age of the extracted sorbent and the age of the young sorbent. This even distribution increases the probability of reaching a target average age that is close to the optimal age for rejuvenating the extracted sorbent.

[0066] Figures 6 and 7 show the actual pollutant reduction measurements for subsamples of the mixture with three and five intermediate average ages, respectively. Figure 6 was obtained by measuring the actual reductions of Seine water sampled near Mont-Valérien after treatment with a 10 μg / L pollutant doping based on an adsorbent sample, in this case a granular activated carbon sample with an age of 45,000 treated bed volumes (BVT). Figure 7 was obtained by measuring the actual reductions of Seine water sampled near Morsant-sur-Seine after treatment with a 10 μg / L pollutant doping based on an adsorbent sample, in this case a granular activated carbon sample with an age of 140,000 treated bed volumes (BVT). All reduction measurements were performed on water samples taken downstream of the adsorption process. The x-axis of these figures shows the reconstructed average age of the subsamples, expressed in BVT. The first y-axis of these figures shows the percentage of adsorbent from the extracted sample and the percentage of fresh adsorbent, respectively. The second Y-axis of these figures shows the actual reduction measured for the subsamples as a function of the contaminant being tested.

[0067] In particular, the last column of these figures shows the actual reduction of each contaminant in the sorbent bed represented by the extracted sample, before any rejuvenation. In this example, in Figure 6 for the sample extracted at 45,000 BVT, the contaminant reductions are seen in this order: metaldehyde (dotted line), metazachlorine (mixed dashed line), metallochlorine at the same point as desethylhydroxyatrazine (both thin solid lines), although desethylhydroxyatrazine had a higher reduction level than metallochlorine before this common point, and finally aminotriazole (thick solid line). In Figure 7 for the sample extracted at 140,000 BVT, the contaminant reductions are seen in this order: metaldehyde (dotted line), aminotriazole (thin solid line), metazachlorine (mixed dashed line), and finally desethylhydroxyatrazine (thick solid line), which shares the same curve as metallochlorine.

[0068] First, these two figures clearly show that breakthrough does not always occur simultaneously and in the same order for a given group of contaminants, depending on the actual history of the adsorbent bed (in this example, granular activated carbon). In particular, in Figure 6, aminotriazole did not cause bed breakthrough, whereas in Figure 7, bed breakthrough by aminotriazole exceeds that of metazaclorin. This observation confirms the advantage of performing actual reduction measurements on samples of the fluid to be treated taken downstream of the adsorption step, as previously mentioned.

[0069] These figures further demonstrate the benefit of reconstituting (or simulating) the rejuvenated mean age of extracted sorbents. In these figures, the sorbents are rejuvenated with mixtures having various percentages of younger sorbents. The range of reconstituted median ages allows for the best determination of the reconstituted ages at which reduction of all tested contaminants can be considered to meet the pre-established reduction targets. In this example, if a 90% reduction of all these contaminants is required, a 45,000 BVT sorbent bed can be rejuvenated to 15,000 BVT by replacing nearly 70% of the sorbent bed with fresh sorbent (Figure 6). For this same reduction target, a 140,000 BVT sorbent bed can be rejuvenated to 100,000 BVT by replacing nearly 30% of the bed sorbent with fresh sorbent (Figure 7).

[0070] In one particularly advantageous embodiment, the measurement of the contaminant reduction of the adsorbent mixture is performed by a Short Bed Absorber test (SBA).

[0071] FIG. 8 shows an SBA tester 60 equipped with eight filtration cartridges 62 filled with the adsorbent of the extracted sample, mixed in various proportions with young adsorbent.

[0072] The proposed method has the advantage that these SBA tests use very little extracted sorbent, allowing tests to provide results in relatively short contact times, such as about 6 minutes, which corresponds to the contact time of the sorbent bed from which the sorbent sample is extracted. The contact time of the SBA test or the contact time of the sorbent bed from which the sample is extracted can vary from 4 to 10 minutes, or from 5 to 20 minutes, to 30 minutes, depending on the operating conditions. These contact times are compatible with SBA beds having volumes of 100 mL or less, 50 mL or less, or even 20 mL or less. Regardless of whether or not an SBA test is performed, the amount of sorbent in the extracted sample can be less than 2 L, preferably less than 1 L, or even less than 200 mL or 100 mL.

[0073] In one embodiment of the proposed method, with or without the use of subsamples, particularly SBA subsamples, the extraction of the sorbent is carried out during an interruption phase of the treatment of the fluid to be treated. If such an extraction is carried out, the interruption phase preferably corresponds to an interruption phase that serves a purpose other than simply extracting a sample of the sorbent, such as an sorbent cleaning phase, particularly an sorbent cleaning phase by injecting air. Alternatively, the sorbent extraction is carried out during the treatment of the fluid to be treated. In this case, sampling is carried out at one or more points in the sorbent bed in order to obtain a sample of the sorbent that is representative of the wear of the entire sorbent bed.

[0074] In a preferred embodiment, the proposed method is carried out as part of an upflow treatment process using activated carbon (hereinafter Carbazur UP), as disclosed in the above-cited and referenced patent application FR 3003477. In particular, the proposed treatment process may be a treatment process involving the upflow of a fluid through a bed of granular activated carbon, which process comprises: at least one filtration / adsorption stage, in which the velocity of the fluid is sufficiently low so as not to cause substantial expansion of a bed of activated carbon that allows both filtration and adsorption of substances contained in the fluid; at least one expansion stage in which a fluid is circulated through the activated carbon bed at a rate sufficient to substantially expand the activated carbon bed, thereby subjecting the activated carbon bed to washing with the fluid; Includes:

[0075] This embodiment is particularly illustrated in FIG. 9, which shows a bed of granular activated carbon operating in an upflow mode. FIG. 9 particularly shows a reactor 20 comprising a bed of granular activated carbon 22. The bed 22 operates with the rising water 28 to be treated. The reactor 20 may comprise an activated carbon inlet 24 and an outlet 26 for renewing the carbon bed. The reactor 20 may comprise an air inlet 34 to allow washing of the bed with air rather than just fluid. The reactor 20 may be in series with a filter 70, which may consist of a downflow bed 72 of activated carbon as shown. After being filtered by the filter 70, the water is discharged via an outlet 33.

[0076] The representativeness of the actual contaminant reduction determined by the SBA test was assessed by comparing it with that of the Carbazur UP method described above. Figures 10 and 11 show comparative tests of the adsorbent-mediated contaminant reduction measured with the SBA and the contaminant reduction measured with a pilot-scale Carbazur UP unit. These pilot-scale Carbazur UP units correspond to laboratory models designed to represent the industrial-scale operation of one and the same line. The x-axis of Figures 10 and 11 shows the different contaminants tested, and the y-axis shows the percent reduction of these contaminants. For each contaminant tested, the left bar represents the reduction measured in the pilot test, and the right bar represents the reduction measured by the SBA test.

[0077] Figure 10 compares the SBA and Carbazur UP tests on granular activated carbon aged 50,000 BVT, using seine water and a contact time of 6 minutes (contact time is the Empty Bed Contact Time - EBCT). From left to right, the contaminants tested are desethylhydroxyatrazine, aminotriazole, metazachlorin, and metallochlorin.

[0078] Figure 11 compares the SBA test with Carbazur UP for a granular activated carbon aged 80,000 BVT, using seine water and an EBCT contact time of 10 minutes. The contaminants tested, from left to right, are desethylhydroxyatrazine, aminotriazole, metazachlorin, metallochlorin, metazachlorin ESA, and metallochlorin ESA (ESA stands for Ethan Sulfonic Acid).

[0079] As shown in Figures 10 and 11, the reduction levels measured according to the two methods are very similar. The SBA test performed in this invention is a method for characterizing the saturation level of a particular compound and represents a pilot. The pilot is designed and sized taking into account relevant factors of the industrial unit (bed height, contact time, etc.). Therefore, the SBA test representing the pilot also represents the industrial unit.

[0080] Also proposed is a fluid treatment installation for carrying out the proposed method described above in all its variants, including measuring the actual reduction of contaminants in a mixture of sorbent extracted from the sorbent bed and young sorbent. The proposed installation therefore comprises a measurement unit. The proposed installation further comprises an adsorption reactor, which contains an sorbent bed held therein, and which allows for the adsorption of contaminants contained in the fluid to be treated.

[0081] This reactor is particularly shown in FIG. 9 above or in FIG. 12 detailed below.

[0082] As shown in FIG. 12 , the proposed equipment 90 includes a reactor 20 with an orifice 26 for at least partially removing used sorbent from the sorbent bed 22. As shown, the sorbent bed 22 can be operated with an upflow of the fluid 28 to be treated. In the proposed equipment, a sample of the sorbent is directed to a mixer 92. Accordingly, the proposed equipment may include this mixer 92. This mixer allows for mixing of the sample of used sorbent with young sorbent, for example, obtained from a young sorbent reservoir 94. The proposed equipment may include this reservoir 94. The mixer 92 can allow for obtaining a single sample of intermediate age or a collection of subsamples having an intermediate average age range between the age of the extracted sorbent and the age of the young sorbent. In the proposed equipment, the obtained mixture, whether forming a single sample or a collection of subsamples, is directed into a measurement unit 96. This measurement unit 96 measures the actual reduction of contaminants in the mixture of sorbent extracted from the sorbent bed and young sorbent.

[0083] As previously mentioned, the measurement unit 96 is preferably a short bed adsorber (SBA), and the test is preferably carried out on a bed having a volume of 100 mL or less, more preferably 50 mL or less, and even more preferably 20 mL or less.

[0084] As mentioned above, the measuring unit 96 preferably allows the determination of a target average age of the extracted sorbent sample for a group of contaminants selected from the group formed by desethylhydroxyatrazine, metaldehyde, aminotriazole, metazachlorin, metazachlorin ESA, metallochlorin, metallochlorin ESA, desethylatrazine, chlortoluron, atrazine, terbuthylazine. Optionally, the group of contaminants also includes contaminants identified as contributing to quality defects in the fluid to be treated and triggering a renewal process.

[0085] Finally, in one particularly preferred embodiment, the measurement unit 96 also receives the fluid processed by the reactor 20, so that measurements are made under conditions as close as possible to the future conditions of the adsorbent bed.

[0086] It is clear that the invention is not limited to the examples and embodiments described and shown, but that many variations are possible.

[0087] In particular, in one variant, the measurement method for detecting water quality defects is a 3D fluorescence analysis, which allows in-line monitoring of the water quality from its entry into the adsorbent to its exit. This method is particularly suitable for pollutant-laden wastewater or sources represented by algae. This method is therefore advantageous for industrial wastewater or leachates formed by liquid effluents from waste storage facilities.

[0088] In another variation, the adsorbent is not limited to granular activated carbon, but may also be resins, clays, zeolites for capturing metals or other specific trace contaminants. These other adsorbents are preferably millimeter sized.

[0089] The adsorbent may also be microgranular activated carbon, i.e., having a finer particle size than granular carbon. For example, the adsorbent may be microgranular activated carbon with at least 85-90% of the granules having a particle size of 300 μm to 800 μm, or preferably at least the majority (50%) of the granules having a particle size of 400-600 μm. The sizes mentioned above are equivalent diameter sizes of the particles when dry-screened or wet-screened.

[0090] In a further variant, the proposed method is carried out using an adsorption step directly on the raw water, and thus in particular without prior decantation or flotation.

[0091] In addition, the proposed method is adaptable to a wide range of fluid velocities, for example, selected from 2 m / h to 20 m / h, preferably 5 m / h to 20 m / h, more preferably 10 m / h to 20 m / h or 5 m / h to 15 m / h. The fluid velocity can be selected in particular as a function of the water temperature.

Claims

1. A method for treating a fluid, in particular water such as surface water or groundwater, or even wastewater, comprising a step (18) of adsorption of contaminants contained in the fluid to be treated by an adsorbent bed (22), the method further comprising a step (40) of restoring the adsorption capacity of the adsorbent bed (22), the step (40) comprising: extracting (42) a sample of adsorbent from the adsorbent bed (22); determining a target average aging level for the extracted sorbent sample by restoring the adsorption capacity of at least a portion of the extracted sorbent sample (44), wherein at the target average aging level, the extracted sorbent exhibits an actual reduction in contaminants corresponding to a pre-established reduction target; Aging (46) the adsorbent bed (22) to the determined target average aging of the adsorbent bed (22). Including, determining (44) a target average aging of the extracted sorbent sample by measuring the actual reduction in contaminants in a mixture of less aged sorbent and at least a portion of the sorbent sample extracted from the sorbent bed; and restoring (46) the adsorbent bed (22) to the determined target average age is achieved by at least partially replacing the adsorbent bed with an adsorbent having a lower age until the determined target average age of the adsorbent bed is achieved. method.

2. 2. The method of claim 1, wherein step (40) can be triggered periodically at a predetermined frequency.

3. 3. The method according to claim 1 or 2, wherein said step (40) can be triggered after a step of detecting a quality defect in the fluid to be treated downstream of said adsorption step (18).

4. The method of claim 3 , wherein the detection of quality defects occurs during the processing of the fluid to be processed.

5. 5. The method according to claim 1, wherein the extraction of the sorbent is carried out at an interruption stage of the treatment of the fluid to be treated.

6. 5. The method of claim 1, wherein the extraction of the adsorbent is performed by sampling at one or more points in the adsorbent bed as the treatment of the fluid to be treated is carried out.

7. 7. The method of claim 1, wherein in determining the target average aging of the extracted adsorbent sample, the actual reduction of contaminants is determined by sampling the fluid being treated downstream of the adsorption step.

8. 8. The method of claim 1, wherein determining (44) a target average aging degree of the extracted sorbent sample is obtained by measuring the contaminant reduction in subsamples of the sorbent obtained by mixing the extracted sorbent with lesser-aged sorbents in various proportions, the set of subsamples having a range of intermediate average aging degrees between the aging degree of the extracted sorbent and the aging degree of the lesser-aged sorbent, and the intermediate average aging degree of the subsample having the actual contaminant reduction that best corresponds to the predetermined reduction goal determines the target average aging degree.

9. 9. The method of claim 8, wherein the number of the intermediate average ages of the sub-samples of the adsorbent is between 2 and 10.

10. 10. The method of claim 1, wherein the one or more measurements of contaminant reduction of the one or more mixtures of adsorbents are performed by short bed adsorber measurements.

11. 11. The method of any one of claims 1 to 10, wherein when determining the target average aging degree of the extracted sorbent sample, the actual reduction of contaminants is determined for a group of contaminants selected from the group formed by desethylhydroxyatrazine, metaldehyde, aminotriazole, metazachlorin, metazachlorin ESA, metallochlorin, metallochlorin ESA, desethylatrazine, chlortoluron, atrazine, terbuthylazine.

12. 12. The method of claim 1, wherein the predetermined reduction target is defined as a minimum reduction in concentration of each contaminant in a group of contaminants being tested.

13. 13. The method according to any one of claims 1 to 12, wherein the adsorbent bed (22) comprises granular activated carbon and the adsorption step (18) is carried out by passing the fluid to be treated, in particular water, through the adsorbent bed in an upflow manner.

14. 14. An installation (90) for treating a fluid according to a method according to any one of claims 1 to 9 or according to any one of claims 10 to 13 further dependent on any one of claims 7 to 9, comprising: a reactor (20) for adsorbing contaminants contained in the fluid to be treated, the reactor (20) having an adsorbent bed (22) held therein and an orifice (26) for at least partially removing used adsorbent from the adsorbent bed (22); a mixer (92) for mixing the sorbent sample extracted from the sorbent bed with less aged sorbent; a measurement unit (96) for measuring the actual reduction of contaminants in the mixture of the less aged adsorbent and the adsorbent extracted from the adsorbent bed; Equipped with, facilities.

Images